The Oxford Handbook of Papyrology

The Oxford Handbook of Papyrology

The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011 Subject: Classical Studies

Online Publication Date: Sep 2012

(p. iv)

Oxford University Press, Inc., publishes works that further

Oxford University's objective of excellence
in research, scholarship, and education.

Oxford New York

Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto

With offices in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam

Copyright © 2009 by Oxford University Press, Inc.

Published by Oxford University Press, Inc.

198 Madison Avenue, New York, New York 10016
www.oup.com

First issued as an Oxford University Press paperback, 2011

Oxford is a registered trademark of Oxford University Press

All rights reserved. No part of this publication may be reproduced,

stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise,
without the prior permission of Oxford University Press.

Library of Congress Cataloging-in-Publication Data

The Oxford handbook of papyrology / edited by Roger S. Bagnall.
p. cm.
Includes bibliographical references and index.
ISBN 978–0–19–517838–8 (hardcover); 978–0–19–984369–5 (paperback)
1. Manuscripts (Papyri). 2. Paleography. 3. History, Ancient—Historiography.
4. Middle East—Historiography. I. Bagnall, Roger S. II. Title: Handbook of papyrology.
Z110.P36O98 2009

Page 1 of 2
The Oxford Handbook of Papyrology

091—dc22 2008031679

135798642

Printed in the United States of America

on acid-free paper

Page 2 of 2
Contributors

Contributors
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011 Subject: Classical Studies

Online Publication Date: Sep 2012

(p. vii) Contributors

Roger S. Bagnall Professor of Ancient History and Director of the Institute for the
Study of the Ancient World, New York University

Adam Bülow-Jacobsen Former Carlsberg Research Professor, University of Copen­

hagen

Guglielmo Cavallo Professor of Palaeography, University of Rome—La Sapienza

Willy Clarysse Professor of Ancient History, Katholieke Universiteit Leuven

Raffaella Cribiore Professor of Classics, New York University

Hélène Cuvigny Directeur de recherches, Centre National de la Recherche Scien­

tifique (Institut de Recherche et dʼHistoire des Textes)

Eleanor Dickey Associate Professor of Classics, University of Exeter

Page 1 of 4
Contributors

Maria Rosaria Falivene Associate Professor of Papyrology, Faculty of Letters and

Philosophy, University of Rome Tor Vergata

Jean-Luc Fournet Directeur dʼÉtudes en papyrologie grecque, École Pratique des

Hautes Études, Section des sciences historiques et philologiques, Paris

Jaakko Frösén Professor of Greek Philology, University of Helsinki

Jean Gascou Professor of Papyrology, University of Paris-Sorbonne

Nikolaos Gonis Reader in Papyrology, University College London

Todd M. Hickey Assistant Professor of Classics, University of California, Berkeley

William A. Johnson Associate Professor of Classics and Head of the Department of

Classics, University of Cincinnati

Alexander Jones Professor of the History of the Exact Sciences in Antiquity, Institute
for the Study of the Ancient World, New York University

James G. Keenan Professor of Classical Studies, Loyola University Chicago

Page 2 of 4
Contributors

David G. Martinez Associate Professor, Department of Classics and the Divinity

School, University of Chicago

(p. viii) Bernhard Palme Professor of Ancient History and Papyrology, University of
Vienna

Timothy Renner Professor of Classics and General Humanities, Montclair State Uni­
versity

Cornelia Römer Director of the Papyrus Collection and Papyrus Museum, Austrian
National Library, Vienna

Paul Schubert Professor of Greek Language and Literature, University of Geneva

David Sider Professor of Classics, New York University

Petra M. Sijpesteijn Professor of Arabic Language and Culture, Leiden University

Dorothy J. Thompson Former Newton Trust Lecturer and affiliated lecturer in an­
cient history, and Life Fellow of Girton College, University of Cambridge

Katelijn Vandorpe Professor of Ancient History, Katholieke Universiteit Leuven

Peter van Minnen Associate Professor of Classics and Ancient History, University of
Cincinnati

Page 3 of 4
Contributors

Uri Yiftach-Firanko Assistant Professor of Classics, Hebrew University of Jerusalem

Page 4 of 4
Figures

Figures
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011 Subject: Classical Studies

Online Publication Date: Sep 2012

(p. ix) Figures

1.1. Ancient papyrus on the lightbox 9
1.2. Diagrams from Hendriks's original article 9
1.3. Papyrus made by Hendriks's peeling method 10
1.4. Papyrus made by H. Ragab 10
1.5. ʼAμμώνις ʼAμμωνίου (wooden label with its string intact) 12
1.6a–b. Letter written on a folded leaf tablet and a concertina tablet 13
1.7. Athenian ostracon 15
1.8. Ostracon in preparation from Mons Claudianus 16
1.9. Example of a shaped ostracon 16
1.10. Giant ostracon 17
1.11. Bookroll with calamus and inkwell and polyptych with stylus 19
1.12. Muse reading from a bookroll 20
1.13a–b. Drawings of bookroll and “rotulus” written “transversa charta” 22
1.14. Reconstruction of a polyptych from “Villa dei papyri” in Herculaneum 23
1.15. Reconstruction of roll from which the first four sheets for a codex were cut 25
2.1. Karanis 35
2.2. Papyrus fragments found at Oxyrhynchus 37
2.3. Rolls of papyrus found in doorway between rooms D and E of 5026 at Karanis 39
2.4. Cartonnage elements at Tebtynis 45
2.5. Crocodile mummies at Tebtynis 46
2.6. Babatha's archive before opening 51
2.7. Babatha's pouch, opened 52
2.8. A layer with ostraca in the garbage dump at Krokodilô 53
2.9. Oxyrhynchus 1902–1903 54
2.10. Didymoi 1999 54
4.1. Unrolling a pyrolized papyrus scroll without dampening 84
4.2. Conserving a mummy cartonnage, steps 1 and 2 89
(p. x) 4.3. Conserving a mummy cartonnage, step 8 90

4.4. The conserved mummy cartonnage 91

Page 1 of 4
Figures

4.5. Three carbonized rolls before conservation 93

4.6. Section of a loosely rolled Petra papyrus 94
4.7. An assembled plate of Petra papyri 95
5.1. PSI IV 325 104
5.2. PSI IV 383 106
5.3. P.Mil. Vogl. VII 307 107
5.4. P.Bad. II 3 108
5.5. BGU III 998 109
5.6. P.Oxy. VII 1061 110
5.7. P.Ryl. II 73 111
5.8. BGU I 197 112
5.9. P.Mur. 108 113
5.10. BKT V.2, 131–139 115
5.11. PSI XI 1214 116
5.12. PSI IX 1091 117
5.13. P.Lond.Lit. 134 118
5.14. P.Oxy. XXXI 2545 119
5.15. P.Flor. I 47 121
5.16. CPR I 32 122
5.17. P.Mur. 114 123
5.18. P.Yadin 18 124
5.19. P.Dura 23 125
5.20. P.Berol. inv. 11532 125
5.21. P.Oxy. VIII 1100 126
5.22. PSI V 446 127
5.23. The Hawara Homer 128
5.24. PSI IX 1086 130
5.25. PSI X 1164 131
5.26. PSI I 1 132
5.27. P.Oxy. I 26 133
5.28. PSI X 1169 134
5.29. P.Flor. III 389 134
5.30. PSI VI 729 137
5.31. P.Herc. 817 138
5.32. P.Vindob. inv. L 112 139
(p. xi)

5.33. P.Grenf. II 110 140

5.34. P.Vindob. inv. L 31 143
5.35. P.Vindob. inv. L 90 144
7.1. List of symbols 176
9.1. P.Gen. inv. 500 205
10.1. Petaus practicing the phrase “I, Petaus, village scribe, have handed
in” (Πεταῦς κωμογραμματεὺς ἐπιδέδωκα) 217
10.2a–c. Demotic archive of the funerary priest Teos and his wife, Thabis 220–222
10.3. Bilingual archive of Totoes, priest of Hathor 223

Page 2 of 4
Figures

10.4. One book, two archives 224

10.5. Four fragments of a single petition by Dryton 225
10.6. “Cantina dei papiri” 228
10.7. Returned loan; the creditor canceled it by crosshatching the text 239
10.8. Language use in archives until 600 CE 242
10.9. Endorsement of estate manager Zenon 243
10.10. Roman tomos synkollêsimos (“pasted roll”) 244
11.1. MP 1233 (P.Lond.Lit. 132) 258
11.2. Anatomy of the bookroll 260
17.1. Memphite Serapeum papyrus (now in the Louvre) with Greek and Macedonian
months written in Apollonios' hand, plus a demotic contract 413
18.1. Example of complete Pehlevi document 420
18.2. Literary papyri in Latin 422
18.3. Letter of recommendation from the rationalis Vitalis to the governor of Phoeni­
cia 425
18.4. Report of proceedings before the comes Aegypti (P.Oxy. LXIII 4381, 375 CE)
426
18.5. Completion of the lawyer Justus in Greek and Latin letters 427
18.6. School exercise in hieratic writing with translation in Old Coptic 432
18.7. Quantitative comparison between Greek and Coptic documents 435
18.8. Coptic documents from the beginning until the Arab conquest 436
18.9. Greek letter by the anchorite Apa Johannes 438
18.10. First dated Coptic text 440
18.11. Beginning of a Coptic legal text (a release) with a Graeco-Arabic protocol 442
18.12. Example of Coptic and Greek graphic styles 443
(p. xii) 21.1. Principal relatives of Kronion and Isidora 505

24.1. Pylon of the Edfu temple 562

24.2a–b. Two oracle questions, one positive and one negative 563
24.3. Bust of Sarapis 564
24.4. Theagoi carrying the mummified crocodile 566
24.5. Semtheus identified with Greek Herakles 567
24.6. Ground plan of the Edfu temple 569
24.7. Votive inscription found in Pisa, Italy 570
24.8. Ptolemy IV making an offering to the gods of Edfu, followed by Ptolemy III, II,
and I with their queens, as synnaoi theoi 576
24.9. Egyptian Neith as Greek Athena, with her son, the sphinx god Tutu 578
24.10. Soknebtynis-Kronos and ithyphallic Min in Tebtynis 580
24.11. Horos cippus 585
25.1a–b. Bodl. MS. Gr. bibl. g. 5 (P), recto and verso 594
25.2. Chester Beatty Biblical Papyrus II (P46; third century) 597
25.3a. P.Amh. I 3 (Pierpont Morgan Library Pap. G. 3), recto 603
25.3b. P.Amh. I 3 (Pierpont Morgan Library Pap. G. 3), verso 604
26.1. Mani, drawing based on a wall fresco found at Khotscho, near Turfan 625
26.2. The Nag Hammadi codices in their original covers 627

Page 3 of 4
Figures

26.3. Nag Hammadi Codex II, 28–29, Apocryphon of John 629

26.4. Cologne Mani codex (enlarged) 634
26.5. House 3 at Kellis 640
26.6. P.Kell.Copt. 25 641

Page 4 of 4
Abbreviations

Abbreviations
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011 Subject: Classical Studies

Online Publication Date: Sep 2012

(p. xiii) Abbreviations

AJA American Journal of Archaeology

AncSoc Ancient Society

ANRW Aufstieg und Niedergang der römischen Welt

APF Archiv für Papyrusforschung und verwandte Gebiete

ARAM ARAM periodical

BASP Bulletin of the American Sociey of Papyrologists

BCH Bulletin de correspondance hellénique

BIFAO Bulletin de lʼInstitut français dʼarchéologie orientale

BSAC Bulletin de la Société dʼArchéologie Copte

BSOAS Bulletin of the School of Oriental and African Studies

CdʼÉ Chronique dʼÉgypte

CErc Cronache ercolanesi

CRAI Comptes rendus de lʼAcadémie des Inscriptions et Belles-Lettres

Page 1 of 3
Abbreviations

CRIPEL Cahiers de recherches de lʼInstitut de Papyrologie et dʼÉgyptologie de

Lille III

CSCA California Studies in Classical Antiquity

EVO Egitto e vicino oriente

GRBS Greek, Roman and Byzantine Studies

JEA Journal of Egyptian Archaeology

JHS Journal of Hellenic Studies

JJP Journal of Juristic Papyrology

JRA Journal of Roman Archaeology

JRAS Journal of the Royal Asiatic Society

JRS Journal of Roman Studies

MBAH Müonstersche Beiträge zur antiken Handelsgeschichte

MDAIK Mitteilungen des Deutschen Archäologischen Instituts (Abt. Kairo)

OLA Orientalia Lovaniensia analecta

QDAP Quarterly of the Department of Antiquities of Palestine

(p. xiv) Quaderni di storia

QS

RAssyr Revue assyriologique

RBi Revue biblique

REB Revue des études byzantines

RHR Revue de lʼhistoire des religions

RIDA Revue internationale des droits dʼantiquité

Page 2 of 3
Abbreviations

SCI Scripta classica Israelica

SDHI Studia et documenta historiae iuris

TR Tijdschrift voor Rechtsgeschiedenis

ZPE Zeitschrift für Papyrologie und Epigraphik

ZSav Zeitschrift der Savigny-Stiftung für Rechtsgeschichte: Romanistische

Abteilung

Page 3 of 3
Internet Resources

Internet Resources
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011 Subject: Classical Studies

Online Publication Date: Sep 2012

(p. xv) Internet Resources

The following list gives only a selection of tools mentioned in this volume. For lists of ad­
ditional resources, with links, see the following two Web sites:

http://www.columbia.edu/cu/lweb/digital/pn/resources.html

http://www.ulb.ac.be/assoc/aip/liens.htm

Advanced Papyrological Information System (APIS). A union catalogue of metadata and

images from many papyrus collections, including a set of links to other online catalogues
of papyrus collections, http://www.columbia.edu/cu/lweb/projects/digital/apis/index.html;
also through http://www.papyri.info

Catalogue of Paraliterary Papyri (CPP). Katholieke Universiteit Leuven, Departement

Klassieke Studies. http://cpp.arts.kuleuven.ac.be/

Checklist of Arabic Documents. http://www.ori.uzh.ch/isap/isapchecklist.html

Checklist of Editions of Greek, Latin, Demotic, and Coptic Papyri, Ostraca, and Tablets.
http://scriptorium.lib.duke.edu/papyrus/texts/clist.html

Cologne Mani-Codex. Images, http://www.uni-koeln.de/phil-fak/ifa/NRWakademie/papy­

rologie/Manikodex/mani.html

Demotistische Literaturübersicht, Katholieke Universiteit Leuven. http://

www.trismegistos.org/dl/index.html

Duke Data Bank of Documentary Papyri. A full-text database only for Greek and Latin
documents. http://www.papyri.info

Fayyum Villages Project. Now included in Trismegistos: places. http://

www.trismegistos.org/geo/index.php

Page 1 of 2
Internet Resources

Heidelberger Gesamtverzeichnis der griechischen Papyrusurkunden aus Ägypten. A data­

base (without texts) of Greek and Latin documentary papyri and ostraca. http://
www.rzuser.uni-heidelberg.de/~gvo/gvz.html; also through http://www.papyri.info

Homer and the Papyri, Harvard Center for Hellenic Studies. http://chs.harvard.edu/chs/
homer___the_papyri

Leuven Database of Ancient Books (LDAB). A guide to Greek, Latin, and Coptic lit­
(p. xvi)

erary texts, including school exercises. http://www.trismegistos.org/ldab/index.php

Mertens-Pack, 3d ed. A digital update of Roger A. Pack, Index of Greek and Latin Literary
Texts from Greco-Roman Egypt, 2d ed. Ann Arbor: University of Michigan Press, 1965. A
listing of literary papyri with full references. http://www2.ulg.ac.be/facphl/services/ce­
dopal/

Nag Hammadi Library. http://www.gnosis.org/naghamm/nhl.html

Namen in koptischen dokumentarischen Texten (by M. Hasitzka). A repertory of personal

names attested in Coptic documents. http://www.onb.ac.at/sammlungen/papyrus/publ/
kopt_namen.pdf

Oxyrhynchus Papyri. Metadata and images of the more recent volumes. http://
www.papyrology.ox.ac.uk/POxy/

Papyrus Archives in Graeco-Roman Egypt. Database with description of archives and

dossiers. http://www.trismegistos.org/arch/index.php

Thesaurus Linguae Graecae. A comprehensive database of Greek literary texts from

Homer to Byzantine times (license required). http://www.tlg.uci.edu

Vindolanda Tablets. Metadata, translations, and images. http://vindolanda.csad.ox.ac.uk/

Page 2 of 2
Introduction

Introduction
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011 Subject: Classical Studies

Online Publication Date: Sep 2012

(p. xvii) Introduction

In a broad sense, papyrology is a discipline concerned with the recovery and exploitation
of ancient artifacts bearing writing and of the textual material preserved on such arti­
facts. For the most part it focuses on what can be called the spectrum of everyday writing
rather than forms of writing intended for publicity and permanence, most of which were
inscribed on stone or metal and belong to epigraphy, in the scholarly division of labor. The
edges of these domains, however, are fuzzy. Papyrology cannot actually be defined by the
material support: Potsherds can belong to epigraphy or papyrology, depending on their
origin and nature, while the great parchment codices of the fourth and fifth centuries are
not usually thought of as papyrological texts. Technique of writing is not an adequate dis­
criminant, for not all epigraphical texts are incised, and some papyrological texts are. A
public/private dichotomy is undermined by papyri put up as public notices, and many
types of content are found in both epigraphical and papyrological texts—edicts of Roman
governors, to give only one obvious example. Nor does geography divide the fields: Both
papyrological and epigraphical texts can be found from Britain to Afghanistan, although,
for environmental reasons, most papyrological material comes from Egypt. Material that
in Egypt would be considered papyrological finds a home in the Corpus inscriptionum
iranicarum when written in a Persian language. In one sense, none of this is a problem
unless one wants to close oneself into a discipline with clear boundaries. But for the edi­
tor of a handbook it poses certain challenges.

Publishing a handbook for a field such as papyrology presupposes some sense of approxi­
mate boundaries. A generation ago, “papyrology” meant Greek and Latin papyrology, and
the borders were thus clear at least in linguistic terms. Neither Coptic nor Arabic papy­
rology had more than a handful of practitioners, and demotic Egyptian unquestionably be­
longed to the Egyptologists. In the summer seminar in papyrology in 1968, at which I re­
ceived my first training, I think none of these languages was ever mentioned. The papy­
rology of the rest of the ancient world was hardly an issue, either; apart from Hercula­
neum, Dura-Europos, and a scattering of other texts, papyrology meant Egypt. The papy­
rological textbooks of that era, most notably Turner (1968, 19802) and Montevecchi

Page 1 of 5
Introduction

(1973, 19882), are essentially and even avowedly about the Greek (and Latin, to some ex­
tent) papyri of Egypt, just as had been the case already for Mitteis and Wilcken (1912),
and the same is explicitly true of Rupprecht (1994).

Today, a broader concept, already partly visible long ago in Peremans and Vergote
(p. xviii)

(1942), is unavoidable. One may trace the change in the Checklist of Editions, which be­
tween its first edition in 1974 and its most recent in 2001 (Oates et al. 2001) has added
demotic and Coptic, and an analogous Arabic checklist has come into being (online). It
seems only a matter of time before the papyri in other Semitic languages are added. Will
the Bactrian documents (Sims-Williams 2000) be next? Papyrologists trained on Egyptian
material have found themselves working on papyri from Petra and tablets from Vindolan­
da. Several volumes of one papyrological series have now been titled “From Herculaneum
to Egypt ” (Papyrologica lupiensia). All of this has in some ways not so much left behind
the old contest between methodological and substantive concepts of the field of papyrolo­
gy as relocated them to a broader plane.

It is, however, all too easy to see these developments uncritically as the papyrological
manifestations of the egalitarian, multicultural spirit of the present. No matter how fuzzy
a set papyrological texts constitute, they do have a core. Greek is still the dominant lan­
guage of papyrology, and the Roman empire its fulcrum. Nearly 80 percent of published
papyri are Greek and Latin (mostly Greek; cf. chapter 27), texts from the period of Roman
rule greatly outnumber those of the Hellenistic period, and the numbers among the un­
published may not be vastly different. The “normality” of the Roman period for papyrolo­
gy is probably not just a matter of the chance of survivals, however; or, to look at it from
another point of view, the survival of documents is probably not simply the product of ar­
chaeological contingency. Roman rule brought with it the development of a society of “no­
tables,” the prosperous elites of both villages and cities who governed them—the cities
especially after Septimius Severus granted them city councils. These groups, the property
they owned, and the public duties they carried out generated an immense amount of pa­
perwork, much of which had not been there in the Ptolemaic period, and these papyri are
a large part of what gives us our impression of the “middle-class” (but really upper mid­
dle or lower upper class) society to which the modern middle-class reader connects so
easily. It is the village societies of the Fayyum and the bourgeoisie of Oxyrhynchos that
have generated most of the stories papyrologists tell about life in Graeco-Roman Egypt.
Greek was the language of power and business in these societies.

The Roman Empire—in an expansive sense, including late antiquity—is also the period in
which the geographical range of papyrological finds outside Egypt is at its greatest. From
the first to the early second centuries there are important finds from the pre-Hadrianic
forts at Vindolanda in northern Britain (Tab. Vindol. I–III), with their snapshot of frontier
military life, and the fort of Masada by the Dead Sea, where, near the other end of the
empire, the Roman army was engaged in putting down a rebellion (Doc.Masada). Second-
and third-century documents from the Dead Sea (P.Yadin) and the Euphrates valley
(P.Euphr., P.Dura) have also helped prevent too Egyptocentric a view of the papyrological
world, as the interplay of (p. xix) Roman, Greek, and local languages and legal norms has

Page 2 of 5
Introduction

given more specificity, bite, and controversy to questions all too easily buried in general­
izations. The army is documented again in third-century Libya with a large find of ostraca
(O.Bu Njem). Later still, Petra and Nessana give us city and village documents linked to
church and military but also highly revealing about private property transactions in the
sixth and seventh centuries (P.Petra, P.Ness.). Yet none of this takes away from the over­
whelming numerical dominance of Egyptian texts.

This handbook reflects these changes in papyrology over the last third of a century; it al­
so reflects the lack of any universally accepted view of the discipline to replace the con­
sensus of the past. The Greek papyri still dominate the book, just as they do the subject.
Limitations of space, differences in the developmental stages of various fields, and some­
times a lack of available contributors have made it impossible to treat all possible sub­
jects. I particularly regret the absence of any substantial discussion of Coptic palaeogra­
phy, a subject much in need of systematic treatment, and the lack of a planned chapter on
hieratic and demotic papyri (although chapters 12 and 17 deal with part of that territory).
Fortunately, these topics will be treated extensively in the forthcoming Oxford Handbook
of Egyptology and Oxford Handbook of Coptic Studies. Readers should in any case recog­
nize that any seeming incoherences of boundaries and coverage accurately reflect the na­
ture of papyrology today in the midst of change.

The divide between the methodological and substantive sides of the discipline will also be
evident. Some chapters are more practical in character, aiming to help the reader under­
stand how papyrologists go about reading, editing, and making sense of their texts. Oth­
ers give some of the results of that process. This divide too was evident in Peremans and
Vergote's Handboek, which contained an entire chapter on the definition of the subject,
then other chapters on writing material, conservation, and decipherment, as well as chap­
ters on political history, language, administration, law, religion, social life, economy, cul­
ture, and private life. The balance is clearly toward the results of papyrology, perhaps not
a surprising outcome in a book written by two scholars who were not editors of papyri. If
the present handbook attempted to cover the full range of these subjects, it would have
required at least two volumes (if it could have been produced at all). It has no sections on
class, ethnicity, economy, trade, gender, family, Hellenization, Romanization, and many
other subjects on which a great deal of good work has been done in recent decades.
Space has been used instead to widen the linguistic range and break “religion” out into
more of its varied constituents. This was hardly an inevitable choice, but it seemed to me
more important to cover papyrology's development into those directions, even if incom­
pletely, than to try to provide a history of Egypt (let alone the entire ancient world)
through the lens of the papyri.

As a collective work, this handbook has of necessity a different character from previous
handbooks or textbooks of papyrology. The twenty-seven authors represented here and
their subjects overlap from time to time, and they do not agree (p. xx) about everything.
Although some repetition has been excised, some remains, and contention remains, too.
There would be no point in pretending that all of the authors speak with the same voice.
One of the purposes of a multiauthor volume of this kind, in fact, is to give the reader a

Page 3 of 5
Introduction

sense of the debates that animate the field. Moreover, different authors have different
conceptions of their audience; that again seems to me inevitable in such a work and per­
haps even desirable. Most of the chapters require no knowledge of any ancient language,
but it was hard to imagine a chapter on the Greek and Latin of the papyri addressed to an
audience that knew nothing of either language.

Handbooks tend to be consulted or read in part rather than continuously. Many different
arrangements of the chapters could have been envisaged, naturally; the one adopted here
made sense to me, but nothing prevents readers from reading chapters in any order they
find helpful.

This is certainly the first papyrological handbook in which electronic research tools play a
significant part. There are few chapters not marked in one way or another by the avail­
ability of major resources in digital form, mainly on the World Wide Web but some still on­
ly on CD-ROM. The authors have somewhat diverse things to say about this revolution,
and I have thought that here particularly some repetition was a good thing. The address­
es of these tools are given above (pages xv–xvi), where the reader will find all of these re­
sources listed with information on access to them.

This book has benefited from the help of many individuals. I want to acknowledge particu­
larly the valuable comments of the participants in the Summer Seminar in Papyrology
held at Columbia University in 2006, who had drafts of the volume available to them. Ed­
uard Iricinschi, of Princeton University, read the entire copyedited volume and improved
it in many particulars, a service for which I am deeply grateful. The financial support of
the Andrew W. Mellon Foundation and of the Institute for the Study of the Ancient World
of New York University has made possible the seminar and this editorial work.

Bibliography

Editions of papyri and papyrological reference works are cited throughout this volume ac­
cording to the abbreviations in Oates et al. (2001) or its electronic version.

Mitteis, L., and U. Wilcken. 1912. Grundzüge und Chrestomathie der Papyruskunde. 4
vols. Berlin.

Montevecchi, O. 1988. La papirologia, 2d ed. Milan: Vita e pensiero.

Oates, J. F., R. S. Bagnall, S. J. Clackson, A. A. OʼBrien, J. D. Sosin, T. G. Wilfong, and K. A.

Worp. 2001. Checklist of Editions of Greek, Latin, Demotic, and Coptic Papyri, Ostraca,
and Tablets, 5th ed. Oakville, Conn.: American Society of Papyrologists. For up-to-
(p. xxi)

date electronic version, see p. xv.

Peremans, W., and J. Vergote. 1942. Papyrologisch Handboek. Leuven: Beheer van Philolo­
gische Studiën.

Pestman, P. W. 1990. The New Papyrological Primer. Leiden: Brill.

Page 4 of 5
Introduction

Rupprecht, H.-A. 1994. Kleine Einführung in die Papyruskunde. Darmstadt: Wis­

senschaftliche Buchgesellschaft.

Sijpesteijn, P. M., and L. Sundelin, eds. 2004. Papyrology and the History of Early Islamic
Egypt. Leiden: Brill.

Sims-Williams, N. 2000. Bactrian Documents from Northern Afghanistan. Vol. 1: Legal and
Economic Documents. Studies in the Khalili Collection; vol. 3: Corpus inscriptionum irani­
carum, II.6. Oxford: Nour Foundation.

Turner, E. G. Greek Papyri: An Introduction, 2d ed. 1980. Oxford: Clarendon Press.

Page 5 of 5
Bibliography

Bibliography
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011 Subject: Classical Studies

Online Publication Date: Sep 2012

(p. xxii)

Page 1 of 1
The Oxford Handbook of Papyrology

The Oxford Handbook of Papyrology

The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011 Subject: Classical Studies

Online Publication Date: Sep 2012

(p. 1) The Oxford Handbook of Papyrology (p. 2)

Page 1 of 1
Writing Materials in the Ancient World

Writing Materials in the Ancient World

Adam Bülow-Jacobsen
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0001

Abstract and Keywords

Paper did not exist in the ancient Mediterranean world. Instead, people wrote on an enor­
mous variety of other materials. While almost every substance imaginable has been used
as writing material at one time or another, this article focuses on the common ones. First,
it considers papyrus since the overwhelming majority of ancient texts are written on this
material. It discusses parchment, ostraca, and wooden tablets which receive considerable
attention. It also discusses linen (e.g., mummy bandages) and stone (mainly Coptic lime­
stone ostraca inscribed with ink). Looking at Coptic documentary texts, which extend
past the end of antiquity, ostraca are the most important medium (47.5%), while papyrus
is second (40.5%). Limestone accounts for 10.5%, while skin (leather/parchment), paper,
and wood represent less than 1% each.

Keywords: paper, ancient Mediterranean world, papyrus, parchment, ostraca, wooden tablet, Coptic documentary
texts, limestone

Paper, as we know it today, did not exist in the ancient Mediterranean world.1 Instead,
people wrote on an enormous variety of other materials. While almost every substance
imaginable has been used as writing material at one time or another,2 in this chapter I fo­
cus on the common ones. First, I naturally consider papyrus since the overwhelming ma­
jority of ancient texts are written on this material Parchment, ostraca, and wooden tablets
also receive considerable attention, while linen (e.g., mummy bandages) and stone (main­
ly Coptic limestone ostraca inscribed with ink) receive minimal attention

An overall view of the use of various writing materials for Greek documentary texts can
easily be acquired from the Heidelberger Gesamtverzeichnis der griechischen Pa­
pyrusurkunden Ägyptens einschlieβlich der Ostraka usw., der lateinischen Texte, sowie
der entsprechenden Urkunden aus benachbarten Regionen (hereinafter HGV).3 Out of a
total (as of April 2004) of 54,312 published documents, the distribution on writing materi­
als is given in table 1.1. In columns 4 and 5 I have added the Wgures and percentages for

Page 1 of 28
Writing Materials in the Ancient World

literary texts, which are taken from the total of 9,875 items incorporated in the Leuven
Database of Ancient Books (hereinafter LDAB).4

The aforementioned figures are for texts in Greek and Latin If we look at Coptic docu­
mentary texts,5 which extend past the end of antiquity, ostraca are the most important
medium (47.5 percent), while papyrus is second (40.5 percent). Limestone accounts for
10.5 percent, while skin (leather/parchment), paper, and wood represent less than 1 per­
cent each. (p. 4)

Page 2 of 28
Writing Materials in the Ancient World

Table 1.1. The Frequency of Various Writing Materials

Material Documents % of Documents Literary Texts % of Literary Texts

papyrus 35,591 65% 7,100 71%

ostraca1 15,195 28% 339 3%

wood2 2,500 5% 148 1%

parchment 349 0.6% 2,575 25%

graffiti 234 0.4%

linen 84

wax tablet 73

stone 67 30

cloth (mummy linen, 30

etc.)

leather, etc. 25

various semiprecious 9
stones

Page 3 of 28
Writing Materials in the Ancient World

limestone 7

bone 7

gold and silver 6

bronze 6

lead 1

iron 1

schist 1

reed 1

(1.) This category includes fragments of ceramic on which the text is written in ink or engraved after firing.

(2.) A good many of these are mummy labels. One could also include wax tablets and the description Klapptafel (4 items in HGV). For
reasons of geography the HGV includes neither the Vindolanda (T.Vindol., 853 items) nor the Vindonissa tablets (T.Vindon, 90 items).
Otherwise, the wooden tablets would account for 6 percent of the total.

Page 4 of 28
Writing Materials in the Ancient World

Papyrus
Where and How Did Papyrus Grow?

According to Theophrastus, whom Pliny copies without acknowledgement,6 papyrus grew

in water no deeper than 90 centimeters (two cubits). Pliny adds on his own account (or
from another source) that it grows in the swamps of Egypt, which are stagnant after the
floods. This was certainly true in his day and at the time of (p. 5) Theophrastus as well,
but before man interfered with the floods of the Nile, where would papyrus have grown in
Egypt? Without human intervention, the floods did not leave stagnant pools for very long,
and the annual change in the water level would not have agreed with the papyrus plant. It
is thus no surprise that Cyperus papyrus L. died out in Egypt when it was no longer culti­
vated for paper making, and I suspect that it never grew there spontaneously but was im­
ported from tropical Africa at a very early period. In the 1960s a surviving specimen was
discovered in Wadi Natrun but typically not on the Nile.7 The papyrus that now grows in
Egypt and is used by modern papyrus makers was imported from the Jardin du Luxem­
bourg, Paris, in 1872 and planted in front of the Egyptian Museum in Cairo. From there, a
number of plants came to the Cairo Zoo and were in turn transplanted by Dr. Hassan
Ragab to his plantation on Jacob's Island. The origin of these plants appears to be Syra­
cuse, where papyrus was introduced (or rediscovered) by the Arabs during the Middle
Ages. In any case, the origin of the plants must have been Egypt.8 Pliny (HN 13, 72–73)
also mentions papyrus in Syria and Mesopotamia.

How was Papyrus Made?

We have three sources of information on papyrus making: analysis of ancient papyri, an­
cient descriptions, and modern experiments with manufacture. If we start with the sim­
plest form of analysis, looking at a piece of papyrus paper, it is obvious that it is made of
two layers of fibers placed perpendicularly to one another. As for descriptions, I consider
that we have no good description from antiquity of how papyrus was made. The Egyptians
apparently never recorded the process, and the only classical author who describes it is
Pliny (HN 13, 74–82), whose account is problematical in several ways. The principles of
textual criticism dictate that we try to reconstitute what the author wrote, but our natural
tendency is to try to make sense of what Pliny wrote since we tend to assume that he
knew what he was talking about. This is, however, not necessarily the case, since Pliny
had never been to Egypt and papyrus paper must be made from fresh papyrus; thus, it
can be made only where papyrus grows. It is therefore almost certain that he had never
witnessed the manufacture of a papyrus sheet, and it is consequently difficult to deter­
mine how we should deal with the obvious shortcomings of Pliny's text. He must have
been excerpting a written source, but we have no idea what it was or whether it was cor­
rect. So, in general, emendations of Pliny's text should be avoided. What we can do (and
what several commentators, including myself, have done) is to try to interpret the words
in such a way that they can be harmonized with what we believe to be the truth. This pro­
Page 5 of 28
Writing Materials in the Ancient World

cedure contains an obvious danger of circular argumentation. To complicate matters, I

believe that Pliny is mixing personal experience of papyrus he had bought and used in
Rome with whatever source he was using to describe its manufacture, and I believe that,
while his source (p. 6) must have given an account of papyrus making in pharaonic times,
his own experience was, of course, of papyrus as produced in his own day.9

I quote a translation of the relevant passage, adapted from Lewis (1974: 37–41):

74. Paper is made from the papyrus plant by separating it carefully10 into
very thin strips as broad as possible. The choice quality comes from the mid­
dle, and after that come the other cuts in order. The (choice) quality, in for­
mer times called “hieratic” because it was devoted only to religious books
has, out of Xattery, taken on the name of Augustus, and the next quality that
of Livia, after his wife, so that the “hieratic” has dropped to third rank.
75. The next had been named “amphitheatric” from its place of manufacture.
At Rome, Fannius' clever workshop took it up and refined it by careful pro­
cessing, thus making a first-class paper out of a common one and renaming it
after him; the paper not so reworked remained in its original grade as “am­
phitheatric.”
76. Next is the “Saitic,” so called after the town where it is most abundant,
made from inferior scraps, and, even more like bark, there is the “Taeneotic,”
named after a nearby place (this is sold, in fact, by weight, not by quality).
The “emporitic,” being useless for writing, provides envelopes for papers and
wrappings for merchandise, and its name accordingly comes from [the Greek
for] merchants. After this there is the end of the papyrus stalk, which is simi­
lar to a rush and useless even for rope except in moisture.
77. Paper of whatever grade is fabricated on a board moistened with water
from the Nile: the muddy liquid serves as the bonding force. First there is
spread flat on the board and quite straight a layer consisting of strips of pa­
pyrus of whatever length they maybe. When the ends are squared of a cross
layer completes the construction. Then it is pressed in presses, and the sheets
thus formed are dried in the sun and joined one to another, in declining order
of excellence down to the poorest. There are never more than twenty sheets
in a roll.
78. There is great variation in their breadth, the best thirteen digits, the “hi­
eratic” two less, the “Fannian” measures ten, the “amphitheatric” one less,
the “Saitic” a few less—and it is not strong enough for malletting—and the
narrow “emporitic” does not exceed six digits. Beyond that, the qualities es­
teemed in paper are fineness, firmness, whiteness, and smoothness.
79. The Emperor Claudius changed the order of preference. The “Augustan”
paper was too thin for writing with a pen; in addition, as it let the ink through
there was always the fear of a blot from the back, and in other respects it was
unattractive in appearance because excessively translucid. Consequently the
vertical (under) layer was made of second-grade material and the horizontal
layer of first-grade. He also increased its width to measure a foot.
Page 6 of 28
Writing Materials in the Ancient World

80. There was also the “macrocolumn,” a cubit wide, but experience revealed
the defect that when one strip tears off it damages several columns of writ­
ing. For these reasons the “Claudian” paper is preferred to all others; the “Au­
gustan” retains its importance for correspondence, and the “Livian,” which
never had any first-grade elements but was all second-grade, retains its same
place.
(p. 7) 81. Rough spots are rubbed smooth with ivory or shell, but then the

writing is apt to become scaly: the polished paper is shinier and less absorp­
tive. Writing is also impeded if (in manufacture) the liquid was negligently ap­
plied in the first place; this fault is detected with the mallet, or even by odour
if the application was too careless. Spots, too, are easily detected by the eye,
but a strip inserted between two others, though bibulous from the sponginess
of (such) papyrus, can scarcely be detected except when the writing runs—
there is so much trickery in the business! The result is the additional labour
of reprocessing.
82. Common paste made from finest flour is dissolved in boiling water with
the merest sprinkle of vinegar, for carpenter's glue and gum are too brittle. A
more painstaking process percolates boiling water through the crumb of leav­
ened bread; by this method the substance of the intervening paste is so mini­
mal that even the suppleness of linen is surpassed. Whatever paste is used
ought to be no more or less than a day old. Afterwards it is flattened with the
mallet and gone over with paste, and wrinkles are again removed and
smoothed out with the mallet.

In general this description corresponds well to what we may deduce from observation of existing
papyri, but a few obscure points remain. The papyrus stalk was harvested and cut into sections,
separating sections from the lower, the middle, or the upper parts. Pliny becomes a little confus­
ing when describing the qualities resulting from these various cuts because the criteria for the
qualities combined both the firmness and opacity of the writing material and the width of the
sheets. The lower part of the stalk contains relatively more pulp between the fibers than the
higher part, so the sections from the lower part of the stem produce a thinner papyrus sheet
than the middle.11 Because of the change of writing implements from reed brush (as used for
Egyptian) to reed pen (κάλαμος) (as used for Greek and Latin), the very fine papyrus favored in
pharaonic times was less attractive for the Greeks and Romans.12 However, the qualities also dif­
fered in the width of the individual sheets.13 When the papyrus was sold in roll form, one asked
for a roll of a given quality, and since the width of the twenty sheets was fixed, the length of the
roll (i.e. twenty sheets of the width appropriate to that quality) was also known for every quality.
The height, on the other hand, could vary. The somewhat confusing statement in 77 would give
the impression that every roll contained all the qualities, which of course is nonsense. What
Pliny means is that the best sheets of the quality in question were put first in the roll for the cus­
tomer to see, rather as strawberries tend to be arranged for sale in the punnet. 14
Another point that may need some explanation concerns the procedures of “Fannius's
clever workshop.” The posttreatment of Fannius has long excited commentators. Pliny
does not tell us what the method involved and the reason maybe that he did not know.
The only thing Pliny does say is that the sheets or rolls were made larger. Fannius pre­

Page 7 of 28
Writing Materials in the Ancient World

sumably guarded his professional secret. Lewis “speculates within (p. 8) the bounds of
reason” that Fannius may have added a third layer of better quality in order to produce a
better writing surface.15 I find it difficult to see how this would enlarge the sheets. C. H.
Roberts is quoted by Lewis for a similar idea, namely that the original papyrus was split
and a layer of better quality was substituted as writing surface.16 Again, I do not see that
this would enlarge the sheet/roll. Besides, such a procedure would have been difficult, not
to say impossible. If I, too, may be allowed to speculate within reason, I believe that the
only way to make an existing sheet or roll larger is to beat it with a mallet.17 This would
inevitably make a dry papyrus sheet more brittle, but if the sheet was first moistened, it
might be possible to increase its size by about 10 percent while making the paper thinner.
The main risk when moistening papyrus, as all restorers know, is that the ink may run,
which is not pertinent in this case. Anyone who has tried his hand at restoration will have
noticed that the fibers regain much of their original flexibility when wet. In fact, I believe
that in 78 Pliny is telling us that the paper was hammered out; he writes that the Saitic
quality is even smaller nec malleo sufficit (and is not strong enough to be malleted). Why
else would he mention the mallet in connection with the size?

Modern Experiments

Figure 1.1. An ancient papyrus on the lightbox

(P.Sorb. inv. 2245). There are no overlaps or bare
patches between the papyrus strips. We clearly see a
kollêsis somewhat to the left of the middle of the im­
age. Photo by Adam Bülow-Jacobsen.

Page 8 of 28
Writing Materials in the Ancient World

Figure 1.2. Diagrams from Hendriks's original arti­

cle: 1. the initial cut into the triangular stem; 2. the
peeling schematized on a cross-section of the stalk;
3. the peeled section; 4. a peeled section seen from
the edge. Reproduced courtesy of Habelt Verlag.

Figure 1.3. Papyrus made by Hendriks's peeling

method. Overlapping between the strips has been
avoided, but there are far too many holes in the
sheet where the fibers have shrunk while drying.
Photo by Adam Bülow-Jacobsen.

Figure 1.4. Papyrus made by H. Ragab. The overlaps

between the strips are much too obvious. Photo by
Adam Bülow-Jacobsen.

Page 9 of 28
Writing Materials in the Ancient World

The best-known modern experiments are those of Hassan Ragab, Cairo, and Corrado
Basile, Syracuse. Both have produced papyrus of a useable quality, and both are sure they
have recreated the ancient procedure, although it is obvious to anyone who handles their
paper that something is wrong. The few examples I have seen of the Sicilian papyrus are
very soft, white, and pliable but do not feel like papyrus at all. The Ragab papyrus feels
like ancient papyrus but has the characteristic “grid pattern,” that is, the individual strips
are seen very clearly, which is not the case with ancient papyrus. The problem is whether
to place the strips side by side (with the risk of gaps forming between them as they dry),
or placing them with an overlap, as Ragab did, thus producing the grid pattern. Pliny's
description (given earlier) does not mention any overlap, and the ancient papyri do not
show any grid pattern. So we still do not know exactly how papyrus was made. In an at­
tempt to find a solution, I. Hendriks proposed that Pliny's diviso acu meant exactly that—
with a needle—and that the papyrus stalk was unrolled by the so-called peeling method.18
The theory created a certain amount of interest at the time,19 but as I have shown, Pliny's
text contains too many counterindications. Besides, having tried it myself, I know that a
papyrus stalk does not react kindly to being peeled. It breaks whenever one tries to “go
around a corner” in order to open the next side of the triangle, and using a needle instead
of a knife tends to tear the pulp. Besides, it has never been clear to me why using a nee­
dle would lead to the peeling method (see figures 1.1–1.4). (p. 9) (p. 10)

(p. 11) Parchment

In Latin, parchment was called pergamena (n. pl.) or, much more often, membrana (f.).
The most common Greek word is διφθέρα, but in the fourth century περγαμηνἠ (διφθέρα)
and δέ ρμα were also used.20 The word parchment comes from the name of the city of
Pergamon in Asia Minor, and the ancients believed that the use of untanned skins origi­
nated there. Pliny quotes Varro as the origin of the following well-known story: King
Ptolemy (V Epiphanes, 205–180 BCE) of Egypt and King Eumenes (II, 197–159 BCE) of
Pergamon competed on creating the best library. To thwart his adversary, Ptolemy
stopped the exportation of papyrus, and so the Pergamenes invented parchment.21 The
story is unlikely to be true, however, for skins were used for writing long before that peri­
od: Aramaic parchment documents from Bactria from the fourth century BCE have been
found (Shaked 2004), and documents on parchment from the early second century BCE
have been found at Dura-Europos.22

Contrary to papyrus, the method of making parchment is well known. Skins, mostly of
calf, goat, or sheep, are cleaned, scraped free of hair, stretched while drying, and treated
with alum and chalk.23 Parchment, or vellum, as it is also called, is different from leather
in that it is not tanned.

When looking at a parchment codex, it is a sobering thought that every double folio page
represents a whole sheep or goat.

Page 10 of 28
Writing Materials in the Ancient World

Wood
Wood in several forms was regularly used for writing.24 Wax tablets, wooden boards
(whitened or not), and concertina leaves are the most important of these. In Greek a
wooden tablet is called πίναξ, πινακίς, δελτίον, δελτίδιον, πvκτίov, or γραμματείον. In
Latin tabula or tabella is used, or, for a wax tablet, cera.

Wax Tablets

The surface of a wooden board was gouged out, leaving a border at the edge, and the hol­
low thus created was filled with beeswax. The writing was scratched into the wax with a
γραφίς (Latin stilus), which was a pointed pin of wood, bone, or bronze, whose opposite
end was normally formed as a spatula for smoothing out when the scribe wanted to cor­
rect something.25 Quintilian recommends writing on wax tablets, although older people
may have difficulties because of the low contrast between the writing and the back­
ground. Writing on parchment with a pen and (p. 12) ink, however, disturbs the flow of
thought—so Quintilian says—because of the frequent need to dip the pen. Also, he says, it
is easier to correct on wax tablets.26 Wax tablets were clearly the everyday notebook for
bookkeeping, business correspondence, and literary drafts. The problem is that the wax
does not often survive, and the writing is then preserved only in the scratchings left in
the wood underneath the wax.27 If holes were drilled in the edge and a string passed
through them, wax tablets could be arranged in a kind of codex. The “pages” between the
first and the last tablet could be hollowed out and waxed on both sides.

Wax tablets were often written in lines parallel to the long side of the tablet. Thus, when
they were bound together into a codex, the notebook would not open with a left and a
right page, but with an upper and a lower page.

Wooden Boards

A wooden board covered with white paint presents a very good writing surface for pen
and ink and must always have been used. We know that such boards, σανίδες, were used
in Athens for the publication of official texts, either impermanent ones or before they
could be carved in stone.28

In Egypt such boards, whitened or not, are found occasionally, first of all as mummy la­
bels; these are small wooden tablets (never whitened as far as I know) on which the name
of the deceased was written in pen and ink or very occasionally incised. The label was at­
tached to the mummy with a piece of string that passed through a hole in the label. La­
bels of similar design were also attached to sacks or baskets that were sent, for example,
to people working away from their families (figure 1.5).

Page 11 of 28
Writing Materials in the Ancient World

Figure 1.5. Ἀμμώνις Ἀμμωνίου. Wooden label with

its string intact (O.Claud. inv. 4271). Photo by Adam
Bülow-Jacobsen.

Figure 1.6a–b. A letter written on a folded leaf tablet

and a concertina tablet. From T.Vindol.I 3839. Repro­
duced by permission of A. K. Bowman.

The most spectacular wooden tablets are the codices from the oasis of Dakhla.29 These
recent finds are unique in their genre so far and also interesting because of their perfect
condition. They are sawn from a block of acacia wood, the (p. 13) (p. 14) norm apparently
being eight leaves from a block. The two outer leaves were sawn to a thickness of about 5
mm, while the inner leaves are 2–3 mm. The separated leaves were marked with notches
by the carpenter, so that the original order could be maintained. Holes were drilled at the
edge, and a string was passed through them. The boards are normally not whitened, but
an inserted leaf in the Isocrates codex is. The text is written in ink in lines parallel to the
short side of the tablets.

Wooden Leaf Tablets

This type of tablet is known from Vindolanda in northern England, where many have been
preserved in anaerobic and humid conditions. Apart from ordinary wax tablets, the site al­
so yielded these unique specimens. They are very thin slices (some as thin as 0.25 mm),
but most are 1–2 mm thick and are of alder or birch. The surface, where it is preserved or
Page 12 of 28
Writing Materials in the Ancient World

can be reconstructed, is 16–20 cm by 6–9 cm. If such a slice were to be used for a letter,
the lines of writing would normally be parallel to the longest side (thus parallel to the
grain of the wood) and in two columns. The leaf was then scored lightly in the middle and
folded, and it could be closed and sealed by a string drawn through holes near the left
and right edges. The address could be written on the outside.30 If the text was an ac­
count, the writing would often be parallel to the short side of the leaf (i.e., across the
grain of the wood). The tablets were again scored and folded, but if the account was a
long one, several such diptychs could be tied together to form a “concertina list” (figure
1.6).31

Ostraca
Potsherds were everywhere in the ancient world, since pots, although they can be reused,
cannot be recycled like glass or metal once they are broken.

We must distinguish several types of ostraca (in the modern usage of the term): (1) the
Athenian type; mostly black glaze (i.e., red-figure) pottery on which ostracisms were writ­
ten by scratching through the black glaze so that letters are shown by the pink pottery
below; (2) the ancient Egyptian type of flat limestone with writing in ink; (3) sherds of
broken pots written on with pen and ink (or brush and ink for the demotic ones); (4)
whole pots inscribed with the contents, the origin, the name of the recipient, or similar in­
formation.

Figure 1.7. Athenian ostracon (Kerameikos Muse­

um). Photo by Adam Bülow-Jacobsen.

The regular Greek word for ostracon is ỏστρακον, whereas Latin does not seem to have a
word that covers all the meanings of the Greek term. Testa or testula are used to translate
ỏστρακον in the Athenian sense (type 1 above) of a voting ballot. Ostracum is found very
occasionally in texts from Egypt. (p. 15)

Ostraca of the Athenian type are of course preserved under most climatic conditions,
while the other two types, even if the ostracon itself is preserved, need a relatively dry
climate if the ink is to remain legible (figure 1.7).

Ostraca of the Athenian type do not seem to have been used for purposes other than bal­
loting. Type 3 ostraca, on the other hand, were used for most kinds of writing in Egypt,
although they were considered a surrogate for papyrus.32 Obviously, ostraca were suited

Page 13 of 28
Writing Materials in the Ancient World

only for short texts and could not easily be archived, nor could they be bound together if
more than one was needed for a longer text, and letters on ostraca could not be sealed to
protect the text from prying eyes. In addition, they were much heavier than papyrus. Nev­
ertheless, all these disadvantages were outweighed by one important advantage: Ostraca
were completely free. In many places one only had to bend down and pick them up. How­
ever, in places like Mons Claudianus, where stonemasons were employed, we sometimes
find ostraca that were prepared for writing with much more care. In a suitable sherd,
holes were drilled to mark the circumference of the desired ostracon, and the worker
then carved out the writing ostracon using these holes as a guide. In this way one could
obtain a pleasant oval or a rounded square. Edges were then beveled, and the writing
surface often smoothed, presumably by polishing it in sand. Such ostraca were sometimes
washed and used again, but this shaping-procedure was exceptional and is not found in
sites where military personnel were predominant (figures 1.8 and 1.9).

Figure 1.8. Ostracon from Mons Claudianus in

preparation. Here the craftsman has chosen a piece
of an amphora that already has an inscription. The
project appears to have been abandoned because the
sherd broke. Photo by Adam Bülow-Jacobsen.

Figure 1.9. A good example of a shaped ostracon

(O.Claud. III 522, natural size). Photo by Adam
Bülow-Jacobsen.

Page 14 of 28
Writing Materials in the Ancient World

Figure 1.10. A giant ostracon (40.5 cm tall) using an

almost complete amphora to write a register of post
riders. (O.Krok I 1). Photo by Adam Bülow-Jacobsen.

The best-known use of Greek and demotic ostraca was for tax receipts, especially in
southern Egypt, but there is mounting evidence of their use for all kinds of writing in the
desert. In particular, the many Roman sites in the EasternDesert that have been excavat­
ed during the last twenty years continue to produce large amounts of ostraca and very
few papyri. This is not difficult to explain: Provisions of wine, salt fish, olives, oil, and
even pickled meat and fish for the people who lived and worked in the desert arrived in
jars, mostly the standard Egyptian amphora of about 6½ liters with pitch on the inside,
which may have been reused on site but were mostly broken (figure 1.10). So there was
never any shortage of ostraca. On the other hand, papyrus had to be brought from the
valley. (p. 16) Letters on papyrus that had arrived from the valley must also have been a
temptation when one was in need of kindling. In Coptic, ostraca were also used for tax re­
ceipts, but the great mass of surviving ostraca, which come from monasteries, contain let­
ters. It is striking how few Arabic ostraca have been found so far. (p. 17)

(p. 18) Ink (μέλαν, Atramentum)

In antiquity, ink was what we now call India ink: soot with a little gum arabic suspended
in water.33 Both in China and in pharaonic Egypt ink was kept in blocks, and the writing
brush could be inked directly on the block if a little water (spittle) was applied to its sur­
face. Since Greek and Latin were written with a calamus (a reed pen), which had to be
dipped into the ink, the scribe had to prepare a quantity of ink every day.34 The earliest
occurrence of metallic ink is from the third century BCE,35 but from the second century
CE and particularly from the third century onward, the mordant metallic inks make their
appearance. These are made from powdered gallnuts, a metallic salt (iron or copper),
gum arabic, and water. While the India inks do not fade, iron-gall inks turn from black to

Page 15 of 28
Writing Materials in the Ancient World

brown with time. They may fade to almost the same shade of brown as the papyrus and
become very difficult to read. The mordant quality of iron-gall ink makes it more penetrat­
ing, but may also eventually damage the papyrus or parchment.

Books in Antiquity: The Volumen and the

Codex
Book (liber, βιβλίον), as far as the ancients were concerned, meant a roll (Lat. volumen).
Although the codex format was known at a fairly early time, it was not until the second
century CE that it really appeared in Egypt, but already in the fourth century the majority
of literary works were written on codices.

Figure 1.11. Bookroll with calamus and inkwell and

polyptych with stylus. (From Praedia di Iulia Felix,
Museo Nazionale, Napoli, inv. 8598). Photo by J.-P.
Brun.

The situation is well illustrated by Ulpian (†228 CE) commenting on Sabinus (first half of
the first century) and Gaius Cassius (mid-first century). The discussion is about what con­
stitutes “a book” when donated in a will:

Under the term books (librorum appellatione) are included all rolls, whether of pa­
pyrus or parchment or any other material. And even if they are of rind of the lime
or linden tree (as made by some) or of some other bark, the same must be said.
But are they due if they are in codex-form, either of parchment or papyrus or ivory
or some other material, or of waxed-tablets? Let us see. Gaius Cassius wrote that
[loose] parchments are due also, when books have been bequeathed. Therefore, it
follows that the others too will be due, unless this is contrary to the testator's in­
tentions.36

So, in Rome, in the first century of our era, a jurist's response was required to decide whether a
codex was a book. Yet, when Ulpian wrote in the early third century, the codex was gaining
steadily on the roll and in another century would (p. 19) replace it almost completely. Why would
that be? Before trying to answer this question, we must look at the anatomy of the roll and the
codex respectively.

Page 16 of 28
Writing Materials in the Ancient World

The Bookroll

The roll was the normal unit in which papyrus was produced and sold.37 As we have al­
ready seen from Pliny HN 13.77, the papyrus sheets were pasted together, twenty at a
time, and sold as rolls. When a scribe wanted to write a document, he cut a sheet of an
appropriate size from the roll, but when writing literature, the scribe presumably used
the roll as it was. If the length of the work he was transcribing did not correspond to the
length of the roll—and there was no reason it should—he would add on or cut off in order
to obtain the right length.

Figure 1.12. A muse reading from a bookroll (Attic

red-figure lekythos from ca. 435–425. Louvre, Collec­
tion Pozzi, inv. CA 2220). Photo by RMN/H.
Lewandowski, courtesy of the Musée du Louvre.

As Pliny has told us, papyrus was commercialized as rolls made up of sheets pasted to­
gether. The reason for this was probably that the individual sheets would each present
four edges, and the edges are the weakness of papyrus, always presenting a risk of fray­
ing. Pasted together into a roll, the twenty sheets would present only four edges in all,
and additional measures were taken to protect the ends. At the beginning of the roll was
the protokollon, an unwritten sheet, while at the end there was probably the ὀμφαλός or
umbilicus, the wooden stick around which the papyrus was rolled, but even if no umbili­
cus was present, the end was protected inside the roll. The sheets of the roll were pasted
together in such a way that the left sheet was always over the right one in any given join.
The joins are called kollêseis (singular kollêsis). If the roll was to be used for demotic writ­
ing (from right to left), it was turned 180 degrees. In this way the writer would always
write “downward” over the join and feel a minimum of resistance when passing over a
“step.” The face used first was always the inside of the roll, where the fibers (p. 20) (p. 21)
were parallel to the length of the roll and to the lines of writing. This is not because it is
easier to write with the fibers rather than across them, nor is it normally because the sur­

Page 17 of 28
Writing Materials in the Ancient World

face on the “back” is less well suited for writing. Given a fragment of papyrus without
original edges, writing, or kollêsis, papyrologists (even experienced ones) will have trou­
ble telling which side is the front and which the back. When a kollêsis is present on a
fragment, it is easy to see which side is the front and which side is the back since the kol­
lêsis on the front makes a break in the fiber pattern. However, it is much more difficult to
see the join from the back, where the edge of the sheet follows the same direction as the
fibers (“vertical”). The reason for having the “horizontal” fibers on the inside and the
“vertical” on the outside was probably that vertical fibers would be squeezed together
and risk detachment if they were on the inside.

The writing would be on the inside, front, in columns (σελίδες, paginae) unless the docu­
ment was written transversa charta (i.e., “having turned the papyrus”), in which case it
would present one long column running down the roll with lines of writing across the
fibers (figure 1.13).38

For practical reasons, a bookroll could be written on one side only. When we find, as we
often do, that there is writing on both sides of a papyrus, we are dealing with an example
of reuse. Quite often a roll would be turned inside out when the primary writing was no
longer of interest, and the back could be used for further writing. Of 3,365 literary rolls
listed in the LDAB, more than 400 are examples of literature written on the back of docu­
mentary rolls that have been turned over. There are also several examples of demotic lit­
erature that was written on the back of Greek rolls, although demotists have a tendency
to consider the demotic text as the “recto” or the front, regardless of the fiber direction
and other evidence.

The height of the roll depended on the constituent sheets, not, as we have seen, on the
quality of the papyrus, and ranges from 15 cm to more than 40 cm (a height of 20–30 cm
is normal). The length of the bookroll was theoretically unlimited, and Ulpian mentions,
for the sake of the argument, the possibility of getting all forty-eight books of Homer onto
one roll.39 Ancient Egyptian rolls could be very long (the longest known exceeds 40 me­
ters), but most of these very long rolls are ornamental copies of the Book of the Dead,
meant to be buried with the deceased, not to be read in this world. Greek rolls were no
longer than 10–11 meters and generally much shorter, but few complete Greek rolls exist,
and the original length of a fragmentary roll is mostly a theoretical projection on the ba­
sis of letters per line and lines per column calculated against a known text.

Page 18 of 28
Writing Materials in the Ancient World

Figure 1.13a–b. Drawings of, respectively, a bookroll

and a “rotulus” written “transversa charta” (W. E. H.
Cockle and E. G. Turner in Turner 1978, 14, 33).

To modern people who are used to the codex format, the disadvantages of the roll seem
many: It does not readily contain more than part of a prose work, like, for example, a
book of an historian; it has to be rolled back when read; and it is difficult to refer to a pas­
sage. Besides, the roll is fragile. The edges fray, especially the lower edge, which may rub
against the reader's clothing, and the roll is easily torn. (p. 22)

Parchment sheets can be made into rolls, the best-known examples being the Hebrew
Torah rolls. However, the one example of a parchment roll that I know and which must
originate from Egypt is clearly an amateurish creation.40 Altogether I know of eighteen
parchment rolls in Greek or Latin, of which seven are Old Testament texts that were
clearly influenced by the Torah format; two are New Testament; and the remainder are a
few fragments of classical authors. However, none of these were unquestionably written
in Egypt, and most of them were (p. 23) probably not. I believe that these texts may give
us a glimpse of what books in the Pergamon library looked like.

The Codex

To understand the development of the codex,41 consider that, if the material is not pa­
pyrus but rather tablets or parchment, the bookroll is not a natural result. Neither tablets
nor parchment have frail edges that need protection, and they are more difficult to con­
catenate. Parchment rolls are sewn together, not glued. The “concertina” tablets from
Vindolanda (mentioned earlier) have been regarded as precursors of the codex but could
also have been an attempt to make a roll. Wooden or waxed tablets might also have been
concatenated like sheets in a roll.42

Page 19 of 28
Writing Materials in the Ancient World

The format adopted when a longer text was to be written on tablets or parchment was the
codex (caudex, pugillares, membranae), which began its career far from the bookroll's
world of classical literature. Letters, drafts, and accounts were routinely written in this
form, not least, of course, outside Egypt, where papyrus was less easily obtained.

A natural way to link tablets together is to bore holes in one edge and bind them with a
piece of string or a leather thong. This way, both sides of the tablet are useable, the inner
surfaces are protected, the “book” can be sealed if it contains a letter, and it is easily
transportable. Such books, with as many as fifteen leaves (thirty pages), are well known
from a number of places in the Roman Empire. The special case of the Dakhla tablet
books has already been described, but waxed tablets were undoubtedly more common.

Figure 1.14. Reconstruction of a polyptych from “Vil­

la dei papiri” in Herculaneum. From Capasso, “Le
tavolette della Villa dei papiri ad Ercolano” in Lalou,
ed. (1992). Reproduced courtesy of M. Capasso.

Latin authors also mention notebooks made of parchment, called membranae.43 The point
of departure here would be a large sheet of parchment that was folded and cut at the
edges, precisely like modern printed books before the (p. 24) bookbinders began to do it
for us. A sheet folded once in each direction will produce four leaves or eight pages, a
“quarto” format. Folding once more makes an “octavo” format of eight leaves, and so on.
These folded sheets, called “quires” in English (derived from Latin quaternio [a set of
four]), are then sewn together with other quires to form a codex. By following this proce­
dure one automatically obtains the aesthetically pleasant effect that any opening of the
finished, cut book presents two pages of “flesh side” or two pages of “hair side.”

All this is quite different, however, if you want to make your codex of papyrus, as would
be natural in Egypt.44 Here the starting point is the roll made of sheets pasted together.
This roll must be cut into sheets twice the width of the desired page and folded once in
the middle. The early papyrus codices were often made as “single quire” codices, in
which the cut sheets were placed in a pile (normally all with the front up), which was fold­
ed in the middle. This method put great stress on the outer leaves and produced an irreg­
ular and fragile front edge. Every possible method seems to have been tried, and, besides
the single-quire codices, there are papyrus codices that range from one to at least five

Page 20 of 28
Writing Materials in the Ancient World

sheets per quire. Eventually, however, a preference for the quaternio (four sheets per
quire) was established.45 The principle of facing pages having the same surface was, as
we saw, automatic with parchment, but with papyrus it was not. Apart from the single-
quire codices, a practice seems to have developed in which the outside leaf in a quire nor­
mally had horizontal fibers and the following ones alternated, so that an opening always
showed two facing pages with the same fiber direction. Sometimes the codex is well
enough preserved to permit reconstruction of the roll from which the leaves were cut (fig­
ure 1.15).

Figure 1.15. A reconstruction of the roll from which

the first four sheets for a codex were cut. As is evi­
dent, no account is taken of the original kollêseis.
From J. Scherer, Extraits des livres I et II du Contre
Celse dʼOrigène dʼaprès le Papyrus no. 88747 du
Musée du Caire, Cairo (1956).

The competition between the roll and the codex lasted a couple of centuries but was
eventually completely won by the codex. It seems that the Christians took to the codex
with alacrity, perhaps because the roll was associated with classical elite, literary culture,
while the first Christians were mostly humble people who were more used to accounts
and business letters than to Homer and Aeschylus. Presumably they also wanted their
books to be different from the Jewish Torah rolls. The codex was also easier to refer to,
simpler to transport, and more economical since the back of the sheet could also be used.
As early as the second century, when the struggle had just begun, only about 4 percent of
1,772 papyri of classical literature were codices, whereas 75 percent of 37 Christian
works were codices. In the third century, 13 percent of classical texts were written in
codex form, while 75 percent of the Christian works were codices. In the fourth century
the codex had already claimed 64 percent of classical literature and 81 percent of Christ­
ian. By the fifth century 90 percent of classical and 95 percent of Christian literature was
in codex form (table 1.2). The era of the literary bookroll had definitely ended.46 (p. 25)

Table 1.2. The replacement of the volumen by the codex

Page 21 of 28
Writing Materials in the Ancient World

Source: Leuven Database of Ancient Books, http://www.trismegistos.org/org/ldab/.

Bibliography
Basile, C. 1977. “Metodo usato dagli antichi Egizi per la fabbricazione e la preservazione
della carta papiro.” Aegyptus 57: 190–199.

———. 1998. “New Discoveries concerning the Fabric of Papyrus.” Annales du Service des
Antiquités de lʼEgypte 73: 28–34.

Bierbrier, M. L., ed. 1986. Papyrus: Structure and Usage. London: British Museum.

Blanchard, A., ed. 1989. Les débuts du codex: Actes de la journée dʼétude (Bibliologia 9).
Turnhout: Brepols.

Brashear, W. M. 1997. “Egyptian Papyrus Then, Chinese Paper Today (or: of Mummies in
Maine and Tea from Cathay).” In Akten des 21. Internationalen Papyrologenkongresses,
ed. B. Kramer et al., vol. 1, 113–131. Stuttgart: Teubner.

———, and F. A. J. Hoogendijk. 1990. “Corpus tabularum lignearum ceratarumque Aegyp­

tiarum.” Enchoria 17: 21–54.

Bülow-Jacobsen, A. 1976. “Principatus medio. Pliny NH XIII, 72 sqq.” ZPE 20: 113–116.

———. 1985. “Magna in latitudine earum differentia” (Pliny NH XIII, 78). ZPE 60: 273–
274.

Cerny, J. 1952. Paper and Books in Ancient Egypt. London. (Repr. s.a. Chicago: Ares Pub­
lishers.)

Clarysse, W. 1993. “Egyptian Scribes Writing Greek.” CdÉ 68: 186–201.

Cockle, W. E. H. 1983. “Restoring and Conserving Papyri.” Bulletin of the Institute of

Classical Studies 30: 147–165.

Cuvigny, H. (ed.), Jean-Pierre Brun, Adam Bülow-Jacobsen, Dominique Cardon, Jean-Luc

Fournet, Martine Leguilloux, Marie-Agnès Matelly, and Michel Reddé. 2003. La route
(p. 28) de Myos Hormos: Lʼarmée romaine dans le désert oriental dʼEgypte. Praesidia du

Page 22 of 28
Writing Materials in the Ancient World

désert de Bérénice. 2 vols. Fouilles de lʼInstitut français dʼarchéologie orientale 48/1–2.

Cairo.

Delange, E., M. Grange, B. Kusko, and E. Menei. 1990. Apparition de lʼencre métallo­
gallique en Égypte à partir de la collection de papyrus du Louvre. Revue dʼÉgyptologie
41: 213–217.

Hendriks, I. H. M. 1980. “Pliny, Historia Naturalis XIII, 74–82, and the Manufacture of Pa­
pyrus.” ZPE 37: 121–136.

———. 1984. “More about the Manufacture of Papyrus.” Atti del XVII Congresso Inter­
nazionale di Papirologia, vol. 1, 31–37. Naples: Centro Internazionale per lo Studio dei Pa­
piri Ercolanesi.

Holwerda, D. 1982. “Plinius über die Anfertigung von ‘charta.’ ” ZPE 45: 257–262.

Howard, M. 1955. “Technical Description of the Ivory Writing-boards from Nimrud.” Iraq
17: 14–20.

Irmscher, J. 1997. “Moderne Papyrusproduktion: Siracusa.” In Akten des 21. Interna­

tionalen Papyrologenkongresses, ed. B. Kramer et al., vol. 2, 1113–1115. Stuttgart: Teub­
ner.

Johnson, R. R. 1970. “Ancient and Medieval Account of the ‘Invention’ of Parchment.” CS­
CA 3: 115–122.

Johnson, W. A. 1992. The Literary Papyrus Roll: Formats and Conventions. An Analysis of
the Evidence from Oxyrhynchus. PhD. diss., Yale University.

———. 1993. “Pliny the Elder and Standardized Roll Heights in the Manufacture of Pa­
pyrus.” Classical Philology 88: 46–50.

Kenyon, F. G. 1951. Books and Readers in Ancient Greece and Rome, 2d ed. New York:
Oxford University Press.

Lalou, É., ed. 1992. Les tablettes à écrire de lʼantiquité a lʼépoque moderne (Bibliologia
12). Turnhout: Brepols.

Leach, B., and J. Tait. 2000. “Papyrus.” In Ancient Egyptian Materials and Technology, ed.
P. T. Nicholson and I. Shaw, 227–253. Cambridge: Cambridge University Press.

Lewis, N. 1974. Papyrus in Classical Antiquity. Oxford: Clarendon.

———. 1981. “Open Letter to I. H. M. Hendriks and E. G. Turner (More on ZPE 39,1980,
113–14).” ZPE 42: 293–294.

———. 1989. Papyrus in Classical Antiquity: A Supplement (Papyrologica Bruxellensia 23).

Brussels: Fondation égyptologique Reine Elisabeth.

Page 23 of 28
Writing Materials in the Ancient World

———. 1992. “Papyrus in Classical Antiquity: An Update.” CdÉ 67: 308–318.

Lukaszewicz, A. 1997. “Diviso acu. Was a Needle Used in Papyrus Manufacturing?” Jour­
nal of Juristic Papyrology 27: 61–67.

Menci, G. 1988. “Fabbricazione, uso e restauro antico del papiro: Tre note in margine a
Plinio, NH XIII 74–82.” Proceedings of the XVIII International Congress of Papyrology, ed.
B. Mandilaras. 2 vols. Vol. 2, 497–504. Athens: Greek Papyrological Society.

Otranto, R. 1997.“Alia tempora, alii libri: Notizie ed elenchi di libri cristiani su papiro.”
Aegyptus 77: 101–124.

Pliny HN = Ernout, A. 1956. Pline lʼAncien Histoire naturelle Livre XIII. (Des Plantes exo­
tiques). Paris: Les Belles Lettres, Collection des Universités de France.

Quecke, H. 1975. Die Briefe Pachoms: Griechischer Text der Handschrift W. 145 der
ChesterBeatty Library. Regensburg: Pustet.

Ragab, H. 1980. Le papyrus [Contribution à lʼétude du papyrus (Cyperus papyrus. L) et à

sa transformation en support de lʼécriture (papyrus des anciens)]. Cairo: Dr. Ragab Pa­
pyrus Institute.

———. 1988. “The Quality of the Recently Manufactured Papyrus as Compared

(p. 29)

with Ancient Egyptian Papyrus.” In Proceedings of the XVIII International Congress of Pa­
pyrology, ed. B. Mandilaras. 2 vols. Vol. 2, 513–523. Athens: Greek Papyrological Society.

Roberts, C. H., and T. C. Skeat. 1983. The Birth of the Codex. London: Oxford University
Press.

Shaked, S. 2004. Le satrape de Bactriane et son gouverneur: Documents araméens du IVe

s. avant notre ère provenant de Bactriane (Persika 4). Paris.

Skeat, T. C. 1982. “The Length of the Standard Papyrus Roll and the Cost Advantage of
the Codex.” ZPE 45: 169–175.

———. 1990. “Roll versus Codex: A New Approach?” ZPE 84: 297–298.

———. 1994. “The Origin of the Christian Codex.” ZPE 102: 263–268.

———. 1995. “Was Papyrus Regarded as ‘Cheap’ or ‘Expensive’ in the Ancient World?” Ae­
gyptus 75: 75–93.

Speidel, M. A. 1996. Die römischen Schreibtafeln von Vindonissa: Lateinische Texte des
militärischen Alltags und ihre geschichtliche Bedeutung. Brugg: Veröffentlichungen der
Gesellschaft Pro Vindonissa.

Täckholm, V. 1974. Students' Flora of Egypt, 2d ed. Cairo: Cairo University.

Page 24 of 28
Writing Materials in the Ancient World

Tait, W. J. 1988. “Rush and Reed: The Pens of Egyptian and Greek Scribes.” In Proceed­
ings of the XVIII International Congress of Papyrology, ed. B. Mandilaras. 2 vols. Vol. 2,
477–481. Athens: Greek Papyrological Society.

Thompson, E. M. 1965. An Introduction to Greek and Latin Palaeography. New York: B.

Franklin. (Orig. pub. 1912.)

Turner, E. G. 1952. Athenian Books in the Fifth and Fourth Centuries B.C. London: H. K.
Lewis.

———. 1968. Greek Papyri: An Introduction. Oxford: Clarendon.

———. 1977. The Typology of the Early Codex. Philadelphia: University of Pennsylvania
Press.

———. 1978. “The Terms Recto and Verso: The Anatomy of the Papyrus Roll.” In Actes du
XVe Congrès international de papyrologie. Part 1, Papyrologica Bruxellensia 16, ed. J. Bin­
gen and G. Nachtergael. Brussels: Fondation égyptologique Reine Élisabeth.

———. 1980. “An Open Letter to Dr. I. Hendriks.” ZPE 39: 113–114.

Whitehorne, J. E. G. 1994. “A Postscript about a Wooden Tablet Book (P. Kellis 63).” In Pro­
ceedings of the 20th International Congress of Papyrologists, ed. A. Bülow-Jacobsen, 277–
283. Copenhagen: Museum Tusculanum Press.

———. 1996. “The Kellis Writing Tablets: Their Manufacture and Use.” JRA Suppl. 19, Ar­
chaeological Research in Roman Egypt: 240–245.

Wiseman, D. J. 1955. “Assyrian Writing-Boards.” Iraq 17: 3–13.

Notes:

(1.) Brashear (1997).

(2.) See, for example, Pliny HN 13, 69.

(3.) Http://www.rzuser.uni-heidelberg.de/~gv0/gvz.html.

(4.) Http://www.trismegistos.org/ldab/.

(5.) Alain Delattre, Banque de données des textes coptes documentaires, http://
dev.ulb.ac.be/philo/bad/copte/base.php?page=rechercher.php.

(6.) Theophr. HP 4, 8, 3; Pliny HN 13, 70.

(7.) Täckholm (1974, 790).

(8.) Ragab (1980, 52–53). Contrast, however, Ragab (1988, 514–515), who states that the
plants in his plantation came from the Sudan. Moreover, Basile (1998, 29) claims that,

Page 25 of 28
Writing Materials in the Ancient World

around 250 BCE, Hieron II had transported the plant from Egypt to Sicily, where it was
used only for cordage since the Sicilians did not know the secret of paper making.

(9.) Thompson (1965, 23): “His description applies specially to the system of his own day;
but no doubt it was essentially the same as had been followed for centuries.”

(10.) This translation is based on a correction of the text, which I propose with some hesi­
tation. The manuscripts have diviso acu (divided by a needle). Attempts to make sense of
this has led to various interpretations, such as Hendriks (1980) (discussed later) or
Łukaszewicz (1997), but all difficulties would disappear if we were to accept diviso ac〈〉
u〈rate〉 and assume a lacuna in the archetype.

(11) (.) Bülow-Jacobsen (1976).

(12.) Tait (1988). See also Clarysse (1993). Delange (1990) gives examples of demotic pa­
pyri written with carbon ink and a reed brush, while, in the same document, the Greek
subscription is written in metallic ink with a calamus.

(13.) Johnson (1993).

(14.) Cf. Lewis (1974, 53n27).

(15.) Lewis (1989, 21–22).

(16.) Lewis (1974, 45n13).

(17.) Lewis (1989, 22): “How did the ‘clever workshop’ make the papyrus thinner? By
malleting?”

(18.) Hendriks (1980).

(19.) See, for example, Turner (1980) but also Lewis (1981).

(20.) In the sense of “parchment book,” δέρμα appears in P.Ashm. inv. 3 (fourth century),
republished by Otranto (1997). I thank Simona Russo for this reference.

(21.) Pliny NH 13, 70.

(22.) Roberts and Skeat (1983, 5–7) do not believe that the story can be true; they cite ar­
guments from a thesis by Richard R. Johnson, “The Role of Parchment in Greco-Roman
Antiquity” (PhD. diss., University of California, Los Angeles, 1968), which I have not seen.

(23.) For references to more detailed descriptions, see Turner (1968, 9 and 9n41).

(24.) Lalou (1992) contains articles by experts on most kinds of tablets.

(25.) Quintilian (Inst. Or.X 4, 1) even claims that the erasing capacity of the stilus is at
least as important as its capacity to write.

(26.) Quintilian, Inst. Or., X, 3, 31.27. See, for example, T.Vindon. = Speidel (1996).

Page 26 of 28
Writing Materials in the Ancient World

(28.) See, for example, Andokides, περì τῶν μυστηρίων 83; Lysias XXVI 10.

(29.) See J. L. Sharp, “The Dakhleh Tablets,” in Lalou, ed. (1992, 127–148), and Sharp's
very full codicological descriptions in P.Kellis IV pp.17–20, and P.Kellis III pp.9–21.

(30.) A perfect example is T.VindolA II 310 (figure 1.6a). A similar type of folded leaf tablet
is P.Yadin 54.

(31.) See T.Vindol.I4.(figure 1.6b).

(32.) Cuvigny et al. (2003 II 470–473) present a more thorough analysis of the use of os­
traca than the space here permits. There are also quotations from unpublished ostraca
that present excuses for not writing on papyrus.

(33.) See also Cockle (1983, 150).

(34.) τò μέλαν τρίβειν (“to grind the ink”); see, for example, Demosthenes, De Corona
258.

(35.) See Delange (1990).

(36.) Justinian, Digesta 32.1.52, trans. A. Watson.

(37.) The most recent and comprehensive description of the bookroll is Turner (1978).

(38.) This was apparently customary when writing to the senate in Rome; cf. Suetonius,
Div. Iul. 52.6.

(39.) Justinian, Digesta. 32.1.52.1.1. On a conservative estimate such a roll would have
been a monster of about 140 meters.

(40.) Chester Beatty Library, inv. W 145. See the description in Quecke (1975). See also
P.Köln IV 174, which is part of the same roll and where the full bibliography may be
found.

(41.) Greek κῶδιξ, but no proper Greek word seems to have existed.

(42.) This format is in fact found in Nimrud, where waxed ivory tablets of the late eighth
century BCE were hinged together to make a concertina (Wiseman 1955; Howard 1955).
These waxed tablets were made of wood and ivory and contain writing on both sides.
Wiseman (1955, 6–8) appears to assume that both the front and the back were used for
the same text.

(43.) Passages illustrating books and reading are conveniently collected in Kenyon (1951:
121–134).

(44.) To me, there is little doubt that the papyrus codex is derived from the parchment
codex, but the great specialists on the matter, Roberts and Skeat (1983), see it differently.

Page 27 of 28
Writing Materials in the Ancient World

Two chapters of their book are devoted to various theories about both this and the Christ­
ian preference for the codex.

(45.) For makeup and statistics on the early codex see Turner (1977).

(46.) LDAB.

Adam Bülow-Jacobsen

Adam Bülow-Jacobsen, Former Carlsberg Research Professor, University of Copen­

hagen.

Page 28 of 28
The Finds of Papyri: The Archaeology of Papyrology

The Finds of Papyri: The Archaeology of Papyrology

Hélène Cuvigny
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0002

Abstract and Keywords

From the first Fayyum find until the First World War, the period of miraculous discoveries
lasted barely forty years. In the present era of slavery to the mass media, A. S. Hunt was
right in noting that the benefactors and institutions that provide financial support do not
have an understanding of archaeology much different from that of the nineteenth century.
Treasure hunting and necrophilia are still the order of the day. It is better to find a royal
mummy than a papyrus and better to find a papyrus than domestic garbage, however
great its informational value. Except for the special case of the Eastern Desert, it is no
longer possible to be sure where to look in order to find texts, but it is still quite possible
that extended excavation of what is left of urban conglomerates will produce impressive
finds.

Keywords: Greek papyri, First World War, archaeology, treasure hunting, necrophilia, papyri deposits, A. S. Hunt

From “Smash-and-Grab” Jobs to Subtle Stratig­

raphy
The Time of the Consuls

Until Napoleon Bonaparte's expedition to Egypt (1798–1801), the papyri had slumbered
undisturbed in their tombs and the ruins of ancient settlements. There were only rare ex­
ceptions, like the roll that came to be known as Charta Borgiana (reputedly found with
fifty others), which was bought by an anonymous Italian merchant at Giza in 1777

Greek papyri completely escaped the attention of the scholars of the French expedition,
who brought back only papyri written in Egyptian This (p. 31) military campaign, howev­
er, precipitated Egypt 's entry into the modern world. The expedition left behind political
chaos, out of which emerged Mohamad Ali, who reigned from 1805 to 1848 and opened
the country to Western influence. France and England then entered into a devious power

Page 1 of 26
The Finds of Papyri: The Archaeology of Papyrology

struggle. Merchants, diplomats, spies, adventurers, tourists, engineers, and technical

counselors were everywhere in Egypt. They discovered an archaeological El Dorado and
became antiquities hunters and dealers. Some of them (not least the consuls) financed ex­
cavations, whether from passion or greed, and assembled fabulous collections, which they
sold in Europe, especially to museums.1

Frédéric Cailliaud, a French mineralogist who had been sent by Mohamad Ali to find the
ancient emerald mines in the Eastern Desert, describes the feverish atmosphere in
Thebes in 1818: “The whole area of the ruins of Karnak was covered with demarcation-
lines that separated French, English, Irish, Italian, &c. excavations from each other. Euro­
pean ladies and other travelers ran around in the ruins and in the catacombs. All were
trying to find or buy antiquities, and nobody thought of the heat and the
fatigue” (Cailliaud 1821, 82). The Greek papyri found during this period are collection
pieces and generally Ptolemaic.

This archaeological fervor did not prevent the destruction of other antiquities. At Anti­
noopolis and Hermopolis, the limestone and marble monuments, known today only from
engravings in the Description de lʼÉgypte,2 became quarries: Blocks were reused in mod­
ern buildings or disappeared into the lime kilns. During his journey in Egypt in 1828–
1829, Champollion was horrified to see that the monuments no longer existed.

In 1835, under the influence of Egyptian thinker Rifaʼa Al-Tahtawi,3 Mohamad Ali ordered
the suspension of all excavations in Egypt and forbade the exportation of antiquities. This
order was ineffectual, however, because the demand was too great and the authorities
were indifferent to the plundering.

In 1858 the French obtained from Viceroy Saïd Pasha permission to create a Service de
conservation des antiquités de lʼEgypte, or Conservation Service. The idea came from Au­
guste Mariette, who had become famous in 1851 by finding the Serapeum of Memphis.
Mariette was sincerely concerned about the pillaging of antiquities, but for him the Con­
servation Service was also a means of staying in Egypt. Although he was assistant conser­
vator at the Louvre, he did not see himself as an armchair scholar. For France, this “ar­
chaeological protectorate,” which the French managed to maintain until 1952, was a low-
cost way of acquiring more influence in Egypt.4 As founder and first director of this
agency, Mariette put an end to the unbridled pillaging and to a large extent managed to
acquire for the Service the sole right to excavate. This practice enriched the holdings of
the Egyptian Museum, which he founded.

(p. 32) Sebâkh and Cartonnage: The Papyrological Excavations

The First Fayyum Find

Mohamad Ali had begun an ambitious program of modernization in Egypt, which was car­
ried on by his successors and much encouraged by the English, who took over administra­
tion of the country in 1882. Mohamad Ali's program had important consequences for the
history of papyrology. First of all, Egyptʼs irrigation system was completely restructured.

Page 2 of 26
The Finds of Papyri: The Archaeology of Papyrology

The traditional basins, in which water from the floods was retained in order to soak the
soil, were partly replaced by canals that made perennial irrigation possible. This brought
about a significant extension of the cultivable land and resulted in double or even triple
harvests. Cash crops were introduced, the most important of which were sugarcane and
cotton. Already at the end of the nineteenth century, archaeologists began worrying about
the deterioration of archaeological sites that came to be surrounded by cultivated land. In
addition, archaeologists and scholars are now concerned about an unforeseen conse­
quence of the building of the Aswan high dam (inaugurated in 1969): the rising water ta­
ble. But the archaeologists of the late nineteenth century also had to contend with the
much more pressing competition from the sehhâkhîn (sebâkh diggers) and the powerful
economic interests that made the peasants excavate for sebâkh.

In Egyptian Arabic, sebâkh is the powdery, saltpeter-rich earth that is characteristic of

Egyptian ruins and is used for manure. Sebâkh consists of decomposed organic matter
mixed with the fine clay that results from the destruction of mud-brick architecture. In
addition, the mud bricks were also salvaged in order to fire them (Bailey 1999, 211).
From the 1830s on, the peasants (faltâhîn) were also put to work digging for sebâkh in
the mounds (kimân, singular kôm) of archaeological sediment into which the Egyptian cli­
mate had transformed ancient settlements. These were sometimes as much as twenty me­
ters high. In 1910 an official inventory of the kimân from which sebâkh could be extracted
numbered 545; most of these were located in the delta, many in Middle Egypt, and a few
in Upper Egypt, which was less populated (Bailey 1999, 213). The industrial crops were
not the only “consumers” of sebâkh, which was also collected in order to extract the salt­
peter and produce gunpowder. In the second half of the nineteenth century, large gun­
powder factories were constructed next to Antinoopolis and Hermopolis, and the mud-
brick city walls of Antinoopolis were swallowed up.

Whole buildings and everything in them were carried off by the camels and donkeys of
the sebbâkhîn, who inevitably also found marketable antiquities. During the winter of
1877–1878 they attacked the kimân Faris, the ruins of the ancient metropolis of the Arsi­
noite nome. Papyri were found by the thousands, constituting the “first Fayyum find,”
most of which was bought by the Austrian dealer and collector Theodor Graf, who sold
them in 1884 to Archduke Rainer. The first (p. 33) Fayyum find marked the beginning of il­
licit excavations in the whole province, and the antiquities market in Cairo was swamped
by enormous quantities of Fayyum papyri. In the last decade of the nineteenth century
the site of Soknopaiou Nesos was plundered by both the fallâhîn and two local antiquari­
ans, who for some time had been granted exclusive “rights” to these excavations. No pa­
pyri from the Fayyum or the neighboring Herakleopolis and Hermopolis found in the ur­
ban ruins during this period are older than the Roman principate. The earlier levels of oc­
cupation had been covered by the Roman and Byzantine levels, and the rising humidity
had destroyed the earlier, lower levels, which were thus less likely to yield papyri.

Page 3 of 26
The Finds of Papyri: The Archaeology of Papyrology

The Excavations for Papyri

In 1882 nationalist troubles led to a British military occupation of Egypt. England took
charge of the country's administration, with the result that every Egyptian minister was
under the control of a British counselor. The Egypt Exploration Fund (EEF), a private,
learned society whose purpose was to finance excavations in Egypt, had just been creat­
ed.5 Unlike Mariette, his successor, the French Egyptologist Gaston Maspero happily
granted authorizations for excavations, and he quickly persuaded the Egyptian govern­
ment to divide the antiquities that excavators brought to light between the excavators
and the Egyptian Museum, of which he was also the director. The EEF's first project, di­
rected by Swiss Egyptologist Edouard Naville, was initiated in 1883 at Tell el-Masquta,
ancient Pithom. The choice of a place mentioned in the Scriptures, in order to prove the
historical validity of the Bible, was meant to attract donors. When Naville was temporari­
ly unavailable the following year, the EEF sent W. M. Flinders Petrie, who was to excavate
in Egypt until 1926.

A bitter rivalry grew up between the EEF's two prime excavators. Petrie had received an
unconventional education and had developed a passion for measuring and surveying; his
archaeological recordings and attention to the humblest objects were indications of the
coming of modern archaeology. On the other hand, Professor Naville, a distinguished aca­
demic, was interested only in inscribed blocks and, when on site, spent part of every day
in his tent. When Petrie accused him of not marking the findspot of every object, he de­
fended himself by saying, “You might as well make a plan of the position of raisins in a
plum-pudding” (Drower 1985, 283).

In 1889, in the Ptolemaic cemetery of Gurob, Petrie found mummies covered in carton­
nage of demotic and Greek papyri.6 From then on, Ptolemaic cemeteries were systemati­
cally plundered, but the raiders were frequently disappointed because humidity had often
a Vected the cartonnage in such a way that it turned to dust at the slightest touch. In
1902–1905 the tombs excavated by O. Rubensohn at Abusir al-Malaq produced carton­
nage from both the first century BCE and the reign of (p. 34) Augustus. It contained pa­
pyri from the Herakleopolite nome, as well as Alexandria, a great rarity. Among the latter
is a royal ordinance that perhaps carries Cleopatra's signature, which made headlines in
2000 (van Minnen 2000).

In 1893 a young Oxford classicist, Bernard P. Grenfell, came to Egypt for the first time.
He worked with Petrie in Koptos, bought some papyri (the future P.Grenfell), and under­
stood the importance ofexcavating postpharaonic sites in order to save as much as possi­
ble from the sebbâkhîn. With support from Petrie he obtained financing for excavations in
the Fayyum. The excavations of 1895–1896, which he undertook with D. G. Hogarth and
A. S. Hunt, were the first papyrological excavations made by Western scholars. The EEF
also financed Grenfell and Hunt's first campaign at Oxyrhynchus (1897). After three
weeks of work in the Roman necropolis and the much-destroyed ancient town, they decid­
ed to concentrate on the enormous, ancient rubbish mounds. During their four months of
excavation there, Grenfell and Hunt found two thousand documentary and three hundred
literary texts. This first season (see chapter 3) was such a success that the EEF immedi­
Page 4 of 26
The Finds of Papyri: The Archaeology of Papyrology

ately created the Graeco-Roman branch, which was intended to finance the papyrological
excavations and the publication of the texts they brought to light.

The so-called Oxford Dioscuri immediately acquired a following. In 1899 the great Ulrich
Wilcken excavated at Herakleopolis. Unfortunately, the papyri he found there went up in
smoke when the ship on which they had been sent to Europe burned in the harbor of
Hamburg. From 1902 to 1906 Otto Rubensohn excavated on behalf of the Berlin muse­
ums at Hermopolis (al-Ashmunayn), where he had to contend with competition from the
Italians, who excavated there until 1909. However, their harvests were modest compared
to that of Grenfell and Hunt at Oxyrhynchus. The sebbâkhîn had already made great finds
at Hermopolis before organized excavations began.

Sebbâkhîn vs. Archaeologists

The competing economic interests were too important to allow the Antiquities Service to
curb the relentless destruction of the kimân. At the end of his campaign at Antinoopolis in
1913, J. de M. Johnson noted that all attempts to protect the kimân would meet with oppo­
sition from the ministries of the interior, finances, agriculture, and public works (the An­
tiquities Service had been part of public works since 1883) (Johnson 1914, 173n1).
Against such powers, the resources of the Antiquities Service counted for nothing, but at
least it tried to limit the extraction of sebâkh. The first decree, inspired by Maspero, was
issued in 1901, and the following ones were in the same spirit: Sebâkh digging was sub­
ject to authorization and must be carried out under surveillance, and the antiquities
found must be handed over to the service.

Figure 2.1. Karanis. The northwestern corner of the

area excavated by the sebbâkhîn. Photo by G. R.
Swain. Kelsey Museum Archives, 5.1707. Courtesy of
The Kelsey Museum, University of Michigan.

But worse was in store: the great landowners began to use railways to transport the se­
bâkh. In 1910 Maspero complained, “Until today, the method of transporting the (p. 35)
sebâkh on camels and donkeys allowed the farmers the time to sift the manure and conse­
quently to collect what they found in it, so that we received our part. With the present
procedure the manure is loaded directly into the dumping wagons. Precious objects are
crushed or broken, papyri are reduced to smithereens and only large pieces resist

Page 5 of 26
The Finds of Papyri: The Archaeology of Papyrology

destruction” (Maspero 1910, 321). In fact, as early as 1884 a dump wagon track laid by a
sugar company led right into the heart of the ruins of Hermopolis. In 1925 at Karanis, the
American excavators from the University of Michigan had to come to an agreement with
the daira Agnelli, an Italian company that was authorized to extract two hundred cubic
meters of sebâkh per day (figure 2.1). Going into what used to be the center of the village,
A. E. R. Boak had the impression of being in “the crater of some extinct volcano.” After
negotiations, the Italians agreed to take only the dirt from the excavations, while the
Americans grudgingly consented to choose their excavation sites with regard to their
richness in sebâkh and proximity to the tracks (Boak and Peterson 1931, 3). However,
they soon came to appreciate the fact that their dirt was removed for free; after the sec­
ond season, when sebâkh output had been unacceptably low and the Italians had threat­
ened to switch to chemical manure, the Americans promptly concentrated again on more
productive locations (Kelsey Museum Newsletter [Fall 2005]: 5). It was not until the
1930s that sebâkh digging became illegal.

(p. 36) Saving Papyri

The excavations conducted before the First World War with the sole purpose of finding
papyri and ostraca had common traits. The concession areas allotted to the excavators
were very large, which permitted them to leave disappointing sites after only a few days
in search of more promising ones. They were poorly financed by public or, in the case of
the Graeco-Roman branch of the EEF, private funds exclusively to find papyri; thus, the
excavators could never be sure they would be able to return the following year. They
were therefore always under pressure to produce immediate results and did not waste
time making plans before beginning to dig. They observed the color of the earth or the
feel of it under the boots, which can indicate the presence of a good layer of afsh, a mix­
ture of earth with straw and other dry, vegetable matter, which experience had taught
them often contained papyri (Grenfell and Hunt somewhat pompously called these indica­
tors “the principles of afsh”). Their global view of the site was vague, and they had no
precise idea of the position of their ditches. In a description to Wilcken of the findspot of
an important cluster of fiscal ostraca, Maspero told him to draw two lines on Mariette's
plan. The house of the ostraca “was roughly at the intersection of the two lines…. Of
course, this is only an approximation, and I could be wrong even by a hundred metres.”7
Plans of buildings are rare and are not indicated in a general plan because those were
never made. Moreover, uninscribed objects were not placed in a context, and the concept
of stratigraphy was absent.

Today archaeologists shudder at the thought of Grenfell and Hunt's “methods,” but we
must take into account the conditions at the time. First, there was no clear distinction be­
tween archaeologists and philologists, and scientific archaeology8 was typically directed
by philologists, who were more interested in written documents than objects. Petrie, who
demonstrated the scientific importance of even the humblest objects, looks like a vision­
ary in this connection. Further, Grenfell, Hunt, and their followers felt the pressure of
competition with the sebbâkhîn, the illicit diggers, and the steadily growing areas of culti­
vation. These were rescue excavations, a concept that is still with us. The papyrus excava­

Page 6 of 26
The Finds of Papyri: The Archaeology of Papyrology

tion was not unlike a race in which the “teams” tried to overtake each other: The papyrol­
ogists constantly frequented the dealers to obtain new leads to where papyri might be
found. As soon as the papyrologists had left, the illicit diggers inevitably took over.

Figure 2.2. Papyrus fragments found at

Oxyrhynchus. Courtesy of the Egypt Exploration So­
ciety.

In addition to having to work faster than the local population, the papyrologists of the
EEF also had to satisfy their donors. Several excavation reports from Grenfell and Hunt
end with an expression of their hope of having enough money to return to Egypt the fol­
lowing year. This is why they unashamedly preferred the papyri that most interested the
donors to the EEF Graeco-Roman branch (among whom there were seven bishops), name­
ly, the literary and the theological texts. This consideration of their donors' preferences,
which apparently coincided with their personal inclinations, entered into their archaeo­
logical choices. Grenfell and Hunt even (p. 37) gave up excavating potentially rich (but
Byzantine) zones, as they explain in their report of the fourth season in Oxyrhynchus:
“The mounds which accumulated in the sixth or seventh century or later have been mere­
ly scratched, and to any one who cares for early and medieval Arabic documents there is
plenty of virgin ground to be explored. But the interest and importance of Greek papyri
after the fourth century wanes rapidly” (EEF Archaeological Report 14 [1904–1905]: 14).

The archaeological methods of Grenfell and Hunt may seem crude, but, thanks to the pa­
pyri, they were nevertheless able to date the layers they were excavating. They also
scrupulously collected the uninscribed material (Grenfell had been trained by Petrie), and
the large number of coins found—even if coins would not normally be thrown away—
shows that their workers were careful. Where papyri were concerned, they were aware of
the archaeological context and made an effort to keep together those that had been found
together (figure 2.2):

Since this rubbish mound had proved so fruitful I proceeded to increase the num­
ber of workmen gradually up to 110, and, as we moved northwards over other
parts of the site, the flow of papyri soon became a torrent which it was difficult to
cope with. Each lot found by a pair, man and boy, had to be kept separate; for the
Page 7 of 26
The Finds of Papyri: The Archaeology of Papyrology

knowledge that papyri are found together is frequently of the greatest importance
(p. 38) for determining their date, and since it is inevitable that so fragile a materi­

al should sometimes be broken in the process of extricating it from the closely-

packed soil, it is imperative to keep together, as far as possible, fragments of the
same document. We engaged two men to make tin boxes for storing the papyri,
but for the next ten weeks they could hardly keep pace with us. (EEF Archaeologi­
cal Report 6 [1896–1897]: 6 f)

The existence of a legal antiquities trade with the possibility of exporting9 must have made the
excavators' life even more difficult since the Egyptian workers were seriously tempted to hide
the best finds and sell them to the dealers, who often came right up to the edge of the excava­
tion. For this reason the bakshish principle, invented by Petrie, was practiced, through which
each worker was paid extra for his finds. One can imagine that the poor scholars must have
spent more time keeping track of the finds of their numerous workers than following the excava­
tion step by step. Friday payment was a nightmare for Hogarth when he was with Grenfell and
Hunt in the Fayyum (Montserrat 1996, 142). Moreover, J. de M. Johnson reports that there was a
constant need of arbitration between the teams of diggers, since the practice was to give each
one a strip several meters wide to excavate (Johnson 1914, 175).
Egypt has fallen victim to its extraordinary archaeological riches and is undoubtedly the
least well excavated of the ancient Mediterranean cultures. But let us not forget that pa­
pyrology as a discipline would not exist without the massive finds made by the fellahîn
and the somewhat uninhibited excavations carried out by a number of lucid and pragmat­
ic scholars of the late nineteenth and early twentieth centuries.

The Scientific Excavations

The Coming of the Archaeologists

Was it the diminishing abundance of papyri that made salvage excavations less urgent
and led to making a virtue of necessity? Or was it the progress in archaeological tech­
nique and in social and economic history, which made for an understanding that precious
archaeological contexts had been destroyed in the unbridled race for sebâkh and papyri?
No doubt both reasons were pertinent.

Page 8 of 26
The Finds of Papyri: The Archaeology of Papyrology

Figure 2.3. Rolls of papyrus found in the threshold

between rooms D and E of 5026 at Karanis: a docu­
mentary cache? Worms feasted on them before papy­
rologists could do so, and they are all still unpub­
lished (unless P. Mich. IX 551, a donkey sale, is one
of them, but layer indications are contradictory).
Photo by G. R. Swain. Kelsey Museum Archives,
5.1801. Courtesy of the Kelsey Museum, University
of Michigan.

From the 1920s on, the excavations of Graeco-Roman sites no longer had the finding of
papyri as their sole objective. This new trend was initiated in the United States by the
philologist Francis W. Kelsey, who was professor of Latin at the University of Michigan.
Kelsey saw the faults of Graeco-Roman archaeology in Egypt and decided that the United
States had a duty to do something about it. The Americans' choice fell on Karanis, which,
thanks to Egypt 's unique climatic conditions, turned out to be an ideal site for the study
of people in antiquity not only through the writings they had left but also by the analysis
of the material world in which they had lived. (p. 39)

The excavation of Karanis, carried out by the University of Michigan from 1924 to 1935,
was exemplary for its time.10 While we wait for the recent excavations of Tebtynis and
Bacchias to progress, Karanis is still the best-known urban conglomerate in the Fayyum.
For the first time, the excavators worked to distinguish different levels of occupation,
carefully mapped the excavated zones, made plans and cross-sections of houses, and
scrupulously recorded the location and level of each object or papyrus (figure 2.3). This
was a vast improvement over previous excavations and one that would remain unparal­
leled for a long time.

Now that modern archaeology demands a very high degree of technical skills, it is no
longer possible to become an excavator just by excavating. And yet, although the papyrol­
ogists hardly dare touch a trowel nowadays, many Graeco-Roman sites are still opened on
their initiative. In fact, Graeco-Roman sites in Egypt have not been very attractive to ar­
chaeologists unless they have a bearing on a larger historical problem like central power,
commerce,11 or the environment.12 In other cases, Graeco-Roman layers have to be re­
moved in order to gain access to pharaonic sites, as is the case with D. Bailey's excava­
tion at Hermopolis, which is essentially a by-product of an Egyptological excavation. The

Page 9 of 26
The Finds of Papyri: The Archaeology of Papyrology

exploration of villages (p. 40) or middle-sized towns that the metropoleis were (which
were allowed to call themselves cities only at a late date and even then relatively briefly)
does not lead directly into the mainstream of historical research. There is no prestige at­
tached to these sites. Their monuments have been dismantled, Egypt is notoriously poor
in Greek or Latin inscriptions, and what is left for the archaeologist are modest mud-brick
structures, rubbish dumps with difficult and unrewarding stratigraphies, and overwhelm­
ing quantities of commonplace material, not the least of which is pottery.

The excavations of Tebtynis (begun in 1988 in collaboration with the Institut français
dʼarchéologie orientale and under the direction of Claudio Gallazzi, professor of papyrolo­
gy at the Università Statale of Milan) have shown that papyrology has nothing to lose by a
methodical excavation, provided that adequate financing is available; thus, several large-
scale seasons can be conducted without fear of disruption. There are no more large con­
centrations of papyri, and excavations must be conducted over a large area in order to
gather a good crop. Although Tebtynis has been excavated since the end of the nine­
teenth century by a succession of official and illicit diggers (not to mention the sebbâkhîn),
the site is rich in ostraca, a commonplace writing material in Upper Egypt but which was
thought to be rare in the Fayyum. This discovery is the result of a more careful and me­
thodical excavation. The earlier papyrus hunters made soundings in order to find a layer
of afsh, which they then followed as one would a vein of ore. Since they were primarily in­
terested in papyri, they paid no attention to the potsherds, which are always abundant in
Egyptian excavations (Gallazzi 2000, 31). Incidentally, it is interesting that, in the deonto­
logically correct excavations of Bacchias, ostraca are handled more casually than papyri.
Papyri are individually located in three dimensions, while ostraca are treated as pot­
sherds and recognized as written sources only when the potsherds are washed and sort­
ed. Ostraca are thus identified with only a layer number (Davoli 2000, 17–18). A final dif­
ference between today's excavations and those of the early papyrus hunters is that the
latter were not interested in the lower layers, where the pressure and the mounting hu­
midity made the presence of good papyri less likely. In this way they cut themselves of
from earlier material and depended, at least in the Fayyum, only on mummy cartonnage
for Ptolemaic papyri.

The exploration of Mons Claudianus, begun a year before that of Tebtynis, on the initia­
tive of a group of papyrologists, of whom I was one, is a special case.13 There the papyrol­
ogists found themselves in a situation similar to that of the papyrus hunters before World
War I. In a remote location between the Nile and the Red Sea, Mons Claudianus had re­
mained almost undamaged for nearly two thousand years, but with the tourism boom on
the Red Sea coast beginning in the 1980s, it was exposed to illicit digging, which in turn
gave the Bedouin ideas. Moreover, because of its isolation, the site was impossible to
guard.14 Mons Claudianus was very rich in texts (not papyri but mostly ostraca), which
were (p. 41) concentrated in the rubbish mounds and easy to find, so there was a great
temptation to dig just to find the ostraca. However, the team was strengthened by the
participation of several archaeologists, who added some archaeological respectability.

Page 10 of 26
The Finds of Papyri: The Archaeology of Papyrology

From the point of view of human experience and method, the seven years of excavation at
Mons Claudianus were not free from friction, but they were all the more interesting in
that the archaeologists had high principles, while we, the papyrologists, were simply ex­
cited by all these ostraca that were there for the taking. Our colleagues could but regard
us as looters, while we tended to see them as killjoys. We undoubtedly learned a lot from
each other. In any case, the experience of Mons Claudianus has shown that the interests
of papyrology and modern archaeology are not necessarily easy to reconcile. Understand­
ably, the archaeologists are loath to dig in places that suggest no other prospect than the
presence of texts. Modern archaeologists are very conscious of the destruction caused by
excavation. Like surgeons, the archaeologists endeavor to use nonin-trusive methods in
order to leave unspoiled samples for exploration by future generations of archaeologists
with newer and even better techniques. Their efforts are concentrated on mapping, plan­
ning, and measuring. Now magnetometry permits analysis of remains under the soil be­
fore (or without) digging. Trial trenches are reduced to a minimum, and soil samples are
taken for analysis and study. It is thus possible to obtain a comprehensive visualization of
a site without destroying it. In Egypt, however, the problem is that the “future genera­
tions” are already at work.

Archaeological Multidisciplinarity as Seen by a Papyrologist

No archaeological course of study fails to point out that it involves multiple disciplines.
However, the joint study of anepigraphic artifacts, soil samples, and texts that have been
collected with the strictest observance of archaeological stratigraphy does not automati­
cally lead to fruitful conclusions. Texts and objects do not always illuminate each other.
Often there is little common ground, and various types of material present different prob­
lems. Often one gets the impression of parallel, unconnected worlds. Sometimes the data
are redundant; sometimes they are supplementary. The collaboration between specialists
is necessary but can be disappointing.

The specialists in uninscribed material who work in the Eastern Desert never fail to ask
us how their type of material is reflected in the ostraca. The confrontation of the data
gives varying results depending on the types of material. It is without a doubt the ar­
chaeobotanists who profit most from the texts.15 Food and provisions are among the most
common subjects in the ostraca, whether they are private letters or administrative. Most
of the cultivated species that the archaeobotanists have identified are also mentioned in
the ostraca, which, on the other hand, give details about the organization of the provi­
sioning. For instance, we learn (p. 42) that certain herbs and vegetables were cultivated
in desert gardens. At Mons Claudianus a number of quarriers' and blacksmiths' pay chits
provide much detail on the workers' diet.

Meat and butchering are also mentioned in the ostraca, although to a lesser degree, but
as meat was rare, the information is more anecdotal (e.g., “buy three suckling pigs,” “I
send you a donkey leg”). The study of the faunal remains gives precious quantitative in­
formation: We learn that donkey was the meat most commonly eaten at Mons Claudianus,
while pork was more common in the forts along the roads to Myos Hormos and Berenike.
This difference is explained by the many work donkeys in the quarries. Archaeozoologists
Page 11 of 26
The Finds of Papyri: The Archaeology of Papyrology

have also been able to state that significant quantities of fish were eaten at Mons Claudi­
anus and that they came almost exclusively from the Red Sea, not from the Nile. The only
Nile fish that has been identified is the catfish, whose presence is certainly not explained
by their fine taste but rather by the fact that these amphibious creatures could arrive
alive at Mons Claudianus. The ostraca give the impression that fresh fish were a rare deli­
cacy and do not inform us of the quantities involved. On the other hand, they put a per­
spective on the means of obtaining fish: At Mons Claudianus, according to one ostracon,
fish from the sea were brought by Bedouin, who are probably to be identified with the
coastal dwellers called “Arab-Egyptian fish eaters” by the geographer Ptolemy. On the
Myos Hormos road the relay post riders of the military stations helped to provide fine fish
for the table of the prefect of Egypt when he was in Koptos.

With regard to leather and textiles, a comparison between the data of the texts and ar­
chaeology is of no particular interest. The leather specialists have been able to identify a
variety of types of shoes, but these are described in the texts by only two words: san­
dalion and the generic hypodêma. The variety in the leather objects is nothing compared
to that of the textiles, where the richness of color and the complexity of weaving dazzle
the experts. Analysis of colors is beginning to show just how advanced dyers from Roman
Egypt were, especially in imitating real purple with vegetable dyes. But, of course, there
is nothing on the subject in the texts written by those who wore these clothes.

By contrast, it is not useless to search the ostraca for names of the ceramic containers
that the ceramologists classify and draw. At Mons Claudianus it goes without saying that
the ubiquitous keramion of the texts is the common Nile silt amphora, which constitutes
90 percent of the pottery found (the so-called AE3 bitronconique). Since complete speci­
mens have been found, it has been possible to measure the contents at 6.5 liters, which is
extremely important for a calculation of wine consumption because the keramion is used
to measure rations. From the middle of the second century, the ostraca from the sites in
the Eastern Desert—at least those that continued to function after about 150 CE—men­
tion a new kind of ordinary container, the kolophôenion. It so happens that, at the same
time and from the same sites, large quantities of costrels appear. It is almost inevitable
that ceramologists identify these costrels with the kolophôenia.

Since the ostraca are often precisely dated, they also permit dating of certain con­
(p. 43)

temporary artifacts. The ostraca from Mons Claudianus have, for instance, allowed D.
Bailey to date the so-called frog lamps to the second century, when they had hitherto
been thought to be from a later period. On the other hand, at the site of Maximianon,
where, as luck would have it, none of the fifteen hundred ostraca found could be dated, it
was the typology of the glass that allowed a dating of the layers and hence the ostraca.

It is the collaboration with the actual excavator that is most productive for the papyrolo­
gist. In spite of the significant information derived from the written documents, we must
not ignore the results of the excavator's austere examination of structures and layers.
Certainly the texts normally tell us the ancient name of the site—except in cases where
the papyri were recycled as mummy cartonnage and found in graveyards (the embalmers

Page 12 of 26
The Finds of Papyri: The Archaeology of Papyrology

reused old papyri that could have come from anywhere). On the other hand, the texts can
be deceptive with regard to the periods of occupation either because whole layers have
disappeared or because the site has not yet been fully explored. This is the case in Tebty­
nis, where the latest papyri date from the third century CE but where the Byzantine part
of the town has not yet been excavated (Bagnall 2001, 234).

In certain cases, the archaeological context in which a document was found can prove es­
sential for its understanding. This is the case with an important group of ostraca found at
Tebtynis in 2003. Their texts are short: a date, a name, and a quantity of beer measured
in dichôra. A few similar texts were already known from stray finds or older excavations,
and the editors had interpreted them either as receipts given to a brewery by those who
had received beer or as delivery notes issued by the brewery. Now, the new texts from
Tebtynis in some cases carry the additional mention of posis zytou (consumption of beer),
which places them in the context of the ritual consumption of an association. In addition,
the ostraca have been found in a banquet hall, which suggests that the names are those
of association members who have presented the beer (Reiter 2005, 133–136). In isolation
(i.e., in a museum or a collection), these documents made no sense and, given their brevi­
ty, would be of little interest except perhaps for some new proper names.

To give another example, let us go back to the letter by Maspero, in which he tries to ex­
plain to Wilcken, using a plan of Karnak, where he had found a group of tax receipts that
had accumulated against the wall of a house. At the time of the discovery, Maspero did
not know that these texts could have clarified the important problem of where tax re­
ceipts were kept. Often these are found, as we should expect, in the house of the taxpayer
who received them when the taxes were paid. However, these receipts from the Theban
region (most of them bought in the 1880s and 1890s) can often be organized into
archives not of the taxpayer but of the tax collector as if they had been kept by the latter.
To explain this apparent paradox, Wilcken suggests that the receipts were issued by the
bank to the tax collectors. In fact, the taxpayer often did not pay the taxes directly to the
public bank but instead to collectors, who (p. 44) then deposited the money. Unfortunate­
ly, these ostraca were not read on the spot and now appear to have disappeared without a
trace in the stores of the Egyptian Museum, where Maspero had duly deposited them, so
we do not know which formula they followed, and Wilcken's theory cannot be verified.

In What Archaeological Contexts are Papyri

Found?
The types of papyri that would most interest the historians, those of the offices of the cen­
tral administration at Alexandria, or the classical philologist, the contents of the library of
Alexandria, have disappeared. This is why papyrology still suffers from a comparative
lack of recognition as an important discipline. The historian will have to make do with pa­
pyri found in the nome capitals and the villages south of the delta. As it happens, the
delta, which was the richest and most populous region, has a geography and a climate
that have not allowed papyri to survive as they have elsewhere in the country. There are
Page 13 of 26
The Finds of Papyri: The Archaeology of Papyrology

only two exceptions to this, namely the carbonized papyri of Thmouis and Boubastos (see
chapter 16). Moreover, archaeological contexts are neither all equally interesting nor
equally favorable to the preservation of papyri.

Books in Tombs

Figure 2.4. Cartonnage elements at Tebtynis (1899–

1900). Courtesy of the Egypt Exploration Society.

Some beautiful literary rolls have been found in tombs. The oldest is the Timotheos pa­
pyrus (fourth century BCE), which was discovered by L. Borchardt in 1902 in a wooden
sarcophagus in Abusir. A roll containing Iliad II (second century CE) was found by Petrie
at Hawara under the head of the mummy of a young woman. The first editor of the pa­
pyrus, A. H. Sayce, credits her with an agreeable, intellectual physiognomy, undoubtedly
Greek. Earlier, in 1858, Mariette had given to the Louvre a papyrus of Alcman found by
the natives at Saqqara, rolled up in linen and placed between the legs of a mummy.16 The
manuscript of Herodas's Mimes (first–second century CE) appears to have been found
north of Assiut (at Meir?), perhaps along with the Constitution of Athens, in the tomb of a
couple. The wife, Sarapous (daughter of Sarapion), had died at the age of fourteen (Mar­
tin 2002, 23–26). This practice was not widespread; in February 1912 J. de M. Johnson,
working for the EEF, spent several days opening approximately a hundred tombs at Qa­
madir, near Oxyrhynchus, “in the vain hope of a papyrus roll.” The people who were suffi­
ciently smitten with literature to be buried with an expensive book must have been statis­
tically rare, for it seems necessary to interpret these pious gifts as a reflection of the
deceased's personality rather than as an imitation of the (p. 45) Egyptian habit of giving
the dead a book of religious-magical formulas as a passport to the hereafter. The Derveni
papyrus (fourth–third century BCE), incidentally, attests to a similar concern in the Greek
world. This papyrus, the only one to have been found in Greece, contains a philosophical-
eschatological text and was found in 1962, carbonized among the remains of a funeral
pyre (Betegh 2002).

Page 14 of 26
The Finds of Papyri: The Archaeology of Papyrology

Cartonnage and Crocodiles

During the whole of the Ptolemaic period and up to the end of the reign of Augustus, hu­
man mummies were wrapped in linen bands and often given a mask, sometimes also oth­
er separate elements such as a foot case, a pectoral, an apron, and leg guards (figure 2.4).
These elements were made of a core of papyrus, linen, or palm fiber, which was covered
with stucco and painted with standardized, protective images. The use of scrap paper for
cartonnage does not seem to have been common before the reign of Ptolemy II
Philadelphos (283–246). Most of the papyri of the third century bce come from such car­
tonnage, with the notable exception of the Zenon archive. Most papyri from cartonnage
are administrative documents, but literary (p. 46) texts are sometimes found as well.
These papyri have the disadvantage of having been cut to fit the part of the body that
they were intended to cover. Moreover, the writing is often weakened, first by the applica­
tion of the stucco and then by its removal.

Figure 2.5. Crocodile mummies at Tebtynis (1899–

1900). Courtesy of the Egypt Exploration Society.

At Tebtynis, Grenfell and Hunt accidentally discovered that a number of crocodile mum­
mies had been prepared with recycled administrative papers, either as wrapping or as
filling but not as cartonnage proper (figure 2.5):

On Jan. 16, 1900—a day which was otherwise memorable for producing twenty-
three early Ptolemaic mummies with papyrus cartonnage—one of our workmen,
disgusted at finding a row of crocodiles where he expected sarcophagi, broke one
of them in pieces and disclosed the surprising fact that the creature was wrapped
in sheets of papyrus. As may be imagined, after this find we dug out all the croco­
dile-tombs in the cemetery; and in the next few weeks several thousands of these
animals were unearthed, of which a small proportion (about 2 per cent.) contained
papyri. (Grenfell, Hunt, and Smyly 1902, vi)

These papyri were of the second century BCE and came from the office of the village scribe at
Kerkeosiris. One might have thought that this was an isolated case, but in 1901 at Talit, a village
neighboring Tebtynis, Grenfell and Hunt found (p. 47) other crocodiles wrapped in Greek and

Page 15 of 26
The Finds of Papyri: The Archaeology of Papyrology

demotic documents of the first century BCE, just slightly later.17 This practice has been related
to the subventions from the Ptolemies toward the burial of sacred animals,18 even though scrap
paper from the administration is perhaps not what one imagines when reading about the “mag­
nificent and famous gifts” that the Rosetta Stone mentions in this respect.

Buildings

Obviously, the dream of every papyrus hunter was to find the public archive still in place
in the bibliothêkê. This miracle almost happened in 1892, when Naville came upon what
must have been the archives of the Mendesian nome in the delta. The rooms of the build­
ing were filled with papyri burned in a ire in the late second century CE. Naville's de­
scription is depressing, although he may purposely have made it even more so:

They are now quite carbonized, like those of Herculaneum, but even in a worse
state. They are most difficult to take out, they crumble to pieces when they are
loosened from the earth which covers them, but, by looking sideways the charac­
ters are still discernible; they generally are Greek, in good handwriting. As for
those which have escaped the fire, they are quite hopeless. The moisture and the
salt in the soil have reduced them to a kind of brownish paste, which seems to be
very fertile, for roots of plants grow in it in abundance. (EEF Archaeological Re­
port [1892–1893: 4])

Naville filled five boxes, which arrived at the British Museum with their contents reduced to
crumbs. Since Petrie suspected that Naville, heavy handed as he was, had not done everything
possible to save what could be saved, the EEF at once undertook a rescue operation directed by
Howard Carter. Carter spent two months looking in vain among the ruins without finding the
bibliothêkê. Naville's indications were not precise enough to find it (Drower 1985, 284). A num­
ber of rolls from illicit digging came into the hands of the Egyptologist Albert Daninos, who con­
ceived the brilliant idea of softening them in rectified alcohol. He then cut the rolls open length­
wise and detached the sheets, which he glued onto cardboard. This is all that is left of the
archive of the Mendesian nome.
The papyri from Dura-Europos in Syria are the only other example of an archive uncov­
ered by an official excavation. And yet, even they do not represent the whole of the
archive of the Cohors XX Palmyrenorum, whose camp had been installed in part of the
town. The room in which the texts were found is just a place where one stored documents
that were no longer of interest. It opened onto another room where the walls were cov­
ered with dipinti, graffiti, and “a great many smudges of ink as if one had used the plas­
ter for wiping pens and fingers” (Rostovtzeff 1934, 152), probably the officium of the
scribes of the general staff.

Grenfell and Hunt established a hierarchy of archaeological contexts in which one

(p. 48)

might find papyri. The best are abandoned houses that have collapsed, thus sealing both
papyri and objects of daily use where the inhabitants had left them.19 But the interest, ac­
cording to them, was that the papyri were better preserved than in the rubbish dumps,
not, as we would now think, that they were part of a coherent archaeological context in
which the various elements could elucidate each other. Strangely, the first editors of the

Page 16 of 26
The Finds of Papyri: The Archaeology of Papyrology

papyri from Karanis, which had been found in advanced excavations with multidiscipli­
nary aspirations, showed the same inhibition and did not take the archaeological context
into consideration in order to explain their texts. As an example, the ostraca found in the
same house were published separately, while the editor did not realize that they could es­
tablish the genealogy of the family that had lived there (van Minnen 1992). The archaeol­
ogists in turn let precious stratigraphic information slip away because they did not know
that a group of papyri found in a trench were homogeneous and consequently belonged to
the same stratigraphic unit (van Minnen 1994). At least the registration of the finds was
so well conceived (even if somewhat rough according to today's standards) that it is still
operational, and now, sixty years later, allows one to make use of the data and show what
can be deduced from what Peter van Minnen has called “a house-to-house approach,” a
method that he has applied to house 17 in state B. Taking into account all the papyri, pub­
lished or unpublished, and all the objects from this house, he has demonstrated that
house B17 was inhabited by a tax collector (praktôr argyrikôn) by the name of Socrates,
who not only lived but also worked there. Thanks to a draft of a petition in Socrates'
handwriting, van Minnen has been able to identify him as the writer of a Karanis tax roll
in which he leaves a personal mark by amusing himself by inventing Greek equivalents of
the taxpayers' Egyptian names. Some of these names testify to a high degree of erudition
(e.g., ἀνδίκτης, which is otherwise attested only in Callimachus, where it means “mouse­
trap”). Socrates uses this name here instead of the Egyptian name, Panpin, which means
“he of the mice.” Callimachus was not among the books found in Socrates' library in
house B17, but a fragment with text by this author has in fact been found in the house op­
posite it. Van Minnen further remarks that Socrates did not live with the woman who was
probably the mother of his twin sons, a Roman citizen who lived several blocks away and
who had declared her sons as of an unknown father in order for them to inherit her juridi­
cal status.

The texts found in Socrates' house were only what was left by chance after abandonment.
Earlier I discussed public archives; on their own level, private archives, which people
guarded carefully in jars or boxes, also have a much greater importance than documents
found in isolation (see chapter 10). Unfortunately, these nests of papyri, the private
archives, are rarely found by archaeologists. We may think of the archives of Zenon,
which were found in Philadelphia under unknown circumstances somewhat before World
War I, or the archive of (p. 49) Heroninos, which was reportedly found at Theadelphia by
fellahîn in a wooden box a short time after Grenfell and Hunt's excavations. One of the
curses of papyrology is, in fact, that so many of the important discoveries have been clan­
destine; thus, the archaeological context is unknown. Predictably, the natives were often
luckier than the professionals. Numbers were in their favor, and they were impeded nei­
ther by time and financing nor by methodological scruples. Of course, scholars have also
made some discoveries of archives, but these were in older excavations of a period when
the excavators were often philologists who saw no farther than the contents of their texts
and did not think of looking for help in the material context.

Page 17 of 26
The Finds of Papyri: The Archaeology of Papyrology

At Kôm Ishqaw in 1905, the classicist Gustave Lefebvre, who was also inspector of antiq­
uities, found the archive of Dioskoros of Aphrodite. His report gives a good impression of
a less pedantic way of doing archaeology: “For a few pounds the owner sold us the right
to turn his plot of land inside out. We excavated right into the road, on the other side of
the wall. Everything was done in three days.” One meter below the surface they came up­
on a house with three rooms, in one of which “there stood a jar with the neck broken off,
0,90 m high, full of papyri…. The inventory was quickly made: at the top of the jar there
came to light, all crumpled up, a codex of eleven leaves: it was the Menander manu­
script…. In the jar there were also some hundred and fifty rolls, mostly Greek, business
papers, wills, contracts, letters &c.” (Lefebvre 1907, x). Unfortunately, there is not even a
photograph of the discovery.20 The excavation reports from the heroic age of papyrus
hunting make much more agreeable reading than today's terse archaeological reports,
but they leave us unsatisfied if we are looking for useful information to elucidate our
texts. When one considers all this wasted archaeological potential, one may perceive
some irony in the technical refinement that is deployed today on the tatters that remain
after the two hundred years of devastation to which Egyptian antiquities have been sub­
jected.

The Document Caches

The documents found in 1960 and 1961 by Yigael Yadin in the Judean desert are to this
day the only archive found by someone worthy of being called an archaeologist and in the
condition in which the owners left them. This brings us to the concept of “document
cache,” which Alain Martin has proposed adding to the roster of archaeological contexts
that produce papyri (Martin 1994). The most famous examples of these treasure troves
not of gold but of documents are undoubtedly the two archives found in the “cave of let­
ters,” one of the inaccessible strongholds in which survivors of the Bar Kochba revolt took
refuge in 132 CE or shortly thereafter.21 The first cache found was a bundle of fourteen
letters on papyrus and one on a folded, wooden leaf tablet (see chapter 1), all of which
was tied up with string and sealed with a clay seal. The documents are written in Greek,
(p. 50) Aramaic, and Hebrew and contain orders, especially instructions for confiscation,

issued by Bar Kochba himself, addressed to several of the persons who had taken refuge
in the cave and who had found it prudent to keep these documents in order to cover
themselves later on, if need be (Yadin 1961).

The other archive is that of Babatha, twice widowed and a strong-willed woman. She had
not undertaken the climb up the cliff in Nahal Hever without the thirty-five documents
that proved, first of all, her ownership of various disputed properties. The archaeological
report shows with what care she had classified and protected these precious documents
(Yadin 1962), which were found in a leather pouch, itself wrapped in linen and tied with a
string (figures 2.6 and 2.7). Finally, the bundle had been put into a water skin, like the Bar
Kochba letters.

Page 18 of 26
The Finds of Papyri: The Archaeology of Papyrology

The discovery of other caches of personal objects (keys, a bird net, everyday utensils) in
the cave seems to indicate that its inhabitants left with the intention of coming back for
their belongings.

The Rubbish Deposits

Rubbish deposits represent the type of ground most likely to produce massive finds of pa­
pyri, even though the proportion of inscribed material will be less in a village mound than
in that of a metropolis. The depositories can be public dumps (some of which are as much
as nine meters high) or private accumulations of garbage in an unused space, whether a
room in a house, an empty cistern, or a silo. The depositories contain domestic, industrial,
and agricultural garbage, rubble, ashes, rags, discarded objects, and masses of pot­
sherds. It is surprising to see that the various kinds of debris have been thrown out ac­
cording to type. The papyri (and in the Eastern Desert the ostraca) are no exception. They
are rarely isolated in the dump but were thrown away several at a time. The papyri that a
person decided to discard were not important documents that ought to be kept, like con­
tracts, title deeds, or literary works. Nevertheless, the excavations at Oxyrhynchus show
that from time to time, in circumstances that we should like to know, it was decided to
clean out documents and rolls in quantity. Some of them were even still in the baskets in
which they had been brought to the dump. Grenfell and Hunt also remark that rolls were
systematically torn before being thrown away.

Figure 2.6. Babatha's archive before opening. After

Y. Yadin, J. C. Greenfield, A. Yardeni, and B. Levine,
The Documents from the Bar Kokhba Period in the
Cave of Letters. Hebrew, Aramaic, and Nabatean
Aramaic Papyri, Jerusalem 2002, II, pl. 1. Courtesy of
the Israel Exploration Society.

Page 19 of 26
The Finds of Papyri: The Archaeology of Papyrology

The abundance of texts from Oxyrhynchus (or to a lesser degree, from the mounds of the
Eastern Desert) must not blind us to the fact that they are only an infinitesimal part of
what was thrown away in antiquity. It took only a shower and then something thrown on­
to the humid surface to seal it, and all of the sealed-up material would rot—turn from afsh
into sebâkh. Even in the climatic conditions of Egypt, the perfect excavation of an undam­
aged ancient site would yield only a fraction of the texts it once contained. It often hap­
pens that one can join parts of (p. 51) excavated ostraca that are in very different states of
preservation. Even if they have spent the intervening years a few centimeters apart, they
have been subjected to very diferent conditions of preservation.

Figure 2.7. Babatha's pouch, opened. After Yadin et

al. 2002 (figure 2.6), pl. 2. Courtesy of the Israel Ex­
ploration Society.

Figure 2.8. A layer with ostraca in the garbage

dump at Krokodilô (Eastern Desert). Photo by J.-P.
Brun.

The garbage dumps are the least interesting type of context from the point of view of
what the archaeological context can contribute to an understanding of the (p. 52) text.

Page 20 of 26
The Finds of Papyri: The Archaeology of Papyrology

This is why the rough-and-ready archaeology of Grenfell and Hunt as applied to the
mounds at Oxyrhynchus has not really done much harm in this respect and the less so
since they took care not to separate papyri that were found together. These rubbish
mounds are mass graves, where the links between objects that used to share a context
have been dissolved. When three inkwells are found in house B17 at Karanis, it is signifi­
cant because we know that the house was occupied by a father and two sons. But the
same three inkwells found in a dump are just material for a (p. 53) typology of inkwells.
Only the contents of the texts and/or the knowledge that they were found together allow
us to reconstitute archives.

Figure 2.9. Oxyrhynchus, 1902–1903. Men at work,

with B. P. Grenfell observing.

Figure 2.10. Didymoi 1999 (Eastern Desert). Men at

work on the garbage dump. Photo J.-P. Brun.

A dump that has formed in the open is infinitely more complex in stratigraphy than a
room filled with rubbish, where the layers are contained by the walls and are formed
more or less horizontally on top of each other. It is formed by successive dumpings, juxta­
posed or superimposed, which will have stayed together or spilled down the sides of the
heap. The lower layers are more or less horizontal, but layers become more and more un­
predictable and oblique as the heap grows. Even hardened archaeologists can get lost in
them.22 Today garbage dumps are excavated like any other archaeological ground, in oth­

Page 21 of 26
The Finds of Papyri: The Archaeology of Papyrology

er words, by removing the layers separately, which demands a high degree of concentra­
tion when they cross over or under each other (figure 2.8). In the excavations of the great
age before 1914 this was done differently: One attacked the mound laterally, caused
slices of it to fall one by one, and sorted out the papyri after each “landslide.” Grenfell
was sometimes alone “supervising” two hundred workers (figure 2.9), while today archae­
ologists feel stressed if they have to follow a dozen workers on a mound (figure 2.10). The
ungratifying work of analyzing a refuse heap is undoubtedly an important factor in influ­
encing scrupulous archaeologists not to search for texts. It is a great deal of work, and af­
terward one has little to write about it; moreover, the descriptions are indigestible for the
reader and, for the author, boring and academically unrewarding. (p. 54)

(p. 55) Conclusion

From the first Fayyum find until the First World War, the period of miraculous discoveries
lasted barely forty years. Even in 1913 Hunt felt that this period was coming to an end:
“Failing the emergence of a fresh fertile source, costly excavations may before long be
found to be not sufficiently remunerative, and there will be a gradual return to the condi­
tions of the first period of discovery.23 Isolated finds will of course continue to be made
for an indefinite time” (Hunt 1914, 91). Perhaps Hunt had not foreseen that the progress
in archaeology and the evolution of ancient history under the influence of social anthro­
pology would soon make the humblest excavated object important. And yet he was right,
for in the present era of slavery to the mass media, the benefactors and institutions who
provide financial support do not have an understanding of archaeology much different
from that of the nineteenth century. Treasure hunting and necrophilia are still the order
of the day. It is better to find a royal mummy than a papyrus and better to find a papyrus
than domestic garbage, however great its informational value. Except for the special case
of the Eastern Desert, it is no longer possible to be sure where to look in order to find
texts, but it is still quite possible that extended excavation of what is left of urban con­
glomerates will produce impressive finds. One might think of the enormous deposits of
papyri that were found at Tebtynis in 1931 and 1934 during the Italian excavations (Begg
1998; Gallazzi 1990). But it is a matter of urgency, for under the pressure of an exploding
demography, the archaeological sites most likely to yield papyri, already threatened by
the rising water table, may soon be sacrificed to the extension of urbanization and agri­
culture (Gallazzi 1994).

Bibliography
Bagnall, R. S. 2001. “Archaeological Work on Hellenistic and Roman Egypt, 1995–2000.”
AJA 105: 227–243.

Bailey, D. M. 1999. “Sebakh, Sherds, and Survey.” JEA 85: 211–218.

Begg, D. J. I. 1998. “It Was Wonderful, Our Return in the Darkness with … the Baskets of
Papyri! Papyrus Finds at Tebtunis from the Bagnani Archives, 1931–1936.” BASP 35: 185–
210.
Page 22 of 26
The Finds of Papyri: The Archaeology of Papyrology

Betegh, G. 2002. “Papyrus on the Pyre: The Derveni Papyrus and Its Archaeological Con­
text.” Acta Antiqua 42: 51–56.

Boak, A. E. R., and E. E. Peterson. (1931). Karanis, Topographical and Architectural Re­
port of Excavations during the Seasons 1924–28. Ann Arbor: University of Michigan
Press.

Cailliaud, F. (1821). Voyage à lʼoasis de Thàbes. Paris: Imprimerie royale.

Cuvigny, H., ed. (2003). La route de Myos Hormos: Lʼarmee romaine dans le désert Orien­
tal dʼEgypte. Cairo: Institut francais dʼarchéologie orientale.

Davoli, P. (1998). Lʼarcheologia urbana nel Fayyum di età ellenistica e romana. Naples:
Procaccini.

———. (2000). Archeologia e papiri. Naples: Eurocomp.

Johnson, J. de M. 1914. “Antinoë and its Papyri.” JEA 1: 168–181.

Jomard, E.-F. and J. B. J. Fourier, ed. 1809–1828. Description de lʼÉgypte, ou, Recueil des
observations et des recherches qui ont été faites en Égypte pendant lʼexpédition de
lʼarmée francaise, Paris: Imprimerie Impériale.

Drower, M. S. 1982. “Gaston Maspero and the Birth of the Egypt Exploration Fund (1881–
3).” JEA 68: 299–313.

———. (1985). Flinders Petrie: A Life in Archaeology. London: Gollancz.

Gallazzi, C. 1990. “La ‘Cantina dei Papiri’ di Tebtynis e ciò che essa conteneva.” ZPE 80:
283–288.

———. 1994. “Trouvera-t-on encore des papyrus en 2042?” In Proceedings of the 20th In­
ternational Congress of Papyrologists, ed. A. Bülow-Jacobsen, 131–135. Copenhagen: Mu­
seum Tusculanum Press.

———. 2000. “La reprise des fouilles.” In Tebtynis I. La reprise des fouilles et le quartier
de la chapelle dʼIsis-Thermouthis, ed. C. Gallazzi and G. Hadji-Minaglou, 3–39. Cairo: In­
stitut francais dʼarchéologie orientale.

Grenfell, B. P., A. S. Hunt, and D. G. Hogarth. (1900). Fayûm Towns and Their Papyri. Lon­
don: Egypt Exploration Fund.

Grenfell, B. P., A. S. Hunt, and J. G. Smyly. (1902). The Tebtunis Papyri I. London: Oxford
University Press.

Hunt, A. S. 1914. “Papyri and Papyrology.” JEA 1: 81–92.

James, T. G. H., ed. (1982). Excavating in Egypt: The Egypt Exploration Society 1882–
1982. Chicago: University of Chicago Press.

Page 23 of 26
The Finds of Papyri: The Archaeology of Papyrology

Lefebvre, G. (1907). Fragments dʼun manuscrit de Mênandre découverts et

(p. 58)

publiés. Cairo: Institut français dʼarchéologie orientale.

Martin, A. 1994. “Archives privêes et cachettes documentaires.” In Proceedings of the

20th International Congress of Papyrologists, ed. A. Bülow-Jacobsen, 569–577. Copen­
hagen: Museum Tusculanum Press.

———. 2002. “Heurs et malheurs dʼun manuscrit: Deux notes à propos du papyrus
DʼHérondas.” ZPE 139: 22–26.

Maspero, G. (1910). Rapports sur la marche du Service des Antiquités. Cairo: Imprimerie
nationale.

Montserrat, D. 1996. “ ‘No Papyrus and No Portraits’: Hogarth, Grenfell, and the First
Season in the Fayyum, 1895–6.” BASP 33: 133–176.

Pack, R. A. (1965). The Greek and Latin Literary Texts from Greco-Roman Egypt. Second
Revised and Enlarged Edition. Ann Arbor: University of Michigan Press.

Petrie, W. M. F. (1891). Illahun, Kahun, and Gurob (1889–90). London: David Nutt.

Reiter, F. 2005. “Symposia in Tebtynis: Zu den griechischen Ostraka aus den neuen
Grabungen.” In Tebtynis und Soknopaiu Nesos: Leben im römerzeitlichen Fajum. Akten
des Internationalen Symposions vom 11. bis 13. Dezember 2003 in Sommerhausen bei
Würzburg, ed. S. Lippert and M. Schentuleit, 131–140. Wiesbaden: Harrassowitz.

Rostovtzeff, M. I. 1934. The Excavations at Dura-Europos: Preliminary Report of the Fifth

Season of Work. New Haven, Conn.: Yale University Press.

van Minnen, P. 1992. “A Closer Look at O. Mich. I 126.” BASP 29: 169–171.

———. 1994. “House-to-House Enquiries: An Interdisciplinary Approach to Roman Kara­

nis.” ZPE 100: 227–251.

———. 2000. “An Official Act of Cleopatra (with a Subscription in Her Own Hand).” An­
cient Society 30: 29–34.

Veen, Marijke van der. 1988. “A Life of Luxury in the Desert? The Food and Fodder Supply
to Mons Claudianus.” JRA 11: 101–116.

Wilcken, Ulrich. (1899). Griechische Ostraka aus Ägypten und Nubien: Ein Beitrag zur an­
tiken Wirtschaftsgeschichte. Vol. 1. Berlin: Gieseke & Devrient.

Yadin, Y. 1961. “Expedition D.” Israel Exploration Journal 11: 36–52.

——— 1962. “Expedition D: The Cave of the Letters.” Israel Exploration Journal 12: 227–
257.

Page 24 of 26
The Finds of Papyri: The Archaeology of Papyrology

Notes:

(1.) Bernardino Drovetti, French consul in Alexandria, pretended to work for the enrich­
ment of the French cultural heritage but primarily wanted to pay off some substantial
debts by trying to sell his collection to the French government.

(2.) This monumental work (first edition 1809–1828) is a sum of all the observations made
by the artists, scientists, and technicians who accompanied Bonaparte's expedition.

(3.) He had been among the first group of Egyptian students who were sent to study in
France, where he was residing in 1822, the year Champollion announced that he had
solved the riddle of the hieroglyphs.

(4.) This fortunate expression was coined by R. S. Poole and is quoted by M. S. Drower
(1982, 312).

(5.) James (1982, 9–36). In 1919 the EEF was renamed the Egypt Exploration Society.

(6.) Petrie (1891, 28). Recycled papyrus from cartonnage was already known; it was men­
tioned by Letronne in 1826 in connection with fragments of cartonnage from the Pas­
salacqua collection (P.Paris, pp. 410 ff.).

(7.) Letter of Oct. 22, 1888, quoted in Wilcken (1899, 26).

(8.) Their descriptions make us smile today, but this is how Grenfell and Hunt report exca­
vations directed by scholars as opposed to sebbâkhîn and other looters (Grenfell, Hunt,
and Hogarth 1900, 20).

(9.) This trade and the possibility of exportation were not made illegal until Law 117 was
enacted in 1983.

(10.) See Davoli (1998, 76) for a good evaluation of this excavation.

(11.) I am thinking of the excavations of the Red Sea harbors by S. E. Sidebotham, W.

Wendrich, and D. P. S. Peacock, as well as of J.-Y. Empereur, archaeologist and epigraphist,
at Alexandria. The exploration of Mons Porphyrites by D. P. S. Peacock and V. Maxfield is
linked to the study of imperial power. It is a model of nondestructive excavation, in which
the search for papyrological documents was a secondary consideration.

(12.) This is also the case with the Dakhleh Oasis Project which has studied that oasis in
the Western desert since 1978 from both a long-term geological and climatic point of view
and a historical perspective.

(13.) Located in the Eastern Desert, Mons Claudianus is a group of grano-diorite quarries
that were exploited by the Roman emperors beginning around the time of Claudius or
Nero. The peak of activity occurred under Trajan, who used the quarries for the numer­
ous monolithic columns in the Basilica Ulpia. Mons Claudianus has also produced most of
the columns in front of the Pantheon in Rome, which was reconstructed under Hadrian.

Page 25 of 26
The Finds of Papyri: The Archaeology of Papyrology

(14.) The “rabbit holes” left by illicit diggers expose the mounds to the rain, which is not
so rare in the Eastern Desert as elsewhere in Egypt, and the previously sealed layers are
quickly penetrated and destroyed. The proximity of modern works—mining, quarries,
road building projects, and so on—also presents risks. Several garbage heaps from an­
cient forts have been disturbed by bulldozers or simply removed. Recently somebody took
a bulldozer right through the walls of the fort Iovis, mentioned in the Itinerarium An­
toninianum.

(15.) See what Marijke van der Veen (1988) deduces from the texts.

(16.) Pack2 78 (Wrst century CE). The circumstances of the find were reported by Mari­
ette; see P.Paris, p. 417.

(17.) The practice remains a local one, however, since Talit and Tebtynis are no more than
Wve kilometers apart as the crow flies.

(18.) See P.Mich. XVIII pp. 91ff.

(19.) Grenfell, Hunt, and Hogarth (1900, 24ff). Second best were rooms that had been
transformed into garbage deposits, and third the actual rubbish mounds. The latter two
categories are discussed later.

(20.) There maybe a good reason for this: In scrutinizing unpublished letters exchanged
between Lefebvre and Maspero, Jean-Luc Fournet has recently realized that the archive
had been found by illicit diggers before Lefebvre's excavations, which were fruitless.

(21.) The cave is in a cliff of the Nahal Hever, one of the valleys on the west side of the
Dead Sea. The exploration of the caves was undertaken in 1960 after it was discovered
that Bedouin had found documents there that they were selling on the Jordanian antiqui­
ties market. The Israeli authorities were enthusiastic about the project, which was car­
ried out in style. The army arranged the logistics for four parallel expeditions with heli­
copters, aerial photography, electricity in the caves, mountaineering equipment, mine de­
tectors, and even a number of soldiers who participated in the excavations.

(22.) On this subject, read the pages full of humor and humility by Jean-Pierre Brun, an
expert “garbologist” (“Méthodes et conditions de fouilles des fortins et des dépotoirs, ou
les affres dʼun Gallo-Romain en Égypte,” in Cuvigny, ed. 2003, 61–71).

(23.) That is the period before the first Fayyum find.

Hélène Cuvigny

Hélène Cuvigny is Directeur de recherches, Centre National de la Recherche Scien­

tifique (Institut de Recherche et dʼHistoire des Textes).

Page 26 of 26
The History of the Discipline

The History of the Discipline

James G. Keenan
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0003

Abstract and Keywords

This article suggests that a strong case can be made for dating the beginning of papyrolo­
gy to 1752, the year in which papyri were first discovered at Herculaneum. Nevertheless,
perhaps because papyrology came to be associated with Egypt and related documents,
not Italy and philosophical texts, papyrologists came to identify 1788 as marking the be­
ginning of their discipline. In that year Danish classicist Niels Iversen Schow published a
Greek papyrus that recorded a series of receipts for work performed in 193 CE on the irri­
gation dikes in the Fayyum district of Egypt. The credit for the first modern edition of an
integrated series of papyri goes to Amedeo Angelo Maria Peyron. An enormous boost to
papyrology is owed to the discoveries of Bernard P. Grenfell and Arthur S. Hunt. The rest
of this article considers papyrus cartels and international organizations.

Keywords: Egypt, papryrology, Herculaneum, Italy, Niels Iversen Schow, Bernard P. Grenfell, Arthur S. Hunt, pa­
pyrus cartels

THE development of papyrology as a discipline presumes the existence in substantial

numbers of finds like those described in the preceding chapter. That being so, a strong
case can be made (Gigante 1984) for dating the beginning of papyrology to 1752, the year
in which papyri were first discovered at Herculaneum, the south Italian city buried by the
eruption of Vesuvius in 79 CE (see chapter 13). These papyri came in the form of cigar-
shaped rolls, crisp and fragile from their ancient incineration. Thus, the earliest history of
papyrology is dominated by the tale of the frustrating and sometimes ruinous attempts to
unroll these recalcitrant pieces (Deuel 1965, chapters 9–10).

Nevertheless, perhaps because papyrology came to be associated with Egypt and docu­
ments, not Italy and philosophical texts, papyrologists came to identify 1788 as marking
the beginning of their discipline. In that year Danish classicist Niels Iversen Schow
(1754–1830) published a Greek papyrus that recorded a series of receipts for work per­
formed in 193 CE on the irrigation dikes in the Fayyum district of Egypt. The papyrus it­
self, a roll with twelve and a half surviving columns, had been bought in 1778 near Mem­
phis by an anonymous merchant. As legend has it, the merchant bought only this one pa­

Page 1 of 21
The History of the Discipline

pyrus of the fifty offered for sale; “the Turks” proceeded to burn the rest, delighting in
the resulting aroma. Details of the story, especially its olfactory coda, have been contest­
ed, but it is certain that the papyrus that escaped destruction was donated to Cardinal
Stefano Borgia. Hence, it is sometimes known after the cardinal as the Charta Borgiana,
but it is also called the Schow papyrus after its editor (full details, Martin 2000b; see Liti­
nas 2007 for a new fragment found in (p. 60) Lisbon). Initially housed in the cardinal's mu­
seum at Velitri, it now resides in the Museo Nazionale Archeologico in Naples (Preisen­
danz 1933, 69–73).

If credit for the first edition of a papyrus belongs to Schow, credit for the first modern
edition of an integrated series of papyri goes to Amedeo Angelo Maria Peyron (1785–
1870), an Italian Jesuit scholar of Coptic and Greek, for his Papyri graeci regii Musei Tau­
rinensis (Greek Papyri of the Royal Museum at Turin) (1826–1827). In this volume Peyron
published a set of papyri from the so-called choachytai dossier. These were long, well-pre­
served Ptolemaic-period documents concerned with litigation between Egyptian mortuary
priests (“choachytai” are literally “water pourers” in Greek) and two Hellenized Egyp­
tians. Not long ago, Peyron's edition was judged a “miracle” (Bingen 1994, 43) not just by
the nonexistent standards of his own day but by universal standards as well. Although
other papyrus editions sporadically followed, none matched Peyron's in editorial skill or in
sensitivity to what documentary papyri reveal about human history (Bingen 1994; Mon­
tevecchi 1994). Nevertheless, great as it was, Peyron's achievement was overshadowed
by the expectations raised by Champollion's decipherment of Egyptian hieroglyphs. In
contrast to what the new decipherment promised, the contents of documents like
Peyron's were regarded as “vulgar” and “negligible” (Montevecchi 1994, 27, quoting An­
gelo Cardinal Mai [1782–1854])—a prejudice against which papyrology has had to strug­
gle ever since.

Because it appeared at a time when the word papyrology did not exist and the discipline
had not yet been distinguished from what is now called Egyptology, Peyron's edition be­
longs, like the Herculaneum discoveries, to the protohistory of papyrology. During these
early years, most papyri that were shipped to Europe were acquired (rather ruthlessly by
today's standards) as parts of collections of other, more valued antiquities. The driving
forces, both through purchase and rough-and-ready excavation, were often diplomats like
Giovanni Anastasi (1780–1860), Bernardino Drovetti (1776–1852), and Henry Salt (1780–
1827) (Fagan 1975; Reid 2002).

The number of acquisitions grew exponentially in 1877 when peasants digging for fertiliz­
er in the ancient mounds of Kimâm Faris, north of the modern capital of the Fayyum
province, discovered thousands of papyri; to distinguish it from a second, somewhat less
spectacular, clandestine discovery, this event came to be called “the first Fayyum find.”
Most of these papyri, along with others of Herakleopolite provenance, were purchased in
Cairo by Austrian businessman and antiquities dealer Theodor Graf (1840–1903) (Preisen­
danz 1933, 110–124). Graf sold the papyri in lots, first to the Louvre and the Berlin Muse­
um and then, in 1883 (or 1884), to Archduke Rainer (1827–1913). Rainer commissioned
Graf to make further purchases in his behalf, a working arrangement that lasted until

Page 2 of 21
The History of the Discipline

1889. Ten years later, the archduke donated his acquisitions to the Austrian National Li­
brary, creating in one stroke what is now the world's second or third (after Berlin) largest
collection. Importantly for their recovery and preservation, papyri had come to be valued
as artifacts in and of themselves.

If the Austrians and others assembled collections largely through purchase, the
(p. 61)

English, encouraged by the successful efforts of William Matthew Flinders Petrie (1853–
1942) in the Fayyum in the late 1880s, led the field in excavating for papyri. There soon
followed another major turning point in papyrology's late nineteenth-century history. This
was not a single event but a series of events over two years, 1891–1892, that served to
put papyrology on the map. The year 1891 is often referred to in papyrological circles as
the “miracle year” (annus mirabilis). It witnessed Swiss-born Irishman John P. Mahaffy's
(1839–1919) publication of the first volume of Petrie papyri, as well as editions by Eng­
lishman Frederic G. Kenyon (1863–1952)of the British Museum's “Constitution of Athens”
and, in separate publications, papyrus rolls with mimes of Herondas and speeches by the
Athenian orator Hypereides. In 1892 there appeared the first fascicle in the extensive se­
ries of Greek documentary papyri from the Berlin Museum (Aegyptische Urkunden aus
den Königlichen [later Staatlichen] Museen zu Berlin, Griechische Urkunden). This led to
the coining of the word papyrologist by Jules Nicole (1842–1921) of Geneva, who in 1896
referred (in French) to the editors of the Berlin papyri as “papyrologistes” (van Minnen
1993, 7, 7n11). In English, “papyrology,” after a hesitant appearance in 1898 (“In the de­
partment of ⋆papyrology, if we may use such a word”), achieved full acceptance by 1900
(“Papyrology is the Greek study which is devouring all the rest”) (Oxford English Dictio­
nary [1933], vol. 7, 442; both citations from anonymous reviews in the literary journal
Athenœum). A year later it had become a “science” (Robinson 1901, 212).

At first papyrology denoted specifically the decipherment and presentation of texts writ­
ten on papyrus (and ostraca), especially in Greek. Texts in the Egyptian language in what­
ever script—hieroglyphic, hieratic, demotic—remained the preserve of the field that came
to be called Egyptology. So did texts in Coptic (Egyptian written in a predominantly Greek
script), at least until recent times, when papyrologists (and New Testament scholars)
have increasingly tried to master Coptic without experiencing the traditional, and some
think essential, initiation to the earlier forms of Egyptian language and scripts. Papyrolo­
gy, however, soon also developed a meaning that extends beyond the editing of papyrus
texts to include their use in the study of Egyptian history, society, and economy from ap­
proximately the third century BCE until the early eighth century CE. This is sometimes re­
ferred to as “the papyrological millennium.”

An enormous boost to papyrology in both senses is owed to the discoveries of Bernard P.

Grenfell (1869–1926) and Arthur S. Hunt (1871–1934). These young Oxford classicists
had, under the auspices of the Egypt Exploration Fund (later, Society), undertaken exca­
vations in Egypt in specific quest of papyri with a special mandate for seeking out Christ­
ian and literary texts. They first excavated in surveylike fashion at various sites in the
Fayyum. But in 1896, with a geographically extended permit, they turned their attention
beyond the Fayyum to the ancient city of Oxyrhynchus, a site conceded to them by Petrie.

Page 3 of 21
The History of the Discipline

Failing on the basis of past (p. 62) experience to uncover papyri in either of the most obvi­
ous locations, the Graeco-Roman cemetery and the town proper, they decided to dig in
the ancient town's rubbish dumps. On January 11, 1897, they hit the mother lode, recov­
ering numerous papyrus fragments of all kinds. The next day they uncovered a special
prize, a small, crumpled codex leaf containing a series of “sayings of Jesus,” later identi­
fied as coming from the apocryphal Gospel of Thomas. And the finds kept on coming in
enormous quantities, not just in 1897 but in later years as well, until 1907, when Grenfell
and Hunt's excavations at Oxyrhynchus, after six seasons of digging, ceased (Deuel 1966,
chapters 9–10; Turner 1982). (These excavations were the source of the world's largest
collection of papyri.)

Almost as impressive as the Oxyrhynchus finds themselves was the swiftness of their pub­
lication. In 1898, within eleven months of the arrival in England of the first batch of pa­
pyri, Grenfell and Hunt produced the first volume of Oxyrhynchus Papyri. Besides the
“sayings of Jesus,” previously published in pamphlet form, other prizes in P.Oxy. I includ­
ed poetic fragments of Alcaeus and Sappho. In all, 158 texts were edited in full, and an­
other 49 in brief description.

Significantly, the first volume of Oxyrhynchus Papyri adopted a new format for exhibiting
papyri in print. Grenfell and Hunt gave every papyrus a number and a title; a heading giv­
ing measurements and date; a concise introduction to the text's contents; Greek text in
modern form (accents included); critical apparatus indicating scribal aberrations and lin­
guistic anomalies; and line-by-line commentary. Positioned before the commentary—and
revolutionary at the time—were translations for most of the texts, provided, as Grenfell
and Hunt say in their preface, “at the request of several subscribers to the Graeco-Roman
branch” of the Egypt Exploration Fund.

For both the interested public and scholars, this template made the Oxyrhynchus papyri
more accessible than those published in antiquated, less congenial formats. Today, any
edition of a papyrus without translation is deemed incomplete: The translation, in effect,
completes the edition. In this view (Youtie 1973, vol. 1, 12), the translation is not added
“for its own sake” but rather to let the reader know “at any moment what meaning the
editor attributed to any and every passage of the text …. [I]t is there as additional and al­
most certainly more effective commentary.”

Grenfell and Hunt are therefore rightly credited with having created a “style,” at once a
hallmark of the Oxyrhynchus editions and a model for modern editions in general. It is,
nonetheless, a particular style that is dominated by the text-based interests of philology
and typology. It does not normally allow for extension into the wider social or historical
ramifications of individual texts or groups of texts. Moreover, within the Oxyrhynchus vol­
umes, papyri were (and still are) grouped in an established order by type (literary, sublit­
erary, documentary); they were thereby detached from their archaeological context and
dissociated from related papyri of different type (Gagos, Gates, and Wilburn 2005). In the
later publication of literary (p. 63) papyri from Oxyrhynchus, this style reached its ideal in
the impeccable but austere editions of Edgar Lobel (1888–1982): “[N]o parade of scholar­

Page 4 of 21
The History of the Discipline

ship, no clutter of bibliography, … an insistence on fact and precision, a distaste for easy
solutions and grandiose speculations” (preface to P.Oxy. L).

It happened that while Grenfell and Hunt were migrating between Egypt and Oxford, ex­
cavating and deciphering, there lived in Germany a scholar who found fascination in even
the most pedestrian of tax receipts scrawled on potsherds (Wilcken 1899). This was Ul­
rich Wilcken (1862–1944), a devoted student of Theodor Mommsen (1817–1903). Momm­
sen has long been credited with originating the prediction that just as the nineteenth cen­
tury had been the century of epigraphy, so the twentieth would be the century of papyrol­
ogy (the credit perhaps belongs to Ludwig Mitteis [1859–1921], Martin 2000a). It was
Wilcken who decided to found the first journal devoted exclusively to papyrology, the
Archiv für Papyrusforschung, whose first fascicle appeared in 1900; the first volume was
completed in 1901. Earlier, papyrologists had found homes for their shorter writings in a
variety of specialist journals. Now, by Wilcken's intention, papyrological research would
have its own Vereinigungspunkt (concentration point) not only for papyrological research
in the strict sense but also for all studies related to papyrology (ancient history, epigra­
phy, numismatics, theology, philology). In his foreword to the first issue, Wilcken enunci­
ated a program for the new periodical. Though not excluding literary papyri, the focus
was to be on documents, both in and of themselves and in their relation to ancient history
and culture. Submissions were not restricted to those in German. Articles in English,
French, Italian, and Latin were also invited.

Wilcken, then at Würzburg, was the journal's sole editor, but as associates he had assem­
bled an international cast of papyrological “all-stars”: Otto Gradenwitz (Königsberg), B. P.
Grenfell and A. S. Hunt (Oxford), Pierre Jouguet (then at Lille), Frederic Kenyon (London,
British Museum), Giacomo Lumbroso (Rome), J. P. Mahaffy (Dublin), Ludwig Mitteis
(Leipzig), Jules Nicole (Geneva), and Paul Viereck (Berlin). Many contributions to Archiv's
first volume seem to have been by invitation. It contained a series of articles (Aufsätze),
reports and reviews (Referate und Besprechungen), and brief communications (Mitteilun­
gen). Wilcken and nearly all of the associates contributed to the first volume, which also
included articles by Mommsen and Ulrich von Wilamowitz-Moellendorff (1848–1931), re­
spectively the premier ancient historian and the foremost philologist of their time (and
quite likely of all time). Among the articles was Wilcken's “general register” of papyri
published to date. Among the reports was one of Wilcken's firm but gracious critiques of
recent papyrus editions; there were also reports on Christian texts by Carl Schmidt
(1868–1938) and on new legal documents from Oxyrhynchus by Mitteis. The brief commu­
nications included Grenfell and Hunt's description of their excavations at Tebtynis for the
University of California in the winter of 1899–1900.

Wilcken intended to organize the new discipline of papyrology on an international

(p. 64)

scale. If Archiv was to be devoted largely to documents, it was also to connect the results
of papyrological research with those of other subfields of Altertumswissenschaft and to
establish a form of quality control over papyrological editions. Once a papyrologist was
assured that documents had been well edited, the next step was to see to their use in oth­
er disciplines, to the mutual illumination of papyri, literature, inscriptions, ostraca, and

Page 5 of 21
The History of the Discipline

coins. But the still higher goal consequent to the foregoing was, as Wilcken put it, “to
grasp ancient culture in all its manifestations in the liveliest possible way” (1901, v).

Conscious of how difficult it was for nonpapyrologists to access the results of papyrologi­
cal research, Wilcken later collaborated with Mitteis, the leading jurist among papyrolo­
gists, to create in 1912 a monumental assessment of the documentary papyri published to
date. This was the Grundzüge und Chrestomathie der Papyruskunde. It appeared in two
halves, one historical, one juristic, each containing a volume introducing its subject and a
volume containing hundreds of illustrative documents. These four volumes, still authorita­
tive today, defined papyrology's major concerns while simultaneously enshrining the divi­
sion of Egypt's papyrological millennium into Ptolemaic, Roman, and Byzantine periods
and identifying juristic papyrology as the discipline's principal subfield. Also influential
over the long term was Wilcken's view (1912, XV) that Egypt held a “special place” (Son­
derstellung) both in the Hellenistic world during the Ptolemaic period and later on under
the Roman Empire. General conclusions could therefore be drawn from the Egyptian evi­
dence only with the greatest caution.

Wilcken's magisterium cast a long shadow, so it is no surprise that the young Marcel
Hombert (1900–1992) assimilated this brief passage from Wilcken, incorporated it verba­
tim into his own inaugural lecture in Brussels on October 27, 1925 (“La papyrologie
grecque,” http://www.ulb.ac.be/philo/cpeg/hombert.htm), and there referred to Egypt in
the Hellenistic world and under the Roman Empire as having “une position toute partic­
ulière.” Wilcken's measured adoption of the idea that Egypt was a “special place” there­
fore had an impact that was specific and practically immediate. It was also lasting and
sometimes exaggerated. Though not Wilcken's intention, it has served to diminish or ex­
clude the use of Egyptian evidence for the general study of ancient (namely, classical) his­
tory. This is most apparent in work on the Roman imperial period, where on many occa­
sions Egypt with its papyrus evidence was judged to be so “untypical” that it could be
readily dismissed. Moreover, what may be called the “Sonderstellung problem” is also in­
extricably bound to the problem of historical continuity, especially that from the Ptolema­
ic period into the Roman. On the one hand, simply put, if the Romans adopted Ptolemaic
forms, then Egypt remained as peculiar under the Romans as it had been under the
Ptolemies; on the other hand, if the Romans introduced their own forms, then Egypt's
place was not so “special” after all, and its evidence could not be casually set aside. In the
past generation, criticism of the model of (p. 65) Ptolemaic-to-Roman continuity (Lewis
1970, 1984) naturally led to some erosion of the “Sonderstellung” model (e.g., Bowman
and Rathbone 1992). In the meantime, papyrologists have composed detailed defenses of
the value of the Egyptian evidence both for the Hellenistic world (Heinen 1989) and for
the Roman Empire (Geraci 1989). Nevertheless, from a programmatic standpoint, per­
haps the most balanced assessment is that advanced years ago by Cambridge ancient his­
torian A. H. M. Jones (Jones 1942, esp. 286–287) in a succinct but pointed statement on
the need to bring history and papyrology together based on the latter's contributions to
the former and the belief that the papyrus evidence had significance that transcended
particularities of place. Jones, of course, amply honored his own recommendation in nu­

Page 6 of 21
The History of the Discipline

merous articles (e.g., Jones 1974) and, most extensively, in his massive The Later Roman
Empire, 284–602 (1964).

Another, perhaps even more important consequence of Wilcken's research agenda is that
the study of papyrus documents came to signify papyrology as a whole. The result was a
lasting tendency to separate, or marginalize, the study of literary papyri, even though
classical scholars, many of whom have only passing familiarity with the documents, dis­
play enormous enthusiasm each time a new literary papyrus is published. All of this aside,
Wilcken was obviously a magnificent synthesizer, justly credited as the first scholar to de­
velop a vision of the field as a whole. Nevertheless, if Wilcken was papyrology's first (and
perhaps only) truly great synthesizer, its first great organizer was his slightly older con­
temporary Friedrich Preisigke (1856–1924) (Montevecchi 1988, 35–36), longtime director
of the Post Office and Telegraph in (then) Straßburg. In retirement Preisigke initiated,
collaborated upon, and supervised four major projects: “la grande tétralogie de
Preisigke” (Bingen 1977, 42; cf. Preisendanz 1933, 199–202):

1. the Sammelbuch (1913–), a series of volumes to reprint documents (papyri, ostra­

ca, inscriptions) that had been published piecemeal in periodicals
2. the Berichtigungsliste (1922–), volumes that record corrections made to already
published texts
3. the exhaustively titled Namenbuch (1922), an alphabetical list of personal names
4. the multivolume Wörterbuch (1925–), a dictionary of the documentary papyri

The generosity of Preisigke's endeavors matched the internationalism of Wilcken's vision of the
field. In truth, papyrology has a long tradition of international cooperation. Originally this was
due to the efforts of scholars like Wilcken and Preisigke as just described and to lively corre­
spondence across national borders. Two examples of more formal international cooperation are
the so-called papyrus cartels and the international meetings that have been held under the aus­
pices of the Association Internationale de Papyrologues (International Association of Papyrolo­
gists).

(p. 66) Papyrus Cartels

In the 1920s and 1930s numerous papyri were being offered for sale on the antiquities
market—an expectedly unregulated market in which papyri were sold at high prices and
scattered worldwide with little tracking. In response, there was formed what has been
called a “cartel,” although it was more precisely “a syndicate of buyers” (Turner 1980,
36). The idea was not new: The Germans had tried something like it earlier in the
Deutsches Papyruskartell, which functioned between 1902 and 1914, purchasing papyri
from local dealers in Egypt and distributing them to German member institutions, includ­
ing, between 1903 and 1906, papyri that turned out to have been filched from Grenfell
and Hunt's site at Oxyrhynchus (Preisendanz 1933, 210–211; Martin 2007).

The new cartel, begun in or shortly after 1920, continued to operate through the
mid-1930s, headed by the British Museum with H. I. Bell (1879–1967) as the principal
keeper of records. Bell inventoried and oversaw the dispersal of papyri to a number of
Page 7 of 21
The History of the Discipline

member institutions, mainly American, according to their particular interests and respec­
tive purchasing powers. These at first included Cornell, Michigan, Princeton, and Wiscon­
sin; Columbia and Yale joined later. Geneva and Oslo were European members, as was, of
course, the BritishMuseum. This was a rather fluid association whose history, in the inter­
ests of what is called “museum archaeology” (Vandorpe 1994)—the virtual reassembling
of papyrus archives and dossiers physically scattered through illegal excavations and the
antiquities market—would be worth reconstructing based on the surviving correspon­
dence between representatives of the member institutions. The most active of these was
the University of Michigan, whose collection was initiated before the syndicate's creation
through purchases made by Francis W. Kelsey (1858–1927), namesake of today's Kelsey
Museum of Archaeology. The University of Michigan's purchases through and apart from
the syndicate continued even during its excavations at Karanis and Soknopaiou Nesos
(1924–1935) (Gagos 2001, esp. 517–525). When buying independently, Michigan seems to
have bought mainly from Maurice Nahman (1868–1948), in whose “petit palais de style
arabe” (Capart 1947; photo: Pintaudi 1993, 165) at 27 Madebegh Street, Cairo, antiqui­
ties were displayed as in a gallery (WWWE 305 [see acknowledgments, this chapter]; pho­
tos: Pintaudi 1993, 157 [Nahman], 167 and 169 [the gallery]).

Part of the syndicate's purpose was to keep a lid on prices. Nonetheless, competition from
nonsyndicate institutions and amateur buyers, together with the wealth and alleged igno­
rance of the U.S. member institutions, which seemed liable to “spoil the market” (letter
from Hunt to Bell, BL Add MS 59512, item 202), continued to drive prices skyward. A
glimpse at buyers' competitiveness is provided by a handful of letters written by Nahman,
three from Paris (Le Grand Hotel, 12, Boulevard des Capucines) and two from Cairo (27,
Rue el-Madabegh; later 27, rue Cherif Pacha), to Medea Norsa (1877–1952) between
1932 and 1935. At (p. 67) this time the site of Oxyrhynchus was being stripped by illegal
excavations whose finds Nahman refused on principle to buy (Pintaudi 1993, 156–169). It
is true that Nahman's customers included American institutions represented by A. E. R.
Boak and Michael Rostovtzeff, but the villain in Nahman's letters turns out to be not the
Americans but an Italian archaeologist-papyrologist from Milan.

A crisis having been reached, Marcel Hombert published in the Belgian periodical
Chronique dʼÉgypte a brief but passionate treatment of the subject (Hombert 1933). In a
largely forgotten but minor classic of Orientalist writing, Hombert pointed to the competi­
tion among institutions to acquire papyri, a limited resource obtainable only in Egypt. He
credited the syndicate with partial success in price control and considerable success in
substance, singling out for special mention its acquisition of the Roman emperor
Claudius's famous letter to the Alexandrians. Hombert continued with an imaginative
sketch in which an amateur, would-be purchaser of papyri, having failed to make any ac­
quisitions from European dealers in Alexandria or Cairo, turns to native antiquities shops
near the fashionable Shepheard's Hotel (Reid 2002, 73, photo on 74). There the hapless
buyer becomes a target for disinformation. He quickly finds himself a neighborhood
celebrity, hassled by salesmen hawking postcards, walking canes, and flyswatters. He is
then cozened by a shady dragoman out into the primitive countryside into the presence of
a surly villager dealer. There he is eventually offered coffee and shown some dust-covered
Page 8 of 21
The History of the Discipline

terracotta lamps, belle époque bronze figurines, and scarabs of dubious authenticity.
When papyri are finally brought out, they prove to be as expensive in the abode of “this
obscure peasant” as “in the most luxurious antiquities gallery in Cairo.”

But despite the frustrations visited upon Hombert's notional purchaser, buyers still went,
and they still bought—at exorbitant prices. What could be done? The only hope seemed to
lie in the equally expensive alternative of excavation—in which a small country like Bel­
gium could hardly expect to compete on its own. At this very time, the University of
Michigan, for example, was still excavating at Karanis (1924–1935), Yale was midway
through its excavations at Dura-Europos (1928–1937), a small city in Syria on the bank of
the Euphrates River (Hopkins 1979), and Italian excavations at Tebtynis for the Società
italiana per la ricerca dei papiri in Egitto were also in full swing (1929–1936) (e.g., Begg
1998).

International Organizations and “schools”

If Hombert's pleas seemed fruitless and if the cartel was relatively short-lived, the institu­
tion of international congresses of papyrology has been useful and lasting. Papyrologists
as a group first came together at an international meeting in Brussels (p. 68) in 1930. At
the invitation of Jean Capart (1877–1947), famed Belgian Egyptologist (WWWE 82–83,
with photo), a special papyrology session was held as part of an “Egyptological week” cel­
ebrating the centenary of Belgium's independence. After the fact this came to be reck­
oned as the First International Congress of Papyrologists. At this meeting an Internation­
al Committee of Papyrology, with Cairo-based Frenchman Pierre Jouguet (1869–1949) as
president and Marcel Hombert as permanent secretary in Brussels, was established to
oversee future meetings. The following year, at Leiden, papyrologists met in “an au­
tonomous session,” once again as part of a larger meeting, this time the Eighteenth Inter­
national Congress of Orientalists, now credited as the Second International Congress of
Papyrologists.

By the standards of today's congresses, both meetings were modest affairs in terms of the
number of participants and the number of papers given, but the names of almost all the
participants are those who were, or became, giants in the field (Habermann 2001, 102–
103, with notes). Many of the papers at the first meeting, whether delivered in person or
by proxy, were progress reports on papyrology in the various countries represented. Of
the others, for example, Victor Martin (1886–1964) of Switzerland provided an update on
a project to reedit the papers of Flavius Abinnaeus, a fourth-century military commander;
a young Belgian, Claire Preaux (1904–1979), spoke on the Zenon papyri, the largest and
most famous archive of the Ptolemaic period; and the eminent ancient historian and Russ­
ian emigré Michael Rostovtzeff (1870–1952) presented results of the Yale excavations at
Dura-Europos.

For the Leiden meeting, the International Committee of Papyrology had set itself the task
of establishing a system of symbols for use in editing texts, partly modeled on those used
in editing inscriptions. In the morning session of September 10, 1931, B. A. van Gronin­
Page 9 of 21
The History of the Discipline

gen (1894–1987) of Holland made the principal proposal but also read supplementary rec­
ommendations from H. I. Bell and A. S. Hunt, who were not in attendance. After discus­
sion, the proposals of Bell and Hunt were accepted as friendly amendments to van
Groningen's proposal. The following day, the papyrologists in attendance unanimously ap­
proved not only what has come to be called “the Leiden system” (i.e., system of editorial
signs) but also the more general standardization of editions recommended by Bell.1

After the Leiden meeting, papyrologists met three more times on their own (Munich,
1933; Florence, 1935; Oxford, 1937), but the politics of the late 1930s and the Second
World War interrupted the series. The meeting planned for Vienna in 1939 (Habermann
2001) was never held. The war itself, based on reports gathered afterward by Marcel
Hombert (Hombert 1947, 1948), wrought terrible personal loss: A young Austrian papy­
rologist and the conservator of the Vienna papyrus collection were both killed in battle; a
well-known French papyrologist from (p. 69) Strasbourg (Paul Collomp, 1885–1943) was
murdered by the German police in Clermont-Ferrand (the University of Strasbourg's
place of exile); a German went missing in action on the Russian front in the war's final
months; a young Dutch papyrologist was executed for his role in the Resistance. In addi­
tion to these tragedies, significant material and cultural destruction was sustained: Pa­
pyri were lost or destroyed, libraries were bombed and burned, and tremendous disrup­
tion occurred, including the shipping of the British Museum's great collection to Aberyst­
wyth, Wales, for safekeeping, which was just one event in the mass relocation of all kinds
of valuable collections at the time (Nicholas 1994).

All of this notwithstanding, several papyrologists met in Brussels in August 1947 with the
intention of restoring international relations among colleagues and devising a plan to
recommence regular meetings. The first such meeting took place in Paris in 1949 (Bingen
1977, 35). There a decision was approved to formalize the organization of the internation­
al congresses through the creation, under the auspices of UNESCO, of the Association In­
ternationale de Papyrologues (AIP). As before, the secretariat was to be in Brussels but
with elected officers and board members representing many countries. A major force in
the organizational effort was (once again) Marcel Hombert, who became the association's
first secretary-treasurer, serving until 1961 when he was succeeded by Jean Bingen
(1920–). Since 1949, with the exception of one four-year interlude, papyrologists have
held their congresses every three years. (The 2007 meeting was held in Ann Arbor, Michi­
gan; Geneva, Switzerland, is the site for the 2010 meeting.)

The coining of what became the association's motto, “amicitia papyrologorum,” is attrib­
uted to juristic papyrologist Leopold Wenger (1874–1953), who as host of the festivities
on the evening of the last day (September 7, 1933) of the Third International Congress
(Munich) toasted “the friendship of papyrologists.” By the fourth congress (Florence,
1935), as the political horizon darkened, “the friendship of papyrologists” could be said to
have become proverbial (sprichwörtlich). This was perhaps in part because of the (in ret­
rospect desperate?) faith in what was then referred to as the “new humanism,” a firm be­
lief in human progress based on a renaissance in classical scholarship, somehow much
dependent upon the “fraternity” of papyrologists, sentiments glowingly expressed during

Page 10 of 21
The History of the Discipline

the Florence congress (Aegyptus, serie scientifica 5, Milan [1936]; cf. Keenan 2002, 226–
227).

The meaning of the motto is usually assumed but rarely defined. It is an ideal that clearly
implies that the field of papyrology is larger than individual papyrologists, no matter what
their several contributions. It alludes to a code of courtesy even in cases of strong dis­
agreement, where criticism is directed at an anonymous “editor” and polemics are
frowned-upon exceptions. It acknowledges that the field is in a constant state of growth
and revision in which all papyrologists are partners. It suggests that the friendship is per­
sonal as much as professional. It also points to the internationalism of the field. As
Wenger put it somewhat later, in 1940, in still darker days, “A realm of knowledge like pa­
pyrology, built to so great a degree upon (p. 70) the collaboration of the international com­
munity of learning, makes it extraordinarily difficult, nay impossible, to separate neatly
the share of the work of an individual nation from the great totality” (http://
www.ulb.ac.be/assoc/aip/amicitia.htm). A few years later, in a symbolically significant ges­
ture, H. I. Bell, at the beginning of an article in the Journal of Egyptian Archaeology (31
[1945]: 75), was pleased to announce that he had received an offprint and with it news
from Italian papyrologist Medea Norsa that she was alive and well despite the current
fighting in northern Italy in and around Florence. Within recent memory, the career that
probably best exemplifies the ideals of the “amicitia” was that of Sir Eric G. Turner (1911
–1983) (Parsons 1987), worthy successor in Great Britain to H. I. Bell in this regard
(Thomas 1966).

There is an irony in all this, I think. That is, despite friendships across real and sometimes
ideological battle lines (Gigante 1986; Habermann 2001), despite Wenger's credo, and de­
spite the AIP's statute (article 1) codifying the ideal of “fostering international collabora­
tion in the field of papyrology,” papyrology's accomplishments have tended to be present­
ed and assessed, as in the modern Olympics, according to the boundaries of nation-states
(e.g., Preisendanz 1933, 160–300). So, for example, the Belgians (Montevecchi 1988, 37–
38) and the Dutch may justly be credited with being in the forefront of combining docu­
ments in Greek and in demotic Egyptian for the study of Ptolemaic Egypt. Such is their
conjoined contribution that it has been possible to refer to the accomplishments of the
“schools of Leuven, Brussels, and Leiden” (Bagnall 1982–1983, 17; my emphasis). A mon­
umental product of this fusion of Greek and demotic studies in the Low Countries is the
Guide to the Zenon Archive produced by six scholars under the editorship of P. W. Pest­
man (Pestman 1981).

The Germans, along with the Poles (especially Raphael Taubenschlag [1881–1958] and
the University of Warsaw's Journal of Juristic Papyrology), have made the most substan­
tial contributions in juristic papyrology (e.g., Rupprecht 1989), a tradition especially
strong at Marburg. In the wake of one of papyrology's pioneering editors, Carl Wessely
(1860–1931) (Gerstinger 1932; Hopfner 1933), Austrian papyrologists and their asso­
ciates have contributed immensely to knowledge of the Byzantine and early Islamic peri­
ods, based on editions of papyri from the Vienna collection, whose origins were sketched
above. French papyrologists have made and continue to make exceptional contributions

Page 11 of 21
The History of the Discipline

to study of these late-period papyri, especially in their historical interpretation. In this

they carry on the tradition of Jean Maspero (1885–1915), initiator with H. I. Bell of the
specialty known as “Byzantine papyrology.” (Maspero, in one of papyrology's greatest
losses, was mortally wounded during an assault at Vauquois on the Lorraine front on Feb­
ruary 18, 1915; G. Maspero 1916, WWWE 278–280.)

Soviet papyrology as revealed by I. F. Fikhman (Fikhman 1999, to cite only the longest in
a series of archival studies) came to center upon Grigorij F. Zereteli (1870–1939), a half-
Georgian (but cultural Russian) later in life based in Tbilisi. (p. 71) Though politically un­
engaged, he was arrested on May 24, 1938 (earlier arrests had occurred in 1918 and
1930), and is presumed to have died in 1939 (I. F. Fikhman, letter, September 25, 2005), a
blow from which Russian papyrology has never recovered. Finally, many Italian scholars,
inspired in Naples by the leadership and dynamism of Marcello Gigante (1923–2001),
founder in 1969 of the Centro Internazionale per lo Studio dei Papiri Ercolanesi, are dedi­
cated to research on the Herculaneum papyri, a phenomenon amply attested in the pub­
lished atti (proceedings) of the Florence Congress of 1998 (Keenan 2002, 222–223). The
Italians have also been indefatigable excavators of archaeological sites of papyrological
interest (Bonacasa 1989; Casini 2001) and remarkably successful in bringing women into
the discipline.

Whether this rough, incomplete classification of interests should lead to extensive consid­
eration of national schools of papyrology in either the institutional or the intellectual
sense is quite another matter. This is something that awaits systematic sorting out.
Montevecchi's compact list (1988, 33–40), outdated despite its addenda (540–541), in­
cludes many items that hardly qualify as schools. It identifies numerous schools in the in­
stitutional sense but makes only a few nods in the direction of schools as intellectually
conceived. A leading example in both senses is the “Istituto Vitelli,” the first of several
Italian schools or institutes of papyrology (Montevecchi 1988, 37). When founded in 1908
in Florence as the “Società Italiana per la Ricerca dei Papiri Greci e Latini in Egitto,” it
was intended as a school dedicated on a national, even nationalistic, basis to the excava­
tion and publication of papyri (Carozzi 1982). Its principal founder, Girolamo Vitelli,
sensed the Italians were entering the field belatedly. In the end, he left a strong imprint
on the institute: in approach, through his distinguished earlier career as a philologist; in
fashion, through his devotion to the severe style in editing (Gigante 1986); and on posteri­
ty, through a directly traceable line of papyrological descendants, including Medea Norsa
and Vittorio Bartoletti (1912–1967) (Pintaudi 1993).

The French scene is both complicated and enriched by the Institut Français dʼArchéologie
Orientale (IFAO) in Cairo, formally established in 1898 in succession to the school that
had existed since 1880 (Reid 2002, 173–175). The institute was geared primarily toward
the study of the Near East in a wide, multidisciplinary sense, with an emphasis on archae­
ology. Papyrology has had an important but not a dominant role. Traditionally important
for the institute have been its links in France, especially with Paris. For example, an
American, Naphtali Lewis (1911–2005), after a seminar at Columbia on the Zenon archive
under William Linn Westermann (1873–1954), went to the Sorbonne to earn his doctor­

Page 12 of 21
The History of the Discipline

ate. There in 1933–1934 he worked under Paul Collart (1878–1946); the following year,
he studied in Cairo under Pierre Jouguet (1869–1949) and assisted in the decipherment of
the Fouad papyri, published in 1939 (Bagnall 1999). In a telephone conversation (Febru­
ary 28, 2005), Lewis respectfully inclined to view the French combination as more
“atelier” (his label) than school, with Collart—a kind, paternal, and caring (p. 72) teacher
—as the guiding light in Paris and Jouguet the major figure in Cairo, assisted by Octave
Gueraud (1901–1987), who first worked as conservator at the EgyptianMuseum, then as
secretary general of IFAO. In Cairo with Lewis, also working on the Fouad edition, was
Jean Scherer (1911–2001), Lewis's exact contemporary and IFAO “pensionnaire” until his
return to France in 1954, where he was soon to become director of the Sorbonne's Insti­
tute of Papyrology (Chamoux 2002).

In Germany, to give in brief a third and more recent example, the Institut für Altertum­
skunde at the University of Cologne, initiated by Joseph Kroll and Reinhold Merkelbach,
dates to the mid-1960s. Though many of its early associates have scattered geographical­
ly and institutionally, the institute has remained an important center for documentary and
literary study and has been responsible for what has become the world's leading and cer­
tainly most prolific journal of papyrology, the Zeitschrift für Papyrologie und Epigraphik,
whose first issue appeared in 1967.

Of course, there cannot be schools without sufficient numbers of teachers and students in
some form of physical and intellectual association. Thus, it must have seemed shocking to
some in the audience when Herbert Youtie, in his Henry Russel lecture at Ann Arbor in
1962, estimated papyrologists to be “a very small group of scholars, about half a dozen in
the United States and Canada, another half dozen in Great Britain, two or three in the
Scandinavian countries, a dozen or more dispersed over the continent of Europe, one or
two perhaps in the Near East—shall we say a maximum of thirty?” The other 270 or so
then members of the AIP were scholars with other interests. They did not edit papyri
themselves but made ample use of papyrus evidence in their various studies. Youtie effec­
tively distinguished between a “private papyrology,” inhabited by an inner circle (it
seems) of papyrus editors, and a “public papyrology,” with a far greater number of “stu­
dents of ancient literature, ancient historians, jurists, grammarians, palaeographers, the­
ologians, Egyptologists, Copticists, Arabists, archaeologists” (Youtie 1973, vol. 1, chapter
1, esp. 11–12).

Youtie's purpose in his 1962 lecture was, I believe, to narrow the scope of his subject so
as to explain the papyrus-editing process to a general audience. Not quite fifteen years
later, in a lecture given in 1976 at an international meeting of classicists (Bingen 1977),
Jean Bingen defined papyrology in more generous terms: “There are nearly as many papy­
rologies as there are papyrologists, and they [papyrologists] are numerous,” said Bingen.
Decipherment remains the center of the enterprise, but papyrological activities are di­
verse, including, for example, literary studies, Ptolemaic chronology, economic history,
and New Testament textual criticism. The unity of these activities resides principally in
the papyrological origin of their data. The ultimate goal, after all, is the kind of synthesis
Wilcken envisaged. In this way, papyrology, as redefined by Bingen, becomes “a kind of

Page 13 of 21
The History of the Discipline

sociology, in the widest sense,” one that is especially focused on the Greek presence in
Egypt's particular geographic and ethnic milieu.

Of course, as we have glimpsed, this tension between papyrology strictly con­

(p. 73)

strued (by Youtie) and papyrology catholically extended (by Bingen) has been present for
more than a century. It is the tension between the editing of papyri and the wider use of
those editions or between philological examination and historical exploitation. In the
United States, one might, roughly, see this as represented in turn by what can be called
the Michigan tradition and the Yale tradition of papyrology. These can properly be called
schools in both senses of the term, institutional and intellectual; they are epitomized re­
spectively by Herbert C. Youtie and C. Bradford Welles. The two may be conveniently
paired as friendly foils.

Herbert C. Youtie (1904–1980) was hired by the University of Michigan in 1929 specifical­
ly for the purposes of the papyrus collection it was acquiring, as already described,
through purchase inside and outside the cartel and through excavations at Karanis (Koe­
nen 2007). He received his AB degree from the University of Cincinnati in 1927 and an
MA from Columbia in 1928. He trained for a year in Paris under Paul Collart (DʼArms
1980), whom he fondly recalled (this is a personal reminiscence) as gently rebuking those
students who began to transcribe their papyri in ink: Collart (significantly) insisted on
pencil. Youtie earned a diploma in Paris but had no time to spend on a doctorate: The
Michigan papyri required immediate editorial attention. An honorary degree from
Cologne, where his influence became considerable (L. Koenen, e-mail, January 17, 2006),
would come forty years later (1969).

In the meantime, Youtie established a reputation as the consummate editor and critic of
papyrus editions, the main successor to Wilcken in these enterprises. His own editorial
masterpiece, in collaboration with A. E. R. Boak (1888–1962), was The Archive of Aurelius
Isidorus (1960), a set of papers of a local official and landowner in the village of Karanis
in the early fourth century CE. Youtie's interests were almost exclusively editorial; it was
Boak who was concerned with the archive's historical ramifications (Gagos 2001, 522–
523). Similarly, the Michigan scholar who in the long run was most open to the archaeo­
logical possibilities of the Karanis materials was Elinor M. Husselman (1900–1996), cura­
tor of manuscripts and papyri of the Michigan Library and curator also at the Kelsey Mu­
seum from 1925 until 1965 (Gagos 2001, 523–524; Wilfong 1996).

Youtie was not interested in such things per se. Rather, he was a philologist who focused
on texts in and of themselves and in relation to other texts. He was also just about the on­
ly papyrologist openly fascinated by the psychological aspects of reading papyri (for edit­
ing and correcting, see also Turner 1980, chapter 5, “How a Papyrus Text Is Edited,” and
Schubart 1970; cf. Turner 1973). So, in the 1950s and 1960s he delivered lectures in Lon­
don, Ann Arbor, and Cologne (Youtie 1974, 1973, vol. 1, chapters 1 and 2) on what might
these days be called “metapapyrology.” These lectures, taken together in their published
forms, have established themselves as “the papyrologist's Bible.”

Page 14 of 21
The History of the Discipline

The Ann Arbor lecture, subsequently printed as “The Papyrologist: Artificer of

Fact” (Youtie 1973, vol. 1, chapter 1), is probably the best introduction to the general
(p. 74) public of what those papyrologists whose deepest commitment is to the editing of

texts actually do—or should do. Probably all papyrologists, at one time or another, in con­
versation with curious nonpapyrologists, have had to explain that they do not (at least ini­
tially) “translate” papyri; rather, they edit them, and in that process the most difficult task
is transcription, the decipherment of ancient writing in often difficult hands on usually
very damaged surfaces. Youtie, while enunciating what might properly be called an ethics
of papyrology, saw the transcribing of texts as an intense form of reading, much slower
and almost infinitely more arduous but in essence the same as everyday reading. Both,
for example, require or (better) are facilitated by stores of acquired cultural knowledge
and strategies based on prediction. Much of this knowledge Youtie seems to have stored
in his mind or recorded on index cards. It was there not for its own sake but to aid in de­
cipherment and commentary.

The story of C. Bradford Welles (1901–1969) (Bagnall 2007) follows a radically different
trajectory from Youtie's. Welles was entirely Yale educated; he earned his BA in 1924 and
his PhD in 1928. He had therefore been at Yale for several years before the arrival of
Michael Rostovtzeff, the premier ancient historian of the first half of the twentieth centu­
ry, in 1925. This led to a long and productive professional collaboration and father-son re­
lationship. As a scholar, Welles had begun as a philologist but later moved into epigraphy.
It was under Rostovtzeff's influence that his dissertation was turned into the distin­
guished epigraphical study, Royal Correspondence in the Hellenistic Period (1934).

Rostovtzeff's scholarly appetite was gargantuan. Archaeology and art were for him impor­
tant historical sources. So were documents of all kinds. These needed to be edited, but
mainly in order to serve the larger purposes of history. It was therefore natural for Ros­
tovtzeff and Welles to collaborate on preliminary editions of the more important Dura pa­
pyri (Welles seems to have been self-taught in papyrology), always attended by extensive
discussion and commentary, exhaustively exploring their full historical relevance. Those
students brought to papyrology through Yale have tended to study the Egyptian papyri
with the same historically driven intentions. This was not by chance, for as Welles once
wrote me (September 5, 1969), “I am less interested myself in making papyrologists than
in showing classical students what there is in the papyri, and how it relates to antiquity in
general.”

This tension between papyrology as focused on editing, represented by Youtie, and papy­
rology as concerned with historical expansion, represented by Welles, is inherent in the
discipline—but it is only one aspect of papyrology's history. This is a history that remains
to be written, largely because papyrologists have generally been too busy “doing papyrol­
ogy” to reflect upon their own disciplinary past. There are always pauses for retrospec­
tion during the international congresses, sometimes in the papers given, always in
necrologies delivered during the closing ceremonies. An exception to the rule is I. F.
Fikhman's work on G. F. Zereteli. The (p. 75) extensive publication of correspondence rel­
evant to Italian papyrology (e.g., Morelli and Pintaudi 1983 and recent issues of Analecta

Page 15 of 21
The History of the Discipline

Papyrologica) is also especially noteworthy. But this merely taps an enormous internation­
al reservoir of letters and records. To read and assimilate all of this would be a long-term
project, whose goal would be a book-length study akin to but more immediate and graph­
ic than Preisendanz's worthy, encyclopedic Papyrusforschung und Papyruskunde (1933;
for a different approach see Canfora 2005). As an alternative one might will Marcel
Hombert back into life and in a long afternoon's conversation learn all that needed to be
known.

Acknowledgments
I am grateful to the editor for reasons too numerous to list and to Todd Hickey for advice
on general substance, certain points of fact, and many bibliographical recommendations.
I am also indebted for specific details to I. F. Fikhman, Traianos Gagos, Nikolaos Gonis,
Ludwig Koenen, †Naphtali Lewis, Anthony A. Long, Herwig Maehler, Alain Martin, Fritz
Mitthof, †John F. Oates, and Dirk Obbink. Any lingering mistakes, of course, are mine
alone. Dates and bibliographical notices for many of the people mentioned in this chapter
may be found in Who Was Who in Egyptology, ed. Morris L. Bierbrier (1995), abbreviated
in my text as WWWE. Dates for papyrologists not included in WWWE are conveniently ac­
cessible on the website of the Association Internationale de Papyrologues (http://
www.ulb.ac.be/assoc/aip), especially in its picture gallery. See also Capasso 2007, not
available to me in full when this chapter was being written.

Bibliography
Bagnall, R. S. 1982–1983. “Papyrology and Ptolemaic History.” Classical World 76: 13–21.

———. 1999. Unpublished notes on a “Visit with Naphtali Lewis, Croydon, New Hamp­
shire, 20–22 August 1999.”

———. 2007. “Charles Bradford Welles.” In Hermae: Scholars and Scholarship in Papyrol­
ogy, ed. M. Capasso, 283–286. Pisa: Giardini Editori e Stampatori.

Begg, D. J. I. 1998. “It Was Wonderful, Our Return in the Darkness with … the Baskets of
Papyri! Papyrus Finds at Tebtunis from the Bagnani Archives, 1931–1936.” BASP 35: 185–
210.

Bierbrier, M. L., ed. 1995. Who Was Who in Egyptology, 3d rev. ed. London: Egypt Explo­
ration Society.

Bingen, J. 1977. “La Papyrologie grecque et latine: Problèmes de fond et prob­

(p. 76)

lèmes dʼorganisation.” In Aspects des études classiques: Actes du colloqué associe à la

XVIe Assemblée Générale de la Fédération Internationale des Associations dʼÉtudes Clas­
siques, ed. J. Bingen and G. Cambier, 33–44. Brussels: Éditions de lʼUniversité de Brux­
elles.

Page 16 of 21
The History of the Discipline

———. 1994. “La papyrologie, dʼavant-hier à demain.” In Proceedings of the 20th Interna­
tional Congress of Papyrologists, ed. A. Bülow-Jacobsen, 42–47. Copenhagen: Museum
Tusculanum.

Boak, A. E. R., and H. C. Youtie. 1960. The Archive of Aurelius Isidorus in the Egyptian
Museum, Cairo, and the University of Michigan. Ann Arbor: University of Michigan Press.

Bonacasa, N. 1989. “Cento anni di archeologia italiana per la conoscenza dellʼEgitto gre­
co-romano.” In Egitto e storia antica dallʼellenismo allʼetà araba: Bilancio di un confronto,
ed. L. Criscuolo and G. Geraci, 291–299. Bologna: Cooperativa Libraria Universitaria Ed­
itrice Bologna.

Bowman, A. K., and D. W. Rathbone. 1992. “Cities and Administration in Roman Egypt.”
JRS 82: 107–127.

Canfora, L. 2005. Il papiro di Dongo. Milan: Adelphi Edizioni.

Capasso, M., ed. 2007. Hermae: Scholars and Scholarship in Papyrology. Pisa: Giardini
Editori e Stampatori.

Capart, J. 1947. “Maurice Nahman.” CdÉ 22: 300–301.

Carozzi, P. A. 1982. “Alle origini della Società Italiana per la Ricerca dei Papiri Greci e La­
tini (dal carteggio inedito di Girolamo Vitelli con Uberto Pestalozza, 1898–1908).” Atene e
Roma, n.s. 27: 26–45.

Casini, M., ed. 2001. One Hundred Years in Egypt: Paths of Italian Archaeology. Milan:
Electa.

Chamoux, F. 2002. “Jean Scherer (1911–2001).” Association Amicale de Secours des An­
ciens Élèves de lʼÉcole Normale Supérieure, vol. 1, 59–61.

Criscuolo, L., and G. Geraci, eds. 1989. Egitto e storia antica dallʼellenismo allʼetà araba:
Bilancio di un confronto. Bologna: Cooperativa Libraria Universitaria Editrice Bologna.

DʼArms, J. H. 1980. “Research Professor Emeritus Herbert C. Youtie 1904–1980.” Obitu­

ary entered into the minutes of the Faculty of Michigan's College of Literature, Science
and the Arts, circulated with a cover letter, November 12, 1980.

Deuel, L. 1965. Testaments of Time: The Search for Lost Manuscripts and Records. New
York: Alfred A. Knopf.

Fagan, B. M. 1975. The Rape of the Nile: Tomb Robbers, Tourists, and Archaeologists in
Egypt. New York: Charles Scribner's Sons.

Fikhman, I. F. 1999. “G. F. Cereteli nei fondi archivistici dellʼex Unione Sovietica (Materi­
ali per un ritratto socio-psicologico dello studioso).” Communicazioni dellʼIstituto G. Vitel­
li 1: 1–73.

Page 17 of 21
The History of the Discipline

Gagos, T. 2001. “The University of Michigan Papyrus Collection: Current Trends and Fu­
ture Perspectives.” In Atti del XXII Congresso Internazionale di Papirologia, ed. I. Andorli­
ni, G. Bastianini, M. Manfredi, and G. Menci, vol. 1, 511–537. Florence: Istituto Papiro­
logico G. Vitelli”.

———, J. Gates, and A. Wilburn. 2005. “Material Culture and Texts of Graeco-Roman
Egypt: Creating Context, Debating Meaning.” BASP 42: 171–188.

Geraci, G. 1989. “LʼEgitto romano nella storiografia moderna.” In Egitto e storia antica
dallʼellenismo allʼetà araba: Bilancio di un confronto, ed. L. Criscuolo and G. Geraci, 55–
88. Bologna: Cooperativa Libraria Universitaria Editrice Bologna.

(p. 77) Gerstinger, H. 1932. “Carl Wessely (27-VI-1860–21-XI-1931).” Aegyptus 12: 250–
255.

Gigante, M. 1984. “Per lʼunità della scienza papirologica.” Atti del XVII Congresso Inter­
nazionale di Papirologia, ed. M. Gigante, vol. 1, 5–28. Naples: Centro Internazionale per
lo Studio dei Papiri Ercolanesi.

———. 1986. Girolamo Vitelli e la nuova filologia. Santa Croce del Sannio: Istituto Storico
“Giuseppe M. Galanti”.

Habermann, W. 2001. “Die deutsche Delegation beim Internationalen Papyrologenkon­

greß in Oxford im Jahre 1937 and der für das Jahr 1939 geplante Papyrologenkongreß in
Wien.” APF 47: 102–171.

Heinen, H. 1989. “LʼÉgypte dans lʼhistoriographie moderne du monde hellénistique.” In

Egitto e storia antica dallʼellenismo allʼetà araba: Bilancio di un confronto, ed. L. Criscuo­
lo and G. Geraci, 105–135. Bologna: Cooperativa Libraria Universitaria Editrice Bologna.

Hombert, M. 1933. “Le commerce des papyrus en Égypte.” CdÉ 8: 148–154.

———. 1947. “Lʼétat des études de papyrologie au lendemain de la guerre.” CdÉ 22: 343–
362.

———. 1948. “Lʼétat des études de papyrologie au lendemain de la guerre.” CdÉ 23: 181–
190.

Hopfner, T. 1933. “Carl Wessely. Geboren 27. Juni 1860, gestorben 21. November 1931.”
Jahresbericht über die Fortschritte der klassischen Altertumswissenschaft 241B: 1–24.

Hopkins, C. 1979. The Discovery of Dura-Europos. New Haven: Yale University Press.

Jones, A. H. M. 1942. “Egypt and Rome.” In The Legacy of Egypt, ed. S. R. K. Glanville,
283–299. Oxford: Clarendon Press.

———. 1964. The Later Roman Empire, 284–602. 2 vols. Norman: University of Oklahoma
Press.

Page 18 of 21
The History of the Discipline

———. 1974. The Roman Economy: Studies in Ancient Economic and Administrative His­
tory. Oxford: Blackwell Publishing.

Keenan, J. G. 2002. Review of I. Andorlini et al., eds. Atti del XXII Congresso Inter­
nazionale di Papirologia, Florence. BASP 39: 213–227.

Koenen, L. 2007. “Herbert Chayyim Youtie (1904–1980).” In Hermae: Scholars and Schol­
arship in Papyrology, ed. M. Capasso, 295–305. Pisa: Giardini Editori e Stampatori.

Lewis, N. 1970. “Greco-Roman Egypt: Fact or Fiction?” In Proceedings of the Twelfth In­
ternational Congress of Papyrology, ed. D. Samuel, 3–14. Toronto: Hakkert.

———. 1984. “The Romanity of Roman Egypt: A Growing Consensus.” In Atti del XVII
Congresso Internazionale di Papirologia, ed. M. Gigante, vol. 3, 1077–1084. Naples: Cen­
tro Internazionale per lo Studio dei Papiri Ercolanesi.

Litinas, N. 2007. “Habent sua fata fragmenta: ‘Donum Borgianum’.” In Akten des 23. In­
ternationalen Papyrologenkongresses, ed. B. Palme, 399–405. Vienna: Verlag der Österre­
ichischen Akademie der Wissenschaften.

Martin, A. 2000a. “Das Jahrhundert der Papyrologie? ” APF 46: 1–2.

———. 2000b. “En marge de la Charta Borgiana” CdÉ 75: 118–125.

———. 2007. “The Papyruskartell: The Papyri and the Movement of Antiquities.” In
Oxyrhynchus: A City and Its Texts, ed. A. K. Bowman, R. A. Coles, N. Gonis, D. Obbink,
and P. J. Parsons, 40–49. London: Egypt Exploration Society.

Maspero, G. 1916. Introduction to J. Maspero, Papyrus grecs dʼépoque byzantine, vol. 3.

Reprint, Milan: Cisalpino-La Goliardica and Osnabrück: Otto Zeller, 1973.

Montevecchi, O. 1988. La papirologia, 2d ed. Milan: Vita e Pensiero.

———. 1994. “Problemi e prospettive della papirologia nelle intuizioni di un pioniere:

Amedeo Peyron.” In Proceedings of the 20th International Congress of Papyrologists, ed.
A. Bülow-Jacobsen, 25–34. Copenhagen: Museum Tusculanum Press.

(p. 78) Morelli, D., and R. Pintaudi. 1983. Cinquantʼanni di papirologia in Italia: Carteggi
Breccia-Comparetti-Norsa-Vitelli. 2 vols. Naples: Bibliopolis.

Nicholas, L. H. 1994. The Rape of Europa: The Fate of Europe's Treasures in the Third
Reich and the Second World War. New York: Alfred A. Knopf.

Parsons, P. J. 1987. “Eric Gardner Turner 1911–1983.” Proceedings of the British Acade­
my 73: 685–704.

Pestman, P. W., ed. 1981. Guide to the Zenon Archive. 2 vols. Leiden: E. J. Brill.

Page 19 of 21
The History of the Discipline

Pintaudi, R. 1993. “Documenti per una storia della papirologia in Italia.” Analecta Papyro­
logica 5: 155–181.

Preisendanz, K. 1933. Papyrusfunde und Papyrusforschung. Leipzig: Verlag Karl W. Hi­

ersemann.

Reid, D. M. 2002. Whose Pharaohs? Archaeology, Museums, and Egyptian National Identi­
ty from Napoleon to World War I. Berkeley: University of California Press.

Robinson, J. J. 1901. Review of Otto Gradenwitz, Einführung in die Papyruskunde. Ameri­

can Journal of Philology 22: 210–214.

Rupprecht, H. A. 1989. “Erwin Seidl als Papyrologe.” In Akademischer Gedächtnis für

Professor Dr. Erwin Seidl, 1905–1987, am 9. Dezember 1988, 7–15. Cologne: Verein zur
Förderung der Rechtswissenschaft.

Schubart, W. 1970. “Herstellung schadhafter Texte.” APF 20: 5–14.

Thomas, J. D. 1966. “H. I. Bell (1879–1967).” Aegyptus 66: 97–99.

Turner, E. G. 1973. The Papyrologist at Work. Greek, Roman and Byzantine Monographs,
no. 6, ed. William H. Willis. Durham, N.C.: Duke University Press.

———. 1980. Greek Papyri: An Introduction, 2d ed. New York: Oxford University Press.

———. 1982. “The Graeco-Roman Branch.” In Excavating in Egypt: The Egypt Exploration
Society 1882–1982, ed. T. G. H. James, 161–176. Chicago: University of Chicago Press.

Van Minnen, P. 1993. “The Century of Papyrology (1892–1992).” BASP 30: 5–18.

Vandorpe, K. 1994. “Museum Archaeology or How to Reconstruct Pathyris Archives.” In

Acta Demotica: Acts of the Fifth International Conference for Demotists, Pisa, 4th–8th
September 1993 (Egitto e Vicino Oriente 17), ed. E. Bresciani, 289–300. Pisa: Giardini Ed­
itori e Stampatori.

Welles, C. B. 1934. Royal Correspondence in the Hellenistic Period: A Study in Greek

Epigraphy. New Haven: Yale University Press.

Wilcken, U. 1899. Griechische Ostraka aus Aegypten und Nubien: Ein Beitrag zur antiken
Wirtschaftsgeschichte. 2 vols. Reprint, Amsterdam: A. M. Hakkert, 1970.

———. 1901. “Vorwort.” APF 1: iii–vi.

———. 1912. Grundzüge und Chrestomathie der Papyruskunde. Leipzig: B. G. Teubner.

Wilfong, T. G. 1996. “†Elinor Mullett Husselman.” BASP 33: 5–10.

Youtie, H. C. 1973. Scriptiunculae. 2 vols. Amsterdam: A. M. Hakkert.

Page 20 of 21
The History of the Discipline

———. 1974. The Textual Criticism of Documentary Papyri: Prolegomena. 2d ed. Bulletin
of the Institute of Classical Studies, Supplement no. 33. London: Institute of Classical
Studies.

Notes:

(1.) The “Leiden conventions” can be found in most volumes of papyri, for example, at the
start of each volume of P.Oxy.

James G. Keenan

James G. Keenan, Professor of Classical Studies, Loyola University Chicago.

Page 21 of 21
Conservation of Ancient Papyrus Materials

Conservation of Ancient Papyrus Materials

Jaakko Frösén
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0004

Abstract and Keywords

Papyrus conservation has had a long history since the first attempts to open the car­
bonized Herculaneum rolls found in 1752 and to unroll the Charta Borgiana, acquired in
1778. New techniques have always been invented and tested, and old treatments have
been revised. This article records the methods used on papyrus materials. These methods
include physical control; cellulose treatment; ink fixing; and conserving papyrus, mummy
cartonnages, and carbonized papyri.

Keywords: papyri conservation, carbonized papyri, Charta Borgiana, mummy cartonnages, carbonized Hercula­
neum rolls

Damage and Survival of the Papyri

From the moment that papyrus was used as writing material in antiquity it was subject to
wear and tear at the hands of scribes and readers. Careless handling caused abrasion,
and the edges became worn. In humid surroundings the papyrus suffered from hydrolysis
and oxidation. When stored in unsuitable conditions, the papyrus could rot or grow mold,
and it was sometimes eaten by rodents and/or insects, especially worms and ants This
phenomenon continued and even increased when the papyrus was buried either inten­
tionally or by accident until the climate of the environment became absolutely dry as a re­
sult of the abandonment of habitation sites that were out of the reach of the annual Nile
flood. Most collapsed buildings were afterward used as a dumping ground The papyri
found in ancient habitation sites are thus mostly quite worn and full of holes, and their
surface is abraded or discolored. (For details about the finds see chapter 2, as well as M.
Fackelmann 1985, 14–19. For the damage see, e.g., Donnithorne 1986, 1–4.)

Almost all of the papyri found in Egypt have been exposed to the desert sand, which con­
tains small crystals of calcium, magnesium, and natron salts, which are visible under a
low-powered microscope. These salts absorb moisture and thus (p. 80) eliminate the dete­

Page 1 of 20
Conservation of Ancient Papyrus Materials

riorating effect of the humidity; on the other hand, when crystallized on the surface of the
papyrus, they can cause decay or disintegration.

The best-preserved papyrus rolls or sheets have been found stored in jars, boxes, and
chests in collapsed buildings in dry areas or in tombs in desert cemeteries. Similarly well
preserved are papyrus rolls and sheets, as well as recycled papyri, deposited in desert
tombs even though the funerary papyri lying alongside their owners in wooden mummy
cases have often been prone to extensive staining from the resin or bitumen poured over
the mummy at burial or from fluids from the body. In particular, animal mummies have
from time to time been wrapped in pieces of rolls or sheets of papyrus, and the animal
bodies, particularly the crocodiles, were sometimes stuffed with papyri. These papyri are
equally well preserved, even if mostly cut or torn into pieces and crumpled for stuffing.
The recycled papyri of the mummy cartonnages found in the cemeteries are described lat­
er in this chapter.

Quite well preserved are papyri found in collapsed buildings filled by their own debris
and windblown sand (a soil low in organic material) in deserted town sites, temples, and
monasteries that have not yet been soaked by the rising groundwater table and never
been filled up with garbage. Those papyri were often deposited and buried in jars and
boxes or between building stones. If they were found on the floor, they were mostly ig­
nored as worthless since they were often already in a poor state of preservation when the
building was abandoned.

In much worse condition are papyri found in ancient trash piles either in separate heaps
or in collapsed buildings, where the objects were deliberately thrown away in antiquity.
They are rarely whole scrolls and often only single leaves or fragments of leaves originat­
ing from a cleaning out of some local, often personal, dossier or library. Unfortunately,
they were first torn to pieces and then broken, crushed, or crumpled when being thrown
away. In the trash piles they were also exposed to other debris, liquids, insects, and mi­
croorganisms, partly rotten, moldy, very dusty, and torn. Sometimes baskets full of dis­
carded papyrus have been found in excavations.

Conservation of Normal Papyrus Material

Physical Control

The purpose of papyrus conservation is twofold. The most important concern is to pre­
vent further decomposition of the papyrus material, that is, to get physical (p. 81) control
over the archaeological objects (for general conservation methods see, e.g., Viñas and
Viñas 1988, 1.1–1.3; for the chemical consistency of the papyrus material see Wiedeman
and Bayer 1983). Once taken from the formerly safe surroundings guaranteed by the ex­
tremely dry sand of the desert, the organic material immediately starts to oxidize and de­
compose. However, more often the objective is to retrieve the ancient texts written on the
papyri for study and publication or for exhibition. Some 80–90 percent of all surviving pa­

Page 2 of 20
Conservation of Ancient Papyrus Materials

pyri consist of the so-called normal papyrus material, which can be subjected to a simple,
neither time-consuming nor expensive, conservation treatment.

Papyri from Excavations and Dealers

In most cases, the conservation of papyri recently found in excavations or stored after­
ward in the papyrus collections of museums and libraries is quite a simple action that
does not call for special instruments or complicated chemicals. What is required is a
steady hand, good nerves, and of course practice, which means knowledge of the material
to be conserved and of the few chemicals used in papyrus conservation. Familiarity with
ancient languages and ancient handwritings is certainly useful, too, but it is absolutely
necessary only in special cases. The same is true for the papyri acquired by purchase. In
general, texts that are obtained from uncontrolled illegal excavations have already been
somehow conserved (or at least flattened), mostly by nonexperts, in order to give them a
more attractive appearance (for more details see the section titled “Old Repairs” in this
chapter).

In the first steps of restoration, reassembling, and arrangement of the fragments, one can
use several aids that the papyrus itself offers, namely the continuity of the writing (on
both front and back), the shape of the handwriting, the color of the material, the design of
the fragments, the wormholes that reappear at successive intervals and in similar shape
in the opened roll or unfolded papyrus leaf, the folds of the leaf, the pattern of the criss­
crossed fibers, and the original joins and overlaps (kollêseis) of the papyrus. Further
steps of restoration are possible only for a papyrologist who is able to read the handwrit­
ing and interpret the contents, which is part of the editing process.

Preparation

As in all conservation work, the first thing to do is to remove all of the loose soil and dirt.
The next step is the documentation, both written and photographic. Every fragment
should be inspected for damage. More often than not, the papyrus has been folded for
carrying and storing, rolled in antiquity, or intentionally crumpled; it may also have be­
come creased and distorted by another object, decayed, and eaten (p. 82) by worms or mi­
croorganisms as part of the archaeological stratum. The more or less three-dimensional
dirty object has to be opened, flattened, and cleaned. It has to be softened by dampening
in order to make it pliable (figures 2.2, 2.3 and 2.6, 2.7).

The Ink

First the ink has to be tested. There are two types of ink (as well as a mixture of those
two) used in the papyrus texts. The typical one, carbon ink, is a mixture of soot and gum
arabic. Very resistant and not soluble in water, it does not bleach when exposed to sun­
light. The second, called iron-gall or ferro-gallic ink, has a metallic base. Originally black,
it usually ages to a brownish color. It suffers from exposure to light but is not water solu­
ble. A mixture of carbon ink and iron-gall ink is not stable and can be water soluble. Pa­

Page 3 of 20
Conservation of Ancient Papyrus Materials

pyrus texts written with both iron-gall and the mixed ink, which fortunately occurred only
infrequently in antiquity, must be handled with great care during the conservation
process and must be stored in the dark. (See also, e.g., Cockle 1983, 150; M. Fackelmann
1985, 28–29; Viñas and Viñas 1988, 2.1–2.2.)

Dampening and Treatment

There are many dampening methods available. One can use a humidor, a moisture cham­
ber (for details see, e.g., Lau-Lamb 2005, 6), a fine sprayer to sprinkle the water, or a wet
piece of cloth positioned under and around the object. One can also place it between
sheets of dampened blotting paper. The object is left in the humid surroundings under a
transparent cover for up to twenty-four hours (or more) until the dry papyrus has ab­
sorbed enough water and become flexible but not totally wet. Then the moistened object
is moved to clean, dry blotting paper, unrolled or unfolded and straightened with sharp-
pointed tweezers, and finally sandwiched between two sheets of blotting paper. The parts
that are folded under should be straightened. Some of the distorted and twisted fibers
can be unbent and fixed in their right places by using tweezers and a thin paintbrush
moistened in water. However, in order to avoid rubbing off either the ink or the thin sur­
face of the damp papyrus, do not try to brush away the dirt at this stage. All brushing
should be done when the papyrus is dry. The papyrus should be allowed to dry out and
flatten under a light weight or in a clamp press; it can also be mounted between two
sheets of glass secured with tight metal or plastic clips (M. Fackelmann 1985, 40–46).

Cellulose Treatment

A solution of cellulose (Klucel or hydroxypropyl celluloid) can be used to dampen the pa­
pyrus material. When absorbed by the papyrus material, it strengthens the cell (p. 83) tis­
sue and the vascular bundles and thus increases its elasticity. (For more details see M.
Fackelmann 1981, 659–660; Nielsen and Doblinger 1985, 22–28.) Other materials (e.g.,
papyrus juice, gum arabic) have also been applied to strengthen the papyrus (A. Fackel­
mann 1970, 145; Cockle 1983, 155–156; M. Fackelmann 1985, 20–24, 45).

Cleaning

After a few days the dry papyri are ready for their next treatment. After they are placed
on dry blotting paper, they can be cleaned mechanically by brushing them lightly with a
bristle brush in the direction of the fibers, from the middle toward the edges. The dirt
(e.g., loose mud, dust, grains of sand, salt crystals) can easily be removed from the dry
surface with sharp tweezers, a scalpel, or a spatula. A damp cotton swab or a ball of
fresh-baked white bread also works well (Cockle 1983, 156). The delaminated and twisted
fibers can be straightened and aligned in their correct position by using tweezers and two
thin paintbrushes moistened in water. If needed, methyl cellulose or a flour-and-water
paste can be used. The damp papyrus should be allowed to dry under a light weight. This
treatment should be repeated until the papyrus is fully conserved. The blotters must be

Page 4 of 20
Conservation of Ancient Papyrus Materials

changed frequently. Both sides of the papyrus sheet should be cleaned with care even if
no text is visible on the back.

Some of the literary papyri have been treated in antiquity with cedar oil, probably to
strengthen the surface and prevent bookworms. This ancient fixative should not be re­
moved (cf. Cockle 1983, 157).

Special Cases

In some special cases, the papyri need very painstaking care and scrupulous attention.
The extracting of recycled papyri from the “cardboard” covers of the mummies, the so-
called cartonnages, as well as the conservation of the paintings on those cartonnages, is a
case in point. The method is similar to that of extracting recycled papyri from a bookbind­
ing.

Figure 4.1. Unrolling a pyrolized papyrus scroll

without dampening. The back of the roll has
been partly consolidated by using Japanese pa­
per. Photo by Jaakko Frösén, Egyptian Museum,
Cairo (1987).

Another special case is the opening and the conservation of charred papyrus scrolls.
Sometimes the so-called normal papyri have become brittle and have disintegrated to the
point that dampening would totally destroy the material. Even if they have not been
burned by fire and are not black, material exposed to the effects of a damp environment
for too long will become pyrolized by the oxygen to the extent that it can no longer ab­
sorb moisture like normal papyrus. The natural adhesive of the papyrus has in this case
totally disappeared, and decay has set in. This kind of material has to be treated like car­
bonized papyrus material. This is mostly the case with scrolls of a very fine and thin pa­
pyrus material used, for instance, for the Egyptian Book of the Dead, which have been ex­
posed, after their retrieval from the extra dry tomb or sand, in a museum exhibition. Usu­

Page 5 of 20
Conservation of Ancient Papyrus Materials

ally the (p. 84) humidity and the temperature have not been kept constant for a long peri­
od of time (figure 4.1).

Fixing Ink and Pigments

When the ink or the pigment of the illustrations is flaky, a very thin methyl cellulose or
wheat starch paste (for recipes see, e.g., Lau-Lamb 2005, 15–16) can be applied under­
neath it by means of a fine sable brush; then the pigment is gently tacked down with a
small spatula (M. Fackelmann 1985, 44; Viñ as and Viñas 1988, 2.2–2.3).

(p. 85) Joining Loose Fragments

Loose fragments can be identified as joining others by their form, the arrangement of
their fibers, and the writing. The identification of loose fragments by the writing is, of
course, more effective when the conservator can read the text or works closely with a pa­
pyrologist. During the conservation process, all of the identified joining fragments should
be physically connected and attached to the main papyrus. The joining tape materials
should expand and contract at the same rate as the papyrus itself, and the adhesive
should never permeate the papyrus. The most ideal method is to use loose papyrus fibers
and starch paste or cellulose as an adhesive. Small, finely cut strips of paper or Filmo­
plast on the back of the papyrus also work well. The adhesive, however, should be water
soluble so that mistakes can easily be corrected (M. Fackelmann 1985, 46–55). Strips of
Japanese paper with a water-soluble adhesive form a suitable tape.

Old Repairs

Removing old repairs and tape is the most time-consuming work of the papyrus conserva­
tor. In many cases, the nonexperts have used different kinds of tapes and adhesives “tem­
porarily” and in a hurry. Some of them were attached at the archaeological site or by the
dealers in antiquities in order to fix and secure the fragments or to make them more at­
tractive, but an opportunity to remove them was never provided. The fragments are not
always aligned correctly, and parts of the papyrus may be folded under. Sometimes the
dealers have produced fakes by gluing together fragments of various original texts, thus
achieving “whole leaves” or even fake rolls of papyrus. The oldest types of repairs made
with, for instance, brown tape, stamp paper, or similar materials can easily be removed
when dampened. However, in many cases, the adhesives have permeated the papyrus and
caused severe problems and losses. Scotch tape can very seldom be totally removed. Usu­
ally the tape turns brownish and dry in four to five years, and in some cases the tape car­
rier comes off when the adhesive dries up, but the glue stays, more often than not, in the
papyrus. Chemical solvents (e.g., ethanol diluted with water 50:50) can be used but only
very carefully. Every adhesive has its own solvent, but chemicals that could harm the ink
or the papyrus material cannot be used. Great care must be taken to free the papyrus of
any traces of chemicals afterward (M. Fackelmann, 60–61).

Page 6 of 20
Conservation of Ancient Papyrus Materials

Early Cardboard Backing

Some early conservation treatments are causing problems today. When a papyrus sheet is
backed with cardboard, as was the custom in the nineteenth century, it should be separat­
ed. Most of the backing materials are of an inferior quality, and the (p. 86) rate at which
they expand and contract is different from that of the papyrus materials, especially with
humidity. Often the backing material is made of wood pulp, which is not acid free and
contains many other harmful additives. Together they have already caused fatal and con­
tinuing changes in the otherwise neutral (pH 7) papyrus material. Different kinds of adhe­
sives have also been used. When they are water soluble, they can easily be broken down
even though it may take two to three days to soften the cardboard enough to remove the
adhesive from the back of the papyrus (see, e.g., Donnithorne 1986, 8–9). In other cases,
depending on the glue, boiling water or an enzyme treatment could be used rather than
chemical solvents (cf., e.g., Cockle 1983, 156–157; M. Fackelmann 1985, 68–71).

Mounting and Storing Papyrus

Papyri can be stored in folders between sheets of acid-free blotting paper. If they are very
fragile or very large or are often used by researchers or for exhibitions, the papyri should
be mounted in a frame between glass plates with a backing of acid-free blotting paper.
The blotting paper lessens the changes in the microclimate between the plates and pre­
vents the papyrus fragments from slipping in the frame. Tags with information on the
pieces (plate numbers, inventory numbers, fragment numbers, etc.) can be put under
glass along with the papyrus. Before the final mounting between glass plates, the papyrus
has to dry thoroughly for several weeks. The binding tape should allow a little air to flow
in and out. When airproof binding material is used, airholes must be left for ventilation.
This prevents the growth of mildew and the residue of other microorganisms in the pa­
pyrus. The papyri must be stored in a climate-controlled area where the humidity and
temperature are kept as constant as possible (50–60 percent relative humidity and 17℃–
23℃).

The ultraviolet part of the daylight spectrum can cause reactions in certain materials;
thus, when exposed to daylight, papyrus will fade. A special film or type of glass can be
applied to frames of papyri in exhibitions to help protect them from the harmful rays.

Plexiglas
Plexiglas or similar acrylic plastic material presents several problems as a mounting
medium for papyrus. The most critical danger is the static electricity that develops be­
tween the two sheets. Whenever the papyrus was mounted long ago between two sheets
of Plexiglas, the static electricity must be discharged before they are opened (e.g., by us­
ing an ionizer). Otherwise, one runs the risk of splitting the thin front and back layers of
the papyrus and of the ink sticking to the Plexiglas because the static electricity creates a
harmful electrolysis in the somewhat humid atmosphere between the two plates (Cockle
1983, 155).

Page 7 of 20
Conservation of Ancient Papyrus Materials

(p. 87) The Mummy Cartonnages

Production of Papyrus Cartonnages

Egyptian papyrus cartonnage is a kind of papier-mâché product for protecting the mum­
mies of both humans and animals. They are whole mummy cases or two- or three-dimen­
sional cover plates that were usually put inside a wooden coffin. The cartonnages were
constructed in pharaonic times of layers of linen and in the Ptolemaic period of layers of
linen and/or discarded, used rolls and sheets of papyrus. The waste papyrus was cut or
torn to pieces and dampened with water to fit the piece of cartonnage, layered, and then
stuck together, sometimes with a sort of plaster or glue. The cartonnage was coated on
both sides with a layer of lime gesso plaster. An idealized image representing the de­
ceased was often painted on the front of the cartonnage, as were representations of
Egyptian divinities and ornaments. As a consequence of the dampening both during and
after the production process, the outmost layers on both sides of the cartonnage, particu­
larly the ink of the writings, covered with gesso, have generally been damaged by so-
called calcium burns. In some cases the pigments of the painting have oozed through the
plaster and stained and damaged the papyrus layer or at least the writing. Most disas­
trous have been the metallic colors, especially the green pigments containing copper,
which have eaten through the papyrus material. Papyri removed from the cartonnage
may have a thick layer of plaster on the surface, and because a lot of water has been used
to dissolve the gesso in the process of dismantling the cartonnage, particles of calcium
carbonate can always be seen under a low-powered microscope in all extracted papyri
(Adams 1966; see chapter 2, figure 2.4).

Conservation of Recycled Papyri from Mummy Cartonnages

Since the first discoveries of the papyrus cartonnage and the realization of the potential
of the cartonnage as a source of papyrus texts in the 1820s, the gesso of the cartonnage,
together with the paintings, has been dissolved in a water with vinegar or hydrochloric
acid bath and removed. The papyri were extracted and saved, but the process led to the
complete loss of the painted gesso surface. Even if practiced occasionally, particularly in
private collections even into the 1960s, this method has been abandoned for obvious rea­
sons as unethical.

The new techniques of papyrus restoration, combined with methods of conserving wall
paintings, have made it possible to extract papyrus successfully while safeguarding the
mummy portraits and other paintings made on papyrus cartonnage. The frequently very
bad state of preservation of the cartonnage has (p. 88) been caused partly by the archaeo­
logical surroundings and partly by poor storage conditions later on. A good deal of the
cartonnage material still unconserved in the archaeological depots comes from excava­
tions carried out at the beginning of the twentieth century. When the paintings have been
removed from their papyrus support and fixed on a new, more stable support, the waste

Page 8 of 20
Conservation of Ancient Papyrus Materials

papyri of the gesso cartonnage can also be extracted and conserved (cf., e.g., M. Fackel­
mann 1985, 67–74).

The recovery of papyrus from cartonnage is still the subject of controversy. Admittedly it
interferes with the integrity of the cartonnage as an artifact. However, as the adhesive in
particular that is used in some cardboards is prone to attack by insects and microorgan­
isms, the cartonnages are in many cases in a very poor state of preservation. More often
than not, in order to save the paintings, the papyrus cardboard has to be replaced. (Cf.
the later section titled “Conservation of Book Bindings.” See Donnithorne 1986, 8; Frösén
1997; Horak 1997; and especially Janis 1997.)

The Greek and demotic Egyptian documentary papyrus texts from the Ptolemaic period
are of significant value for investigating the historical outlines of Hellenistic Egypt, as
well as specific topics of economic and social history. The dated papyrus documents also
enable us to date the scattered literary fragments found in the same cartonnage (termi­
nus ad quem) and the cartonnages themselves with their paintings (terminus post quem).

Twenty Steps in Conserving the Paintings and the Papyri

This is a refined version of the method employed by Stohler-Zimmermann and Harrauer

(Stohler-Zimmermann 1981; cf. Wright 1983). It works without complicated chemicals
and without enzymes. The technique is inexpensive and, due to the drying, requires ap­
proximately one week to complete. Nevertheless, during this time several cartonnages
can be handled.

1. Loose dirt is removed by brushing, and the painted surface is cleaned with ace­
tone.
2. Holes in painted gesso or lime plaster are protected (“blocked off”) with calcium
carbonate (figure 4.2).
3. The paint layer is consolidated (“fixed”) by using Paraloid B-48 (30 percent in ace­
tone).

Page 9 of 20
Conservation of Ancient Papyrus Materials

Figure 4.2. Conserving a mummy cartonnage,

steps 1 and 2. The holes in the cleaned surface
have been protected with calcium carbonate.
Photo by Jaakko Frösén, Petrie Museum of
Egyptian Antiquities, London (1985).

4. The paint layer is faced with a new temporary cartonnage by using pieces of (a)
Japanese tissue paper (ca. 50 g) and a Paraloid B-48/acetone emulsion (20 percent)—
one layer; (b) Japanese paper (ca. 70 g) and a Planatol BB/ water dilution (40 per­
cent)—two layers; (c) linen cloth and a Planatol BB/ water dilution (40 percent)—one
layer; and (d) Japanese paper (ca. 70 g) and a Planatol BB/water emulsion (40 per­
cent)—one layer. (p. 89)
5. The object (three-dimensional only) is supported by a negative form by using, for
example, fluid polyurethane.
6. The object in its negative form is placed in water (60℃) for about ten minutes (or
longer if needed).
7. The inner layer of gesso is dampened with water and removed with a brush.
8. The papyrus cartonnage is removed from behind with a scalpel (figure 4.3).
9. The cartonnage is placed for a few minutes in a hot water bath (90℃). The use of
acid in water will cause the gesso to effervesce and loosen the layers of papyrus,
which are sometimes pasted together; afterward, however, they must be deacidified.
10. When wet, papyrus layers are mechanically separated with tweezers and scalpel.
11. Fragments of papyri are dried between sheets of blotting paper.
12. The painting with its gesso layer is consolidated from the back with Paraloid B-48
(20 percent in acetone).
13. The object is backed with a new cartonnage support consisting of Japanese pa­
per (70–90 g), linen cloth, and a Planatol BB/water dilution (40 percent).
14. The negative form of the three-dimensional object is mechanically removed.
15. The temporary facing support is removed with tweezers and brush by using ace­
tone.

Page 10 of 20
Conservation of Ancient Papyrus Materials

Figure 4.3. Conserving a mummy cartonnage,

step 8: removing the papyrus cartonnage from
behind the painted gesso layer. Photo by Tapio
Tyni, Helsinki (1981).

16. The painted surface is cleaned with a brush and acetone (figure 4.4). (p. 90)

17. When dry, papyri are cleaned with a brush.

18. Papyrus fragments are mechanically cleaned (by sandwiching them between
pieces of wet cloth or blotting paper) using a brush, tweezers, and palette knives,
leading to the final removal of the remaining pieces of gesso.
19. Papyrus fragments are dried between sheets of blotting paper under pressure.
20. Papyrus fragments are stored between glass plates.

The Paraloid B-48 is an ethyl acrylate/methyl methacrylate copolymer (made by Rohm and Haas
Ltd., UK). The Planatol BB is a polyvinyl acetate (PVA) (made by Planatolwerk W. Hesselmann,
Germany) that is used, for example, by bookbinders. Instead of acetone, other solvents such as
toluene or ethyl methyl ketone (which are easier and safer to use) can be substituted.

Page 11 of 20
Conservation of Ancient Papyrus Materials

Figure 4.4. The conserved mummy cartonnage.

The painted gesso layer has been backed with a
new cartonnage support. Photo by Jaakko
Frösén, Petrie Museum of Egyptian Antiquities,
London (1985).

In some cases, especially in the earlier cartonnages, various glues have been used in
forming the papyrus support. Depending on the glue, boiling water or an enzyme treat­
ment could be utilized rather than chemical solvents in order to extract the glued papyrus
layers (for more details see, e.g., Cockle 1983, 156–157; M. Fackelmann 1985, 68–71).
(p. 91)

Conservation of Book Bindings

The boards of ancient and medieval book bindings are generally of poor quality as they
were normally covered with leather or cloth. Instead of wooden boards, pasteboard was
frequently used for covering the codices. The production of the pasteboard was similar to
that of papyrus cartonnnages except that more glue was applied in order to make the
product as solid as possible. When recycled literary or documentary papyri were used,
they were cut from discarded remains of rolls; alternatively, whole sheets were glued to­
gether on top of each other so as to resemble cardboard. The covering leather and the an­
imal glue adhesive or the starch paste of the cardboard are prone to attack by insects.
Thus, more often than not, the boards have to be replaced, but in some cases the leather
can be reused for covering the more solid new boards. In extracting the papyri from the
cardboard of the bindings, an enzyme treatment is usually needed (Cockle 1983, 156–157;
cf. Donnithorne 1986, 8). (For the book bindings, see, e.g., Viñas and Viñas 1988, 5.1–5.4.)

Page 12 of 20
Conservation of Ancient Papyrus Materials

Conservation of Carbonized Papyri

The Carbonized Papyri of Herculaneum: An Exception

Carbonized rolls or codices can only rarely be unrolled or opened carefully, page by page.
This kind of exceptional situation prevailed in respect to the carbonized rolls (p. 92) found
in Herculaneum in 1752 (see chapter 13). The building and the room in which the papyri
were stored did not collapse, and the papyrus rolls had been taken out of “bookshelves”
and gathered (possibly for evacuation) in the middle of the floor before the fire broke out.
During the eruption of Vesuvius, the Villa dei Papiri was also covered by a thick layer of
eruption material, a concretelike tuff, which provided an effective obstacle to the subse­
quent penetration of, for example, water, air, and roots into the charred strata, which con­
sisted of organic material. For more than two hundred fifty years, many attempts have
been made to unroll these literary papyrus rolls. All kinds of physical (among them Anto­
nio Piaggio's famous machine), chemical (solvents and gases), and biochemical (enzymes)
procedures have been tested with varying success (A. Fackelmann 1970, 145–147; Cockle
1983, 158; M. Fackelmann 1985, 63–66). For the latest steps in conserving the Hercula­
neum rolls see Capasso (1990, 1991).

The Normal Finds of Charred Papyri

Usually the situation is quite different from that in Herculaneum: The charred papyrus
material has been found in archaeological or illegal excavations in the middle of collapsed
ruins. It has already been affected, immediately during and after the fire and perhaps lat­
er on (repeatedly) by oxygen and water; the papyri have been subjected to the pressure of
stones, sand, and other collapsed materials for quite a long time. Roots of plants have
found their way to the organic charred layers and worked their way into the papyrus
lumps. Thus, the rolls or codices cannot typically be unrolled or opened simply by lifting
the layers one after another. In most cases, the very thin charred layers of papyrus are
stuck and more or less pasted together (ten or even twenty layers within one millimeter)
and sometimes attached to each other by a net of tree and bush roots. The condition of
the fragments of rolls or codices is thus by no means uniform.

Page 13 of 20
Conservation of Ancient Papyrus Materials

The Carbonized Papyri from Egypt and Elsewhere

Figure 4.5. Three carbonized rolls before conser­

vation. The Petra papyri. Photo by Jaakko
Frösén, American Center of Oriental Research,
Amman (1995).

The carbonized papyri of Tanis and Thmouis in the Nile delta, found partly during the ex­
cavations of 1884 and 1902–1904 and between 1892–1893 and 1906, respectively, as well
as partly in the “unofficial” excavations prior to and in between the official ones, fall into
the category of normal finds of charred papyri. This was also the case of the charred liter­
ary roll of Derveni (Greece), found during systematic archaeological excavations in 1963,
as was the case of the carbonized papyri of Boubastos (Nile delta), discovered during ille­
gal excavations in the late 1950s or early 1960s and removed without any scientific con­
trol or documentation. The carbonized papyri of Petra (Jordan), which were exposed and
removed with painstaking care, control, and documentation during the systematic archae­
ological excavations of the Byzantine church in the city center in December 1993 were al­
so normal charred papyrus finds (figures 4.5–4.6). (p. 93)

Page 14 of 20
Conservation of Ancient Papyrus Materials

The Idea of Conservation

Figure 4.6. Section of a loosely rolled Petra pa­

pyrus. Photo by R. Henry Cowherd, American
Center of Oriental Research, Amman (1995).

Figure 4.7. An assembled plate of Petra papyri.

Digital image by Steven Booras and Gene Ware,
American Center of Oriental Research, Amman
(1998). Prepared for publication by Antti Nur­
minen and Vesa Vahtikari (2006).

The principal idea of the conservation of carbonized papyri is to stop the harmful chemi­
cal process from continuing to destroy the delicate organic material: the burning process
—the pyrolysis—as well as the deterioration caused by physical contact, pressure, and
tremor. The conservation method applied to save the (p. 94) carbonized papyri is, in prin­
Page 15 of 20
Conservation of Ancient Papyrus Materials

ciple, the same as that used by Daninos Pasha at the end of the nineteenth century for
conserving a small part of the Thmouis papyri. It was also employed by U. and D. Hage­
dorn, together with L. Koenen, in 1961 when opening two lumps from the Boubastos
archive in Cologne. In addition, it was applied and described by Anton Fackelmann (Pa­
pyrus Collection of the Austrian National Library) in the 1960s for dismounting some of
the Herculaneum papyri in Naples and the Derveni scroll in Thessaloniki and tested by
him on the Boubastos (p. 95) scrolls in Vienna (A. Fackelmann 1970, 145–147). It was used
(with some amendments by Michael Fackelmann [1981, 1985] after 1970 and by J.
Kampichler [1984] from 1978 to 1983) on the Boubastos papyri in Vienna. I myself used it
on the Boubastos papyri in Vienna and Cologne from 1982 to 1985, and my team applied
it to the Petra papyri in Amman in 1994–1995 (Lehtinen 2002). The successful treatment
of the material by this method demands special knowledge of the carbonized (p. 96) mate­
rial and extensive practical experience. Moreover, a knowledge of ancient languages and
ancient handwritings is indispensable for the restoration work.

The Method

Separating the Layers

One begins by mechanically separating first the scroll fragments and then the layers of
these fragments one by one with a thin palette knife, scraper, or other tool and lifting
them with tweezers. It is essential to record with care the order of the pieces. To facili­
tate the loosening process, it sometimes helps to dampen and warm the object, thus caus­
ing the papyrus to curl layer by layer. Often one's breath is enough (“just talk to them”);
otherwise, one may carefully introduce some humidity with a thin paintbrush or by
putting the object under a light bulb (approximately 40℃) or both. This is Anton
Fackelmann's method, but it is not a question of static electricity or of “the electromag­
netic action of the bulb” as stated by him (A. Fackelmann 1970, 145–146). Nevertheless,
when working on large fragments, one usually cannot separate and lift the whole layer at
once; this must be done piece by piece. In a later phase it is quite easy to put the pieces
together to form a whole layer. If the layer is very fragile and does not react to slight
moisture and heat, it can be fixed from the back by using a dilution of water and glue
and, if needed, small pieces of Japanese paper (see the second step). Thereafter, one can
loosen and lift the layer with the aid of the paper. This is, of course, possible only when
the back is blank and when working on the outside of the object.

Fixing the Separated Layers

To provide some elasticity, one has to strengthen the fragile layers. The pieces have to be
put together according to the writing on the front in order to reconstruct the whole layer.
Then the layer is turned over in order to stabilize the entire construction from the back,
which is usually blank. (If the back of the papyrus also contains writing, the pieces cannot
be consolidated at all—sometimes only very carefully between the inscribed lines.) The
reconstruction of the layer is best made on a piece of acid-free Japanese (or Indian) tissue
paper (9–11 g) somewhat larger than the papyrus layer and placed on a sheet of glass. Af­
ter that, one puts a piece of wax paper (or comparable material, e.g., plastic paper) over

Page 16 of 20
Conservation of Ancient Papyrus Materials

the whole reconstruction and then covers it with a glass plate. Sandwiched in this man­
ner, the whole thing is easy to turn over.

The carbonized layer can be strengthened by carefully applying the diluted adhesive
through the Japanese paper from the back with a thin paintbrush and letting the glass
plate slide aside, step by step. The whole back must be glued (avoiding all air bubbles).
The glue must not be allowed to seep deep into the charred material. The glue will dry to
a thin clear film, thereby attaching the piece of papyrus to the Japanese paper and thus
consolidating and protecting it. If the (p. 97) adhesive permeates the thin papyrus layer, it
will form a reflective film surface on the front of the papyrus; this will eliminate the con­
trast between the dull black ink and the shiny black background and make the text invisi­
ble. For the glue, one can use cellulose or a dilution of neutral (pH 7) polyvinyl acetate
glue and water (e.g., Planatol BB—1 part glue to 8–10 parts water, depending on the con­
sistency of the adhesive). Then the reinforced piece of papyrus can be turned over by
grasping the Japanese paper; one may make the indispensable notes in the blank corner
of the paper. The layers must be sandwiched between two pieces of wax paper, and blot­
ting paper or pasteboard under light pressure, letting them dry for a couple of days.

The Two-Dimensional Reconstruction or Jigsaw Puzzle

The two-dimensional reconstruction of the opened papyrus scroll—according to the writ­
ten text of the papyrus, one's own notes, and the numbering of the fragments—must be
made on acid-free paper. To facilitate the subsequent photography, gray acid-free paper
should be used rather than white. When putting the fragments together, the superfluous
margins of the Japanese paper must be removed with sharp scissors. The fragments of
the opened layers can be glued together one after the other from the back (i.e., from the
back of the Japanese paper glued on the back of the charred layers) with thin strips of an
archive-safe Scotch tape or Filmoplast. Papyrus fragments that have writing on both sides
cannot be consolidated or glued together.

Mounting between Glass Plates

The next step is to put the reconstructed sheet of charred papyrus on the acid-free paper
and mount it in a frame of two glass plates. Plexiglas should be avoided not only because
the elasticity and the static electricity it produces could make it very difficult to keep the
fragments in their correct positions but also because the static electricity will create a
harmful electrolysis in the somewhat humid atmosphere between the two plates; this in
turn will destroy the charred papyrus layers. Along the edges, the plates may be sealed
with tight metal or plastic clips or long-lasting, acid-free frame sealing tape, which ad­
heres well to glass. The corners of the plates must be left unsealed for ventilation and to
prevent an injurious and fungi-fertile microclimate from developing between the plates.
Tags with information on the pieces can be put under glass along with the papyri. The
plates can be used as double sided.

Page 17 of 20
Conservation of Ancient Papyrus Materials

Storage
The plates of carbonized papyri should be kept, when not on exhibition or in working
rooms, in a place where the climate can be kept as constant as possible (17–23℃ and 50–
60 percent relative humidity) since every change in temperature or humidity will create
(p. 98) a movement in the charred material. The plates should be kept horizontally on cab­

inet shelves, not more than three plates in each stack and acid-free paper or pasteboard
between the plates in order to avoid excessive pressure and scratches on the glass. Un­
painted and unvarnished wooden cabinets and shelves are preferable because of the
more constant conditions they provide for the conservation of organic material.

The Photographic Documentation

For documentation, the plates must be photographed by using special methods of black-
and-white overexposure. An approximately two-thousand-watt light (preferably incandes­
cent bulbs with a reddish light) is needed. It should come from three or four points at an
angle of 45° to avoid reflection and be dispersed as evenly as possible over the whole ob­
ject. When the plates are exposed to the lights, the heat will create a strong movement in
the pieces and cause the carbonized papyrus fragments, which are glued to Japanese pa­
per, to curl drastically. To avoid the curling and ensure stability, the fragments must be
photographed under the sheet of glass. This increases the reflection of the lights and also
of the ceiling, the camera, and other surrounding objects. In addition, the glass decreases
the effectiveness of the infrared photographs. By using a high-contrast film for black-and-
white prints (e.g., Kodak TechPan 25 ISO), one can use an exposure time of one second
with an aperture of 11. These are, of course, only rough approximations. However, the
best results can be achieved by using the multispectral digital technique (figure 4.7).

Papyrus Conservation: Past and Present

Papyrus conservation has a long history since the first attempts to open the carbonized
Herculaneum rolls found in 1752 and to unroll the Charta Borgiana, acquired in 1778.
New techniques have always been invented and tested, and old treatments have been re­
vised. (For a short history of papyrus conservation and for old techniques see Cockle
1983, 153–158; Donnithorne 1986, 5–6; Parkinson and Quirke 1995, 74–83.) In this chap­
ter I have recorded the methods I have used on papyrus materials. The sources in the bib­
liography discuss additional techniques and treatment of other materials. They will also
lead to other procedures.

Although there is no systematic training program for papyrus conservators, various train­
ing programs in paper conservation provide some instruction. Otherwise, those who are
interested in and want to take charge of treating the papyrus material in various collec­
tions share their experiences and knowledge both privately and at international confer­
ences and thus build a common knowledge of (p. 99) best practices. The papyrus collec­
tion at the Austrian National Library in Vienna has played an important role in this
process.

Page 18 of 20
Conservation of Ancient Papyrus Materials

Bibliography
Adams, C. V. A. 1966. “The Manufacture of Ancient Egyptian Cartonnage Cases.” Smith­
sonian Journal of History 1: 55–66.

Bierbrier, Morris L., ed. 1986. Papyrus: Structure and Usage. British Museum Occasional
Paper 60. London: British Museum.

Capasso, M. 1990. “Problemi di conservazione, restauro e svolgimento di papyri carboniz­

zati.” Rudiae 2: 41–49.

———. 1991. “Problemi di conservazione, restauro e svolgimento di papiri carbonizzati

II.” Rudiae 3: 25–29.

Cockle, W. E. H. 1983. “Restoring and Conserving Papyri.” Bulletin of the Institute of

Classical Studies 30: 147–165.

Donnithorne, A. 1986. “The Conservation of Papyrus in the British Museum.” In Papyrus:

Structure and Usage, ed. M. L. Bierbrier, 1–23. British Museum Occasional Paper 60. Lon­
don: British Museum.

Fackelmann, A. 1970. “The Restoration of the Herculaneum Papyri and Other Recent
Finds.” Bulletin of the Institute of Classical Studies 17: 144–147.

Fackelmann, M. 1981. “Neuerungen in der Papyrusrestaurierung.” In Proceedings of the

XVI International Congress of Papyrology, ed. R. S. Bagnall, G. M. Browne, A. E. Hanson,
and L. Koenen, 657–663. American Studies in Papyrology 23. Chico, Calif.: Scholars
Press.

———. 1985. Restaurierung von Papyrus und anderen Schriftträgern aus Ägypten. Studia
Amstelodamensia 23. Zutphen, the Netherlands: Terra.

Frösén, J. 1997. “Der Wert des Kontextes fur die Deutung der Kartonage-Papyri.” Akten
des 21. Internationalen Papyrologenkongresses, Berlin 1995, ed. B. Kramer et al., vol 2:
1079–1082. Stuttgart: Teubner.

Harrauer, H., ed. 1985. Bericht über das 1. Wiener Symposion für Papyrusrestaurierung
4.–8.6.1984. Mitteilungen aus der Papyrussammlung der Österreichischen Nationalbi-
bliothek NF 19 (with an extensive bibliography by M. Doblinger, “Literaturübersicht zur
Papyrusrestaurierung,” 63–76). Vienna: Holzhausen.

Horak, U. 1997. “Die Bedeutung der Malerei auf Papyruskartonage aus ptolemäischer
und augusteischer Zeit für die antike Ikonographie und für das Verständnis einer antiken
Kunstindustrie.” Akten des 21. Internationalen Papyrologenkongresses, Berlin 1995, ed.
B. Kramer, W. Luppe, and H. Maehler, 1091–1096. Stuttgart: Teubner.

Janis, K. 1997. “Die Bearbeitung eines ptolemaaischen Mumienpektorals im Inter­

essenkon-flikt zwischen Papyrologe und Restaurator.” Göttinger Miszellen 161: 87–95.

Page 19 of 20
Conservation of Ancient Papyrus Materials

Kampichler, J. 1985. “Konservierung verkohlter Papyri.” In Bericht über das 1. Wiener

Symposion für Papyrusrestaurierung 4.–8.6.1984, ed. H. Harrauer, 35–36. Vienna:
Holzhausen.

Lau-Lamb, L. 2005. Guidelines for Conservation of Papyrus. Advanced Papyrological In­

formation System (APIS). http://www.lib.umich.edu/pap/conservation/
guidelines.html.

Lehtinen, M. 2002. “Conservation and Reconstruction.” The Petra Papyri I, ed. J.

(p. 100)

Frösén, A. Arjava, and M. Lehtinen, 11–16. Amman, Jordan: American Center of Oriental
Research.

Nielsen, I., and M. Doblinger. 1985. “Hydroxypropylcellulose zur Festigung von Papyrus.”
In Bericht über das 1. Wiener Symposion für Papyrusrestaurierung 4.–8.6.1984, ed. H.
Harrauer, 22–28. Vienna: Holzhausen.

Parkinson, R., and S. Quirke (with contributions by U. Wartenberg and B. Leach). 1995.
Papyrus. Austin: University of Texas Press.

Stohler-Zimmermann, A. 1981. “Das Ablösen der Malerei von Mumienkartonagen.” In

Proceedings of the XVI International Congress of Papyrology, ed. R. S. Bagnall, G. M.
Browne, A. E. Hanson, and L. Koenen, 665–676. American Studies in Papyrology 23.
Chico, Calif.: Scholars Press.

Viñas, V., and R. Viñas. 1988. Traditional Restoration Techniques: A RAMP Study (original
in Spanish). General Information Programme and UNISIST. UNESCO, Paris (PGI-88/ WS/
17). http://www.unesco.org/webworld/ramp/html/r8817e/r8817e.jpg.

Wiedemann, H. G., and G. Bayer. 1983. “Papyrus: The Paper of Ancient Egypt.” Analytical
Chemistry 55: 1220–1230A.

Wright, M. M. 1983. “A Method of Extracting Papyri from Cartonnage.” Studies in Conser­

vation 28: 122–126.

Jaakko Frösén

Jaakko Frösén, Professor of Greek Philology, University of Helsinki.

Page 20 of 20
Greek and Latin Writing in the Papyri

Greek and Latin Writing in the Papyri

Guglielmo Cavallo
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0005

Abstract and Keywords

This article briefly sets forth some methodological considerations in the history of Greek
and Latin writing in the papyri. The writing exercises attested in papyri, ostraca, and
tablets offer many examples of training at various levels. The distinction that emerges
from them is thus not between documentary and literary hands but between cursive and
semicursive writing styles, in which the greater or lesser velocity of the ductus modifies
the traces and forms of the letters, and regular or rather calligraphic handwritings. Start­
ing with the fourth century, manuscripts preserved in libraries begin to be available
alongside those found in archaeological excavations. Paleographical evidence can emerge
from the comparison of dated or datable documentary writing and undated literary
hands.

Keywords: Greek writing, Latin writing, cursive styles, calligraphic handwritings, paleographic evidence

IT is difficult—both methodologically and in choice of material—to trace concisely the his­

tory of Greek and Latin writing in the papyri over many centuries, especially because it is
impossible to discuss and defend adequately the dates proposed, the criteria for assess­
ing handwritings, and one's acceptance or rejection of the views of other scholars. In­
stead, I first briefly set forth some methodological considerations.1

In the area of method, I have sought to avoid too drastic a distinction between documen­
tary and literary hands, for handwriting is a unitary phenomenon, the various manifesta­
tions of which need to be assessed not so much on the basis of the use to which they are
put—documents or books—as by a consideration of their development. The basic patterns
of the letters always constitute the point of departure, patterns taught first at the elemen­
tary level and then, through successive exercises, developed either toward greater rapidi­
ty and cursiveness or toward calligraphic deliberateness. The writing exercises attested
in papyri, ostraca, and tablets offer many examples of training at various levels.2 The dis­
tinction that emerges from them is thus not between documentary and literary hands but
between (1) cursive and semicursive writing styles, in which the greater or lesser velocity

Page 1 of 37
Greek and Latin Writing in the Papyri

of the ductus modifies the traces and forms of the letters, and (2) regular or rather calli­
graphic handwritings, which remain closer to the original graphic structure of the letters.

A discussion limited to the Greek and Latin handwriting of the papyri (including
(p. 102)

parchment fragments) also suffers methodologically from being unable to take into ac­
count all of the graphic manifestations on other supports (not only papyrus and parch­
ment but also tablets, ostraca, stone, marble, plaster, metal, and other materials). Nor
can one ignore the fact that, starting with the fourth century, manuscripts preserved in li­
braries begin to be available alongside those found in archaeological excavations.

The choice of examples from the nonetheless vast range of material offered by the papyri
has involved balancing two constraints: The first is the need to examine significant
pieces; the second, that of selecting readily accessible material. Taking account of these
two limitations, the choice of documents has fallen especially on dated or datable items,
while literary materials, which generally lack explicit indications of date, have been cho­
sen on the basis of various other criteria—archaeological, papyrological, and palaeo­
graphical. One type of archaeological criterion is provided by the discovery of mummy
cartonnage in which both documents and books were used as raw material. Another in­
stance, particularly fortunate because it offers not only archaeological but also textual ev­
idence, is provided by the discoveries at Herculaneum. Because the majority of these con­
tain works of Philodemus of Gadara, they must of necessity be dated in the middle of the
first century BCE and in any case before 79 CE, the year of the catastrophic eruption of
Mount Vesuvius. Papyrological evidence may come from considering the criterion of rec­
to/verso in cases where a roll was written on both sides, one bearing a document, the oth­
er a literary work; if the document is dated or approximately datable, the literary hand
will be older than the document if on the recto, later if on the verso.3 More directly,
palaeographical evidence can emerge from the comparison of dated or datable documen­
tary writing and undated literary hands. In the absence of any other criterion for dating,
only a palaeographical assessment remains.

The Origins of Greek Handwriting

Familiarity with writing and the ability to read and write go back in the Greek world to a
period well before that to which we can assign the oldest documentary and literary papyri
known so far, which are not older than the second half of the fourth century BCE and the
opening decades of the third.4 In these, we can identify some characteristics of Greek
writing: alpha with horizontal bar; epsilon and sigma of an archaic type, with straight
strokes; kappa with short oblique lines; and omega with a convex central curve. We en­
counter these characteristics in SB XIV 11942, a military order from the years between
331 and 323, which displays, all the same, a highly rigid (p. 103) writing, which can hardly
be considered a typical hand of its day. A more normal hand appears in late fourth-centu­
ry accounts found in the same necropolis at Saqqara (SB XIV 11963), written in two
hands. A handwriting with a faster and more fluid ductus appears in the marriage con­
tract P.Eleph. 1 (310 BCE), the oldest dated papyrus, and at the start of the third century

Page 2 of 37
Greek and Latin Writing in the Papyri

in the contracts P.Eleph. 2 (285/284) and 3 (284/283). In these, besides a slight inclination
of the letters to the right, one can even observe some forms that appear later, like alpha
with an oblique cross stroke, beta with flattened curves, large mu, and nu with a raised
middle stroke.

The same graphic characteristics and evolutionary phases reappear in the literary papyri
assignable to the last decades of the fourth century or the first decade of the third. In
P.Derveni, which is interesting because it comes from near Salonika, in Greece, and is ar­
chaeologically datable to around 340–320, we see alpha with horizontal crossbar, archaic
epsilon and sigma, and omega with convex central curve. We can also see zeta with paral­
lel outer horizontal strokes and vertical middle stroke and theta with its central element
reduced to a point. Other examples of this early type of writing in literary papyri are
P.Berol. inv. 9875, the Persians of Timotheos, found at Abousir, the so-called curse of
Artemisia in UPZI 1, P.Hib. I 6, and BKTV 2, 56–63, which contain scholia and elegies, the
writing of which shows, like the documents of Elephantine, the transition from fourth-cen­
tury to third-century forms. In the writing of the last of these, a constant characteristic of
third-century scripts (especially in literary rolls) is visible: the contrasting sizes of large
letters (alpha, eta, mu, tau, and sometimes nu and pi) and other narrower or smaller let­
ters (epsilon, theta, omicron, sigma).

In the oldest examples, then, no substantial differences are visible between documentary
and literary Greek hands or between papyrological and epigraphic hands. Perhaps a basic
script had been taught at the elementary level as early as the end of the fifth century and
was intended to be used for general purposes. Beginning around the fourth to third cen­
tury, however, as the Elephantine papyri show, a process of differentiation began that,
from around 275 BCE on, led to increasingly different styles of execution. The rigid writ­
ing of the old style remained in use, however, as we can see both from certain documents
like the letter P.Lille I 17 (probably Ptolemy II Philadelphos) or literary papyri like P.Petr.
II 50 and P.Petr. I 5–8, which contain, respectively, the Laches and the Phaedo of Plato. In
documents dated a bit later we observe graphic forms becoming ever faster, as is evident
in P.Rev. (259/258) and P.Hamb. II 187 (246/245).

Figure 5.1. PSI IV 325. Photo courtesy of the Bib­

lioteca Medicea Laurenziana.

Page 3 of 37
Greek and Latin Writing in the Papyri

A rather stylized script, called “Alexandrian chancery,” is found in a series of documents

from the Zenon archive produced in Alexandria, especially letters from the dioikêtês Apol­
lonios to Zenon (P.Lond. VII 1973 [254] and PSI V 514 [251]) or circular letters from Apol­
lonios to officials (e.g., PSI IV 324 and 325 [261] [figure 5.1] in the same hand). The most
obvious characteristic of this script is the placement of letters along a guideline that de­
termines their shape, so that the horizontal strokes of gamma, epsilon, pi, and tau are ex­
tended to touch the following letter, forming a kind of pseudo-ligature, with oblique
strokes descending from the left of (p. 104) (p. 105) alpha, delta, lambda, and nu, and the
middle strokes of mu transformed as far as possible into horizontal strokes also aligned
with the guideline. Curved strokes, like the bowl of upsilon and the second half of omega,
undergo a similar transformation, and beta, iota, omicron, sigma, and rho, reduced in
size, look as if hung from the horizontal guideline.5

Less formal chancery scripts are attested in documents written in Egypt, Caria, and
Palestine, all of which are from the Zenon archive: PSI VI 616, PSI V 518, and P.Cair.Zen. I
59037.6 In addition, the same Alexandrian chancery writing, in a less rigid and stylized
form, was likely adopted in literary papyri, as in the two funerary epigrams commissioned
by Zenon for his hunting dog, Tauron (P.Cair.Zen. IV 59532).

Developed Ptolemaic Style

The further lines of evolution of Greek writing from the middle of the third century to the
early second century may be followed in hands that, although not completely devoid of el­
ements drawn from the Alexandrian chancery style, differ from it in lacking the relentless
horizontal linear drive of the writing as a whole, with individual letters taking on more
regular proportions and softer strokes. The ductus is sometimes slower and more calli­
graphic, sometimes faster and informal. Examples include PSI IV 341 and P.Lond. VII
2011, with more rounded forms visible in PSI IV 383 (figure 5.2). This papyrus allows us
to see characteristics of the period, such as a contrast in size between broader and nar­
rower letters, alpha, lambda, and delta with a slightly concave oblique stroke descending
from left to right, and mu and pi with their right hasta (i.e. the vertical stroke) trans­
formed into a curved line; however, it shows no trace of stylization along the lines of the
chancery style. Instead, small ornamental apices appear at the ends of right-hand strokes
in some letters (mu, nu, omega), a habit that would later be extended to other letters. In
the late third century we also find writing in an older style, as in P.Eleph. 17, P.Hamb. I
24, and SB XXII 15762, all from 223–210 BCE.

Page 4 of 37
Greek and Latin Writing in the Papyri

Figure 5.2. PSI IV 383. Photo courtesy of the Bib­

lioteca Medicea Laurenziana.

The same characteristics may be found with minor variations in the hands of literary pa­
pyri, beginning in the middle of the third century and lasting through the end of the cen­
tury. Sometimes, however, the ductus is slower, and the forms more calligraphic and
rounded. Good examples for these central decades are the pseudo-Epicharmus (P.Hib. I
1), the Hippolytus of Euripides (P.Lond.Lit. 73), the Odyssey fragment PSI VIII 979, Thucy­
dides in P.Hamb. II 163, and the mathematical exercises PSI VII 763, this last showing
similarities particularly with P.Lond. VII 2011 (244). The tetrameters of Archilochus in
P.Lond.Lit. 54 and Menander's Sicyonius in P.Sorb. inv. 2772b, which show a letter axis
slightly inclined to the right, may be dated a little later. The book hands of the later third
century can also be represented by the (p. 106) Posidippus papyrus P.Vogl.Mil. VIII 307
(figure 5.3), which is associated with documents from the period 222–213,7 the Odyssey
papyrus P.Sorb. inv. 2245, and the tragic fragment PSI II 136.

The transition from the third to the second century brought no radical change in Greek
writing, but evolution continued in the cursive and semicursive hands. In documents such
as two petitions from the 160s (P.Lond. I 24 recto and P.Lond. I 44), as well as a judicial
decision probably from 135/134 BCE (PSI XIII 1310), the writing recalls the more or less
rigid forms of the third century, but semicursive hands also appear, as in P.Mert. 5 (149–
135), PSI III 166 (118), and P.Bad. II 3 (109) (figure 5.4). At the same time, however, we
witness the birth of the first true cursive writing, visible in documents like P.Amh. 35 (pe­
tition, 132) and BGU III 998 (contract, 101) (figure 5.5) and in receipts on ostraca (BGU
VI 1440 [143/142], VI 1340 [120/119], VI 1341 [104]).

Page 5 of 37
Greek and Latin Writing in the Papyri

Figure 5.3. P.Mil.Vogl. VII 307. Photo courtesy of the

Istituto di Papirologia, Università degli Studi, Milan.

Along with these semicursive and cursive hands flourishes a line of skilled and regular
hands, visible sometimes in the same documentary production and even more widely in
book production. Such hands appear in documents like P.Lond. II 223, P.Köln V 222, and
P.Bad. II 2, ranging from 179/178 (or 168/167) to 130. In the formal handwritings used in
second-century book production, we can distinguish two basic tendencies: There are both
accurate but fluid hands and more elaborate and elegant hands, characterized by more or
less marked decorative apices more at the ends of strokes. We thus witness the birth of
highly calligraphic hands. An example of the first class is the skeptical treatise P.Louvre
E7733 recto (Pack2 2579), (p. 107) while a fluid ductus can be seen in P.Par. 2 (Chrysip­
pus); examples of the style with apices are P.Würzb. 1 (history) and P.Tebt. I 4 (Iliad),
while a particularly artificial style appears in P.Oxy. XV 1790.

Page 6 of 37
Greek and Latin Writing in the Papyri

From Hellenistic to Roman

Figure 5.4. P.Bad. II 3. Photo courtesy of the Institut

für Papyrologie, University of Heidelberg.

Figure 5.5. BGU III 998.

Graphic elements of an archaic type, like the horizontal orientation of strokes, sharp cor­
ners, contrast in size between broad and narrow letters, and the absence of ligatures,
which all tend to disappear around the end of the second century BCE, vanish in the fol­
lowing century, during which we see a transition from Hellenistic handwriting to Roman.
In the documents semicursive and cursive writings dominate, although there is a continu­
ous grade between these two categories rather than (p. 108) (p. 109) a clean distinction. In
the semicursive part of the spectrum we may place P.Lond. III 833 (89/88) and P.Oxy. VII
1061 (22 BCE [figure 5.6]); toward the cursive end are PSI V 549 (41 BCE), P.Ryl. II 73
(33–30) (figure 5.7), and PSI X 1099 (6/5).

In the second half of the first century BCE, particularly from the reign of Augustus, the
process of “cursivization” reaches full maturity. This period opens a phase of changes in
the morphology of the letters, whereby the speed of the ductus brings about reductions

Page 7 of 37
Greek and Latin Writing in the Papyri

and simplifications in the internal strokes of letters, and ligatures change the letters they
join. The letters beta and kappa are a particular locus of change; in some cases these are
written in a single movement and take on the form of two curved hastas joined by a
curved base, rather like the Latin u (for beta, see P.Ryl. II 73 and, for kappa, P.Oxy. VII
1061 [figure 5.6]).8

In Egypt, the cursive that matured in the first century BCE, written more or less rapidly,
continues in the following centuries. In the first century CE we can distinguish two types
of cursive. One displays rounded strokes, characterized in some versions by a certain sin­
uosity; examples are RRyl. II 183a (16 CE) and SB I 3924 (= P.Select. II 221, 19 CE or
shortly after). The other type of cursive shows more headlong forms and a more marked
tendency toward leaning to the right. This is found in documents like P.Ryl. II 131 (31 CE),
P.Ryl. II 119 (54–67 CE), and BGU I 197 (17 CE) (figure 5.8).

Figure 5.6. P.Oxy. VII 1061.

Page 8 of 37
Greek and Latin Writing in the Papyri

Figure 5.7. P.Ryl. II 73.

In the period from the first century BCE to the first CE, the panorama of skilled hands—
those that retain a regular composition of the letters from strokes and sometimes reach a
high level of formality—is more articulated and varied than in earlier centuries. In P.Ryl.
IV 586 + P.Oxy. IV 802 (99 BCE), only a few cursive elements appear. Although less ele­
gant, P.Oxy. XIV 1635 (44–37 BCE) offers a clear, regular hand, (p. 110) (p. 111) as does
BGU IV 1114 (5 BCE). The narrow, headlong forms of PSI X 1160 (after 23 BCE; account
of an Alexandrian embassy to Augustus), by contrast, imitate or are inspired by chancery
forms: The letters are almost entirely detached, made with a rather slow ductus, some­
times with caudate hastas (particularly in kappa and upsilon). But in the Augustan period
(7–4 BCE) we also find in P.Lond. II 354 a petition, a writing that, even if fluid and fluent,
is also rather regular and even calligraphic.

Figure 5.8. BGU I 197.

Page 9 of 37
Greek and Latin Writing in the Papyri

Turning from documents to contemporary literary rolls, we find a rich series of

(p. 112)

skilled and calligraphic hands. These, indeed, in the course of this period not only reach
full maturity but also exhibit a typological variety that allows us to distinguish a whole se­
ries of graphic streams sometimes also visible in documents, thereby enabling us to ar­
rive at better-founded dates.

Figure 5.9. P.Mur. 108. Photo courtesy of the Israel

Antiquities Authority.

A first and rather widespread graphic stream is represented by P.Herc. 1050, with its very
regular hand, using elegantly rounded forms. This type of writing is found in the same
century (the first BCE) not only in the Herculaneum papyri but also in P. LouvreE7733
verso (Pack2 2911) and in the following one in the document P. Oxy. II 246 (66 CE). The
writing of P.Mur. 108 (a fragment perhaps of a philosophical text) (figure 5.9) is also close
to these in type. Another graphic stream, identifiable also from the first century BCE, is
evident in BKT V 2, 131–139 (anapests) (figure 5.10), which can be dated to the first half
of the century, in which we find a round script adorned with marked apices at the ends of
hastas, paralleled in a document from 99 BCE. The same graphic forms are found in a
fragment of Deuteronomy, P.Fouad inv. 266 (Van Haelst 56, mid to late—first century
BCE), and in the pinnacle of contemporary calligraphy, the famous papyrus of the Coma
Berenices of Callimachus, PSI IX 1092. (p. 113)

The continuation of this style in the first century CE is evident in PSI XI 1214
(p. 114)

(Sophron, Mimes) (figure 5.11). A related graphic stream is represented by the Iliad
papyrus P.Fay. 6, the Epicurean fragment P.Heid. inv. 1740 recto, and the Carchedonios of
Menander (P.Oxy. XXXIII 2654); P.Oxy. XXXIII 2654 is assigned to the first half of the first
century CE, and the first two date to its beginning or even the end of the previous centu­
ry. This writing has simple, even rough, forms, mostly devoid of apices; it appears first in
some papyri from Herculaneum (e.g., P.Herc. 1005) and later in the Iliad papyrus
P.Lond.Lit. 6, which displays a rather less formal script, which should be assigned to a

Page 10 of 37
Greek and Latin Writing in the Papyri

date some decades before the reign of Domitian (81–96), when a document was written
on the verso. A calligraphic script, but written with a fluid ductus, appears in a series of
literary papyri connected to a graphic stream; it is also attested in a document from the
end of the first century BCE, P.Lond. II 354, but is widespread above all in the following
century in book production, as P.Lond.Lit. 30 (Odyssey), PSI XII 1285 (Alexandrian ro­
mance), and P.Gen. II 85 (Ninus romance, before 100/101 [verso]) show. Further exam­
ples are PSI IX 1091 (figure 5.12), PSI VIII 978, and P.Berol. inv. 6845, the script of which
has a close parallel in P.Fay. 110, a document from 94 CE.

A graphic stream entirely different from those mentioned so far appears in the Hercula­
neum papyri, for example, P.Herc. 994 +1676 +1677 +1074 +1081, certainly from the
first century BCE, as well as in Egyptian papyri like P.Oxy. XXIII 2359 (Stesichorus?) and
P.Oxy. XXII 2318 (Archilochus). This graphic stream, beginning with the end of the first
century CE and continuing throughout the second, has more rigid forms, even while
sometimes using softer strokes, as the looped alpha, the mu with its middle strokes fused
into a curve, and the omega with more curved lines illustrate. This phase is evident in pa­
pyri like the so-called Harris Homer (P.Lond.Lit. 25), P.Oxy. XVII 2066 (Sappho), P.Oxy. XV
1809 (Plato), and PSI IX 1088 (Isocrates), all of which were written not later than the first
century CE, and in P.Oxy. VIII 1083 (Sophocles), a luxurious roll from the second century.

Figure 5.10. BKT V.2, 131–139.

There are two final, particularly characteristic graphic typologies. The first picks up some
forms from P.Oxy. XV 1790; one of its first exemplary instances is the Hyperides papyrus
P.Lond.Lit. 134 (second century BCE) (figure 5.13), with its elaborate, fairly regular writ­
ing. In the same classification we can place the Philodemian De libertate dicendi (P.Herc.
1471) and the less elegant P.Oxy. IV 659 (Pindar) and P.Oxy. XXXI 2535 (hypomnema; to
be assigned to the first century CE). The other rather characteristic script of this period is
what we may call “epsilon-theta style,” from the marked tendency in these letters to re­
duce the middle element to isolated central points, buttons, apices, or short curved
Page 11 of 37
Greek and Latin Writing in the Papyri

strokes. The flourishing of this style, which may be dated to the second half of the first
century BCE and the beginning of the first century CE, is reflected in papyri like P.Herc.
1044 (life of Philodemus), P.Lond.Lit. 48, and P.Oxy. XXXI 2545 (figure 5.14). (p. 115)

The Imperial Peak

Figure 5.11. PSI XI 1214. Photo courtesy of the Bib­

lioteca Medicea Laurenziana.

Figure 5.12. PSI IX 1091. Photo courtesy of the Bib­

lioteca Medicea Laurenziana.

Page 12 of 37
Greek and Latin Writing in the Papyri

Figure 5.13. P.Lond.Lit. 134.

Figure 5.14. P.Oxy. XXXI 2545. Photo courtesy of the

Egypt Exploration Society.

The period from the first to the third centuries CE represents the peak in numbers of pre­
served papyri, with the pinnacle occurring in the second century. In the most common
cursive scripts the two dominant tendencies observed even in the early Roman period
continue. The cursive of a more rounded type is sometimes uneven and messy, with de­
forming ligatures and inversions of strokes, because of the practiced speed with which
the vast quantity of documents was produced. (p. 116) (p. 117) (p. 118) (p. 119) (p. 120) Al­
though sometimes this writing keeps an upright axis, it increasingly acquires a more or
less strong slant to the right. Good examples of the more upright variety are PSI X 1138
(107), PSI XII 1225 (156/157), and BGU III 807 (185). But more common, from the second

Page 13 of 37
Greek and Latin Writing in the Papyri

century on, is the inclined variant of this cursive, as illustrated in P.Bad. II 22 (126), P.Flor.
I 47 (213–217) (figure 5.15), and P.Ryl. II 117 (269). The same cursive is found also in the
third-century PSI X 1173. The other cursive, which was documented in the first century
but became more widespread in the second and particularly in the third century, is the
more headlong, sharp variety with a more or less marked slant to the right; examples in­
clude BGU II 544 (ca. 138–161), P.Flor. I 23 (145), BGU I 92 (187), CPR I 32 (218) (figure
5.16), and BGU IV 1073 (275).

Many of the handwritings found in papyri outside Egypt in this period are similar to those
of the Egyptian documents. There are, however, some distinctive characteristics, particu­
larly the widespread use of semicursive scripts characterized by very rigid tracing, angu­
larity, and compact strokes, which give the writing a coarse appearance. These character­
istics appear in documents from Palestine and Mesopotamia.9 Palestinian examples in­
clude (from Murabbaʼat) P.Mur. 116 (first half of the second century) and P.Mur. 114
(probably 171) (figure 5.17); the same awkward, heavy scripts are also attested in docu­
ments from Nahal Hever (e.g., P.Yadin 14, 15, 17, and 18 [figure 5.18], all from 125–128
CE). There are also examples from Dura-Europos, like P.Dura 18 (87), 23 (134) (figure
5.19), and 17 (second century).

Alongside the cursive hands there appear in the second century scripts in which the let­
ters often have a cursive base but overall are traced with a more or less slow ductus. Gen­
erally these show an upright axis, with ligatures frequent at times but in any case not af­
fecting letter shapes, and with some artificial shapes and extended hastas. These scripts
were used particularly (although not exclusively) in bureaucratic settings. A more clearly
distinctive stream is that inspired by the chancery style; it shows artificial strokes, fre­
quent presence of cursive forms written with a more or less slow ductus, and a modest
use of ligature. For the second and third centuries, examples are PSI XII 1227 (188), P.Ryl.
II 196 (196), P.Flor. I 6 (210), and P.Ryl. II 110 (259).

Page 14 of 37
Greek and Latin Writing in the Papyri

Figure 5.15. P.Flor. I 47. Photo courtesy of the Bib­

lioteca Medicea Laurenziana.

Figure 5.16. CPR I 32. Photo courtesy of the Pa­

pyrussammlung, Österreichische Nationalbibliothek.

Page 15 of 37
Greek and Latin Writing in the Papyri

Figure 5.17. P.Mur. 114. Photo courtesy of the Israel

Antiquities Authority.

This script sometimes shows hints of a rather characteristic chancery style, which asserts
itself at the turn of the third century, a style of which the most exemplary witness is the
well-known P.Berol. inv. 11532 (figure 5.20) and which exercised a more or less marked
influence on contemporary documentary writing. The papyrus was written in Alexandria
and is thus a rare example of writing from a central chancery. It is possible that this styl­
ization (characterized by some narrow letters, as can be seen in official documents like
BGU I 73) was already in use in chancery documents at the start of the second century.
The letters, rigorously perpendicular to the baseline, at times have an exaggerated pro­
longation in their vertical dimension (beta, epsilon, theta, kappa, omicron) and sometimes
a marked reduction in size (alpha, delta, omega). The script as a whole appears formal,
artificial, severe, and thus highly stylized. The same forms appear in the third (p. 121) cen­
tury in the official communication PSI XII 1247, perhaps also the product of the prefect's
office in Alexandria, and in the official letter P.Oxy. XIX 2227 (probably 215/216). It also
appears in less stylized forms, especially in copies made in minor offices, military docu­
ments, fiscal and judicial reports, as well as in private letters, contracts, and various oth­
er documents. For the period that encompasses the end of the second century and all of
the third, one may mention, for instance, P.Flor. II (p. 122) (p. 123) 278 (203/204), P.Lond.
II 345 (194), BGU I 296 (219/220), and P.Oxy. XXXI 2612 (288–290). Chancery hands also
appear in literary papyri, apparently written by functionaries: Examples are P.Beatty V
(Genesis), PSI II 127 (Judges), and PSI VI 727 (astrology), all from the late second or early
third century.

Formal scripts, devoid of any chancery stylization, are also found in documents from the
same period, particularly in bureaucratic milieus; these are to be categorized with the
rounded and sometimes calligraphic scripts such as those in the Gnomon of the Idios Lo­
gos (BGU V 1210, ca. 170) and P.Oxy. VIII 1100 (figure 5.21), a copy of an edict of the pre­
fect (206). These resemble the scripts also found in literary papyri. The writing of the
Gnomon, for example, is close to that in the famous papyrus of the Mimes of Herondas
(P.Lond.Lit. 96), assignable to the start of the second century, while the script of P.Oxy.
VIII 1100 is found in the same period in the papyrus of Favorinus (P.Vat. 11), later than
190/191, in BKT IX 58 (Iliad), before Septimius Severus, and in the Christian papyri P.Ryl.
III 463 (Gospel of Mary) and PSI VIII 980 (Psalms). Elegant calligraphic hands that are al­

Page 16 of 37
Greek and Latin Writing in the Papyri

so flowing appear particularly in PSI V 446 (prefect's edict, 133–136) (figure 5.22), P.Oxy.
III 473 (138–160), and P.Mur. 113 (Murabbaʼat). A similar but more rigid and less elegant
handwriting is found in BGU III 895 (138–161), a will probably written in Syria.

Figure 5.18. P.Yadin 18. Photo courtesy of the Israel

Exploration Society.

Figure 5.19. P.Dura 23. Photo courtesy of the Bei­

necke Rare Book and Manuscript Library, Yale Uni­
versity.

Page 17 of 37
Greek and Latin Writing in the Papyri

Figure 5.20. P.Berol. inv. 11532.

Figure 5.21. P.Oxy. VIII 1100.

Page 18 of 37
Greek and Latin Writing in the Papyri

Figure 5.22. PSI V 446. Photo courtesy of the Bib­

lioteca Medicea Laurenziana.

The skilled hands found in literary papyri of the second and third centuries display a
great variety of graphic solutions. In this period the rise of new cultural currents, circles
of readers, literary practices, texts, and public and private libraries drove a more exten­
sive and varied production of books than the past. At this time, the most notable phenom­
enon in writing found in the domain of skilled and (p. 124) (p. 125) (p. 126) (p. 127) calli­
graphic hands is the development of normative scripts (i.e., handwritings that follow pre­
cise rules and are repetitively stable in their technique and manner of execution, with the
result that they have great staying power).

Figure 5.23. The Hawara Homer.

The first group is made up of the regular and rounded scripts that constitute the continu­
ation of analogous hands in previous centuries. Here, from the first half of the second
century (or possibly the end of the first), we find PSI VIII 978 and PSI XI 1197 + PSI Con­
gr. XVII8 + P.Oxy. II 226; from the second century we find P.Oxy. III 454 and, in a less for­
mal style, PSI IX 1095, P.Oxy. X 1231, and P.Berol. inv. 9782. The assignment of these
Page 19 of 37
Greek and Latin Writing in the Papyri

hands to the second century is supported by the affinity they display with documents like
P.Berol. inv. 6854 (Trajan) and the Gnomon of the Idios Logos (BGU V 1210, ca. 170). To
the turn of the second to the third century and to the (p. 128) third century we attribute
P.Beatty VI (Numbers), and to the turn of the third century to the fourth, P.Beatty IX
(Ezekiel and Esther). Among these curvilinear hands the so-called round majuscule (also
called the “Roman uncial”) stands out, representing one of the pinnacles of Greek callig­
raphy. Because its characteristics (with only minor variation) are identical in many exam­
ples, it was clearly a normative script; the letters are drawn with maximum regularity, the
strokes tend to curvilinear forms (note the mu), and even oblique lines sometimes present
a slight curvature. The script as a whole is markedly bilinear and adorned with decorative
apices.10

Among the most important examples are the famous Hawara Homer (P.Haw. 24–28) (fig­
ure 5.23), PSI XI 1213, PSI I 8, and BKT V 1, 28–30, all of the second century, the floruit of
this remarkable calligraphy. This flourishing was, however, ephemeral, essentially limited
to the Antonine period, with no examples after the second century. An imitation was pro­
duced in the late fifth century, perhaps in pagan Alexandrian circles. It appears, in fact, in
the famous Ilias Ambrosiana, in the Iliadic fragments PSI VII 748 + 749 and in two Platon­
ic fragments, P.Duke inv. 5 and P.Ant. II 78.

A new graphic grouping with normative claims, the so-called biblical majuscule, makes its
appearance starting in the late second century.11 Its basis was a sober and undecorated
script, which can be observed at the turn of the second to the third century in P.Oxy. XVIII
2169 (Callimachus), P.Berol. inv. 9968 (Homer), as well as P.Mich. II 135 + P.Mil. I 13 (Ec­
clesiasticus). Among these scripts emerges (p. 129) the true biblical majuscule, which re­
flects in its penmanship the base models of the letters and is carried out with a visible
contrast between thin horizontal strokes and fatter vertical ones (particularly gamma, pi,
tau), while oblique strokes appear in between (alpha, delta, lambda). Rho and upsilon
project below the baseline, and the hastas of phi and psi project both up and down.

The number of witnesses to the biblical majuscule—by no means all biblical—is very
large. For the first centuries of its existence one may cite the classical texts in P.Ryl. I 16
(before 255/256), PSI IX 1086 (figure 5.24), P.Berol. inv. 7499 + 7502, P.Oxy. IX 1179, and
P.Oxy. XXII 2334, the last of the third to the fourth century. This handwriting was widely
adopted for Christian texts, in particular for the Scriptures, to the extent that it was used
for the great biblical codices (Vaticanus, Sinaiticus [both fourth century] and Alexandri­
nus [fifth]). Its continuing use in book production is attested up to the eight century and
beyond in numerous examples on both papyrus and parchment, secular and especially
Christian. Besides the biblical codices already mentioned, important examples for the
fourth century are the parchment leaf P.Oxy. XIII 1621 and P.Beatty IV (Genesis); for the
fifth, P.Oxy. III 411 (Thucydides), and P.Amh. I 1 (ascension of Isaiah); for the sixth, the fa­
mous illustrated codex of Dioscorides in Vienna, produced at Constantinople in 513 or a
bit later, and the parchment fragment of Jonah, PSI X 1164 + BKT VIII 18 (figure 5.25).
Around the seventh century the biblical majuscule begins to disappear in the last Greek
books produced in Egypt, while it survives, even if only sporadically, in Greek book pro­

Page 20 of 37
Greek and Latin Writing in the Papyri

duction in other areas. Among the late examples, as can already be observed in the Vien­
na Dioscorides, the script shows a stronger contrast between fat and thin strokes and
decorative buttons at the extremities of the latter, in particular on the horizontal strokes
of gamma, delta, epsilon, pi, and tau.

Figure 5.24. PSI IX 1086. Photo courtesy of the Bib­

lioteca Medicea Laurenziana.

Figure 5.25. PSI X 1164. Photo courtesy of the Bib­

lioteca Medicea Laurenziana.

Yet another type that emerged in the second century is that of a script showing curving
strokes and a more fluid ductus, which appears from the reign of Antoninus in documents
like PSI V 446 (figure 5.22) and P.Oxy. III 473. This type of handwriting later turned into a
normative style, the “Alexandrian majuscule,” which was also widely used in both classi­
cal and Christian book production.12 The first signs of this script appear in the second-
century gospel codex P.Egerton 2; in P.Bodm. II (John) and P.Ant. I 28 (Hippocrates) of the
third century; in PSI II 125 (Acts, fourth century); P.Ant. I 12 (2 John, fifth century); and
PSI I 1 (Matthew, fifth–sixth century) (figure 5.26). In the last of these, as in many other
sixth-century manuscripts (e.g., P.Amh. II 191, 192; BKT VIII 4), the Alexandrian majus­
cule appears with all of its normative characteristics. The script is soft and lowing, with
Page 21 of 37
Greek and Latin Writing in the Papyri

oblique lines descending from left to right and tending to be prolonged on the baseline to
form pseudoligatures; alpha, mu (with internal strokes fused into a single curve), upsilon,
and omega present variably open or closed loops; epsilon sometimes has its upper curve
brought down to the middle stroke; and delta and lambda generally have a curl at the up­
per end of the right-hand oblique stroke. (p. 130) (p. 131)

This same script, with curved strokes and fluid ductus, appears in the second century in a
narrower version, with epsilon, theta, omicron, and sigma particularly squeezed, eventu­
ally turning into an Alexandrian majuscule that differs from that just described in a sharp­
er contrast between wide and narrow letters. Good early examples are P.Oxy. XXXII 2631
(second century, lyric) and P.Lund. IV 13 (260–270); nonetheless, this script remains rare
until the fourth century and then comes into wider use from the fifth century on. Exam­
ples from this period are P.Berol. inv. 13418 (fifth century), P.Oxy. XV 1820 (fifth–sixth
century), P.Oxy. XX 2258 (sixth century), PSI XIV 1400 (seventh century), P.Louvre E 7404
(seventh–eighth), and PBad. IV 58 (eighth). The dating of the Alexandrian majuscule has a
series of pegs in some festal letters issued by the Alexandrian patriarchs assignable to
precise or highly probable dates. The best known of these are P.Grenf. II 112 (577) and
BKT VI 55–109 (713 or 719). This script was used not only by the patriarchal chancery of
Alexandria but also (adapted to the Coptic alphabet) as an ecclesiastical script used for
centuries for Coptic texts, such that it seems to have been particularly widely used in
Egypt and its vicinity; there are examples from Nessana in the Negev.13

Figure 5.26. PSI I 1. Photo courtesy of the Biblioteca

Medicea Laurenziana.

A further type of script that began spreading in the second century is that identified by
Wilhelm Schubart, who called it “severe style,” a term still sometimes used. It is a script
based on rigid and angular hands, characterized by a contrast in size between broad let­
ters (mu, nu, tau, omega) and narrow letters (epsilon, theta, omicron, sigma), and seen al­
ready in the Hellenistic period. The severe style has a greater degree of simplicity and a

Page 22 of 37
Greek and Latin Writing in the Papyri

tendency not only to squeeze some letters but also to shrink them. Schubart mentioned as
particularly significant examples of the severe style P.Oxy. X 1234 + XI 1360 + XVIII 2166
(Alcaeus) and especially the (p. 132) famous Bacchylides papyrus P.Lond.Lit. 46 + PSI XII
1278, to which one may add P. Oxy. IX 1174, P.Oxy. I 26 (both second century) (figure 5.27),
and P.Oxy. VII 1016, PSI X 1170, and P.Oxy. XVII 2098 (early third century), where a slight
inclination of the axis is noticeable. In the severe style, this slant toward the right is
marked to the point of becoming a true sloping script. It appears both in documents (e.g.,
P.Oxy. XXII 2341 [202] and P.Flor. II 259 [before 260]) and in a considerable number of lit­
erary papyri, both classical and Christian. Examples are PSI XI 1203 (late second centu­
ry), P.Oxy. V 842 (second—third), P.Ryl. I 57, PSI X 1169 (figure 5.28), and P.Lond.Lit. 5
(verso) + 182 (all third century).

Figure 5.27. P.Oxy. I 26.

In the following centuries, the upright severe style seems to evolve into the Byzantine
script called “upright ogival majuscule,” taking on a certain contrast in the thickness of
the strokes and thus exhibiting a stylistic taste of a general nature ever more widely dif­
fused in late antiquity.14 Fourth-century examples are P.Oxy. XI 1352 (Psalms) and PSI X
1171 (Aristophanes), and fifth-century, P.Flor. III 389 (Sibylline oracles) (figure 5.29).
However, instances from this period are rare; its floruit is middle Byzantine, thus rather
later. By contrast, scripts that lean more to the right were widely dif used from the fourth
century on. Prominent among these are the “inclined ogival majuscule,” which, as the
Byzantine period progresses, also (p. 133) acquires some degree of contrast in the width
of the strokes. A faster, more cursive version appears in the fourth century in P.Herm. 4
and 5 (ca. 320–325), P.Oxy. XXXIII 2656 (= LXIV 4408), and P.Bodm. IV; a slower and
more rigid form that occurs in the same period appears in P.Beatty XI, PSI X 1165, and
P.Oxy. XXXIV 2699. In this last papyrus the inclined ogival majuscule takes on an uncom­
monly marked contrast in the thickness of strokes. This characteristic becomes normal in
the following centuries both in Egypt and in Palestine.15 Examples include the famous
Menander codex P.Cair. 43227 (Wfth century), P.Oxy. XV 1817 and 1818 (sixth), PSI XIII
1296 (seventh), and P.Ness. 6 (eighth).

Page 23 of 37
Greek and Latin Writing in the Papyri

Figure 5.28. PSI X 1169. Photo courtesy of the Bib­

lioteca Medicea Laurenziana.

Figure 5.29. P.Flor. III 389. Photo courtesy of the

Biblioteca Medicea Laurenziana.

The more or less regular scripts (used particularly in book production) that we have ex­
amined so far, along with their successors in the seventh and eight centuries, do not ex­
haust the graphic range of the Roman period. They constitute only some of the more
characteristic, widely diffused, and durable scripts of the period. The complex and varied
range of other handwritings makes it impossible to give a detailed and ordered classifica­
tion. Many are generic scripts that do not fall into well-defined typologies but reflect the
diverse tendencies of the period and combinations of individual forms. Some are minute
and densely written, like P.Oxy. V 843 or P.Flor. II 112; others are larger and more broad­
ly spaced, as in P.Lond. I 130 or P.Beatty VII + P.Mert. I 2 + PSI XII 1273; while others are
forms derived from cursive, as in P.Oxy. VII 1019 + P.Oxy. XLI 2948 and PSI X 1181. Still
others resemble more formal scripts, like P.Oxy. X 1250, or are even inspired by chancery
style, like PSI VIII 982. (p. 134)

Page 24 of 37
Greek and Latin Writing in the Papyri

(p. 135) The Byzantine Period

In papyrological parlance, the Byzantine period for cursive and semicursive writing in
Egypt begins with Diocletian. Although the documentation remains abundant, it is less di­
verse in type than that of the Roman period, concentrated as it is in groups connected
with the great families of state functionaries and landowners and with the new institu­
tions represented by the churches and monasteries. Over time, the production of books
moves away from the texts of the classical tradition—apart from those used in education
—to Christian scriptures and literature.

Bureaucratic and chancery usage of scripts in the fourth century, whether cursive, semi­
cursive, or regular, witnesses both continuity with the hands of the second-to-third centu­
ry and significant transformations. Continuity in the cursive is evident, for example, in the
inclined writing of BGU II 408 (312), while in the tradition of older semicursive scripts
with some chancery elements, we find documents like PSI X 1107 (336) and P.Flor. I 75
(380). More marked elements of second-to-third-century chancery script (but without ele­
ments like the contrast in letter size) appear in BGU II 405 (348), where a gentle slope in
the axis of the letters is observable.

From Diocletian on, Greek majuscule bureaucratic handwriting comes increasingly into
contact with Latin minuscule writing, resulting in the formation of certain graphically
equivalent but often phonetically different signs; examples are d = delta (but also t + i =
rounded delta), h = eta, m = pi + iota. This phenomenon (the so-called Graeco-Latin
graphic koinê) initiates a rapid process of the transformation of majuscule graphic forms
into minuscule, even if the latter in some cases had already been found in Greek writing
for a long time. The phases of this process are noticeable beginning in the fourth century
in cursive and semicursive hands, whether upright or inclined. Particularly fine speci­
mens of the rapprochement between Greek and Latin handwriting appear in judicial and
administrative practice, like the bilingual P.Lips. I 38, a trial before the praeses Thebaidis
in the year 390, and PSI XIII 1309 (early fifth century), similarly a trial. However, the
same graphic phenomenon can be traced in purely Greek papyri written in bureaucratic
hands. The archive of Abinnaeus furnishes interesting examples for the mid-fourth centu­
ry; for the fifth, good illustrations are PSI XII 1265 (426) and P.Flor. III 315 (435).

Beginning at the latest with the fifth century, minuscule forms also enter the scripts in or­
dinary use. In documents like PSI XII 1239 (430) and BGU II 609 (442) we observe the
substitution—still unsystematic—of some minuscule letters of the koinê for those of the
traditional type. A more pervasive presence of minuscule elements, both letters and liga­
tures, is found, for instance, in PSI X 114 (454) and P.Flor. I 94 (491), which show that at
the turn of the fifth to the sixth century the transition in Greek writing from majuscule to
minuscule, at least in documentary practice, was essentially complete.

During the same time span, however, and even later, the traditional majuscule
(p. 136)

survives, even if sometimes mixed with certain minuscule forms. Its range includes semi­
cursive and detached handwritings, both upright and inclined, used both for documents

Page 25 of 37
Greek and Latin Writing in the Papyri

and for books, both classical and Christian. Documents include SB VIII 9907 (388), P.Köln
III 151 (423), SB XVI 12865 (576), and PSI I 60 (595). These scripts also occur in literary
papyri like P.Köln III 134 (fourth century) and P.Oxy. XIII 1614 and P.Egert. 5 (Wfth centu­
ry). These hands all appear very rough and disorderly. Calligraphic writing by this point is
limited to styles long established by tradition—biblical majuscule, Alexandrian majuscule,
upright and inclined ogival majuscule—which, as we have seen, continue in use.

As it emerged from the Graeco-Latin koinê by the sixth century, Byzantine minuscule cur­
sive continued its life down to the Arab period in two parallel styles of writing. The most
widely attested one is swift, decisively inclined to the right, rich in ligatures, and charac­
terized by elongated strokes that extend above and below the line, along with artificial
swirls and flourishes. Good examples are BGU I 255 (599), P.Flor. I 70 (627), and P.Berol.
inv. 13371 (ca. 650), which bring us into the Arab period, during which the same type of
writing appears in many papyri of the correspondence of Qurrah ben Sharik (e.g., P.Lond.
IV 1348 [710]). The papyri from Petra and Nessana of the sixth and seventh centuries
show that this manner of writing was in use also outside Egypt.

The other type is a stylized chancery script with a slower ductus and an upright or only
slightly inclined axis, lacking deforming ligatures, tending to isolate and round the body
of the letters, and exhibiting modestly elongated and hooked hastas. Some letters in par­
ticular (lambda, mu, pi, tau) show a form different from that encountered in the other
stream of Byzantine cursive.16 This stylization is in use and widely diffused above all in
the Arab period, as is apparent in P.Apoll. 9 (675/676 or 660/661), P.Lond. IV 1408 (709),
and above all P.Lond. I 32, an official letter of the dux Attyia ben Guʼaid (698/699, less
probably 713/714), and SB I 5638 (710).17 What is striking is that, in the course of the
Arab period, this chancery stylization begins to be adopted also for nondocumentary use.
It appears, in fact, even in some Christian papyri with devotional texts, like MPER n.s. IV
31 + XVII 7, MPER XVII 34, MPER XVII 53, or even P.Aberd. 72a.

When around 800 CE new cultural movements arose in the Byzantine world and the an­
cient and monumental majuscule scripts for book production, now regarded as slow and
tiring in execution, began to disappear, it is the chancery stylization of the Byzantine cur­
sive that, stripped of its most bureaucratic elements (e.g., its swirls and excessive elonga­
tion of hastas) is promoted to become also a book handwriting, thus turning into the nor­
mal Byzantine Greek minuscule. Nonetheless, in this period the torch of Greek culture
and script had passed from the Nile and Alexandria to the Bosporus and Constantinople.

(p. 137) Latin Handwriting

Latin papyri are relatively scarce compared to the immense number of Greek papyri;
moreover, most of them come from Egypt, whereas Latin was written mostly in Italy and
the western provinces of the Roman world. There is, all the same, enough material to doc­
ument the course of Latin handwriting at least from the first century BCE.

Page 26 of 37
Greek and Latin Writing in the Papyri

To begin, cursive writing is not attested in the available material before the first century
CE, when we meet it in BGU II 611, the famous papyrus of the oratio Claudii on reforming
justice, which is later than 41–54. This is the oldest example of authentic Latin cursive. It
shows that Latin writing had already, at least for some purposes, freed itself from the
hard strokes natural to incision on tablets. In particular, certain characteristic forms of
Latin cursive are evident, like B with its belly to the left, E with oblique middle stroke, and
Q with a long tail. See PSI VI 729 (77 CE) (figure 5.30) for an illustration of the same type
of writing.

Figure 5.30. PSI VI 729. Photo courtesy of the Bib­

lioteca Medicea Laurenziana.

Figure 5.31. P.Herc. 817. Photo courtesy of the Bib­

lioteca nazionale Napoli and Brigham Young Univer­
sity (Provo, Utah). All rights reserved. Courtesy of
the Ministero per i Beni e le Attività Culturali, Italy.

Another type of writing is that which, even if executed in ink, seems to represent the
transposition of an incised script meant for wax tablets. It shows a geometric course; iso­
lated letters; an absence of ligatures; hard, cutting strokes; broken curves; and some­
times prolongation of hastas as in incised texts. This script appears in a fragmentary let­
ter of ca. 29–22 BCE, P.Vindob. inv. L 1b(ChLA XLII 1241), P.Berol. inv. 8334 (ChLA X 417;
83–86, copy of imperial codicil), P.Berol. inv. 11649 (ChLA X 424; letter), all of the first
century, and the Severan BGU II 628r. This rigid and disarticulated writing is found in the
West in a number of tablets from Vindolanda and in the East in P.Lond. 229 (166, Seleu­
cia). This script was that most commonly used on papyri from the first century BCE to the
second century CE. In the same graphic track and contemporary to these documents we
also find literary texts like P.Heid. L 1a, b (CLA VIII 1220) and P.Mich. VII 430. This same
script is sometimes more compact and leans slightly on its axis, as in P.Vindob. inv.L 1c
Page 27 of 37
Greek and Latin Writing in the Papyri

(ChLA XLIII 1241), PSI XIII 1307 and 1308, and P.Mich. VII 442 (second century). (p. 138)
Once again, Vindolanda's ink tablets offer parallels in the early second century (Tab.Vind.
II 291, 310, 345). The presence of this script in a legal roll (P.Mich. VII 456, first century)
and in a literary text of the second century, P.Oxy. XVII 2088, shows that it was also used
in book production.

Figure 5.32. P.Vindob. inv. L 112. Photo courtesy of

the Papyrussammlung, Österreichische Nationalbib­
liothek.

Before further considering the second and third centuries, let us look at the more formal
and calligraphic scripts, which are among the oldest witnesses to Latin writing on pa­
pyrus, those that show, taken together, a normative handwriting. This is the capital nor­
mally used in books and whose characteristics can be summarized as follows: substantial­
ly epigraphic letter forms; soft drawing; usually a contrast in thickness between strokes;
frequent addition of elegant apices at the ends of vertical strokes.18 One of the first exam­
ples of these capitals is P.Berol. inv. 13956 (ChLA X 428; second half of the first century
CE); in a more calligraphic form, the same script appears in PSI XI 1183 (45–54), with an
outer text in book majuscule and an inner text (the same) in a harsh and broken cursive.
If the same scribe wrote both, as seems likely, it seems clear that some Latin scribes were
trained in two types of writing. A book majuscule is also attested in BGU VII 1689 (122–
145) and P.Cair. JdʼE 39513 (ChLA XLI 1191; first–second century). (p. 139)

However, it is obviously in books that this capital writing is most amply attested. The Her­
culaneum papyri offer examples that were certainly produced in Italy, but so do Egypt
and other Mediterranean areas. All of these were presumably written in the Roman west
or by hands trained there. Among the Herculaneum finds the majuscule is sometimes
more fluid, as in P.Herc. 817 (figure 5.31), the famous Carmen de bello Actiaco, and some­
times more formal, as in P.Herc. 1475 and 1067, securely datable before 79 CE (the first
after 46 BCE and the second after 27 BCE). Egyptian examples include P.Oxy. VI 871,
P.Hamb. 167, and the Sallustian fragments P.Ryl. I 42 and P.Ryl. III 473 (both second cen­
tury). Particularly in the oratorical fragments from Herculaneum and in P.Ryl. I 42 the
writing demonstrates a large, squared module characterized by soft, thick oblique lines
from left to right and by soft little strokes crowning the hastas, as if traced by a little pen.
Page 28 of 37
Greek and Latin Writing in the Papyri

This is the highest-quality majuscule used in Roman books from the time of Augustus on.
A particularly interesting instance is the “Cornelius Gallus papyrus,” found at Qasr Ibrim
in Nubia.19

Another type of regular capital appears in documentary practice. Its most distinctive
characteristic is the absence of contrast between thick and thin strokes. Instances are
P.Vindob. inv. L 112 (ChLA XLV 1323; after 129) (figure 5.32), P.Iand. inv. 209 (ChLA XI
491; second century), and BGU II 696 (156).

Figure 5.33. P.Grenf. II 110.

Among the scripts in use in chancery practice, which were developing already in the late
first century and continued to emerge and develop in the second, is a headlong cursive
with thin strokes decisively inclined to the right, in which certain forms and ligatures
gradually develop into a system. This particular hand emerges fully in the third century in
a near-normative documentary script.20 The most (p. 140) characteristic letters are A
made with only outer oblique lines that meet along the stroke descending from left to
right, B with its belly to the left, D with an open or “laced” eyelet, E in the shape of V, M
and N with their middle strokes almost always elevated, P with the upper curve reduced
to an oblique stroke and R with it turned into a wavy line, and Q with its descending hasta
turned from the left.

Second-century examples are P.Oxy. VII 1012, PSI IX 1026, P.Grenf. II 108, and P.Gen.Lat.
8; third-century instances are P.Ryl. IV 553, P.Dura 59, 63, and P.Oxy. XLI 2951. These
same chancery forms, although not so formal, are also found in many ostraca from Bu
Njem in Libya, written in ink perhaps with a soft, small reed pen; O.Bu Njem 76–80, mili­
tary letters of 259 written by a single hand, are good examples.21 In addition, the same
style of writing was sometimes used for paraliterary texts like P.Oxy. XI 1404. This
chancery writing virtually disappears at the end of the third century and survives only in
the imperial chancery, which alone was permitted to use the so-called litterae caelestes.
The imperial rescripts from 436–450 in Paris and Leiden are notable examples.

It is possible to document a freer evolution of everyday cursive script between the first
and the third centuries, a handwriting usually called “common script.” This evolution may
be followed in both incised and inked writing, which show a process of reduction or sim­
Page 29 of 37
Greek and Latin Writing in the Papyri

plification of the letter strokes leading to the creation of the minuscule. This process can
be observed already underway in some letters during the late first century in the wall in­
scriptions of Pompeii and on plates and (p. 141) bowls from La Graufesenque (ancient
Condatomagos in Gaul),22 as well as in some first—second-century tablets from Vindolan­
da, such as the epistolary diptych Tab.Vind. II 343. In the papyri, some minuscule forms
appear as variants in less formal examples of the chancery style of the second–third cen­
tury, like P.Dura 60 and 82. However, from the third to the beginning of the fourth centu­
ry, other documents show more fully the evolution of the common script in the sense of
the minuscule; an example is the cursive script, alternating with capitals, of P.Mich. III
164 (242–244), P.Grenf. II 110 (293) (figure 5.33), PSI I 111 (287–304), and P.Argent.Lat. 1
(317–324). The brief sections owed to an official in P.Dura 59 are also substantially in mi­
nuscule.

In all of these, apart from occasional appearances of residual forms of the majuscule, the
script can be considered to be minuscule, which takes over definitively from the majus­
cule in the fourth century. In a petition of Flavius Abinnaeus, P.Abinn. 1 (341/342), the
new cursive shows all of its characteristics. It is the final result of the evolution of the
everyday cursive during the two previous centuries and the product of a more wide­
spread literacy. If only a few examples from this period appear in the documentation, that
is due to the dominance of official documents in the Egyptian evidence and to the near
absence of the products of everyday writing in the West.

In the period from the late second to the fourth century, the evolution of Latin writing to­
ward the minuscule and its replacement of the majuscule brought about changes as well
in the use of more formal scripts, although calligraphic capitals continued to be used in
various documents and books. Instances of the latter are the Feriale Duranum (P.Dura 54,
225–235) and the Aeneid codex P.Ant. I 30 (late fourth century). The majuscule still had a
long life ahead of it, above all in high-quality books; the great late antique Vergilian
codices of Italian origin and preserved in libraries, some of them elegantly illustrated, are
sufficient to make the point.

In the skilled hands, minuscule forms (d, h, q) appear as early as the second century in
the codex P.Oxy. I 30, but it is mainly from the third century that they appear ever more
frequently alongside the majuscule as the latter falls into disuse. These hands are difficult
to categorize, like the Epitome of Livy, P.Oxy. IV 668, assigned to the third century (but
perhaps instead from the third–fourth), which displays a minuscule with traces of the ma­
juscule (A, G, N). Another such hand appears in a fragment of the Sententiae of Paul, CLA
X 1577, and in the liturgical fragment P.Ryl. III 472, approximately contemporaneous.23
Moreover, a number of versions of a genuine minuscule were carried out with slow duc­
tus and more or less marked separation between letters. One example is PSI VII 848, as is
the codex now in Montserrat, containing Cicero's Catilinarians, the so-called Latin Alces­
tis, and a responsory psalm: With these papyri we come to the fourth century. From this
point on, especially in the fifth century, “papyrological” material properly speaking

Page 30 of 37
Greek and Latin Writing in the Papyri

is joined by codices and documents preserved continuously in libraries and

(p. 142)

archives. I can give only a few references to the latter here.

In daily use and private documentation, the new cursive remains the dominant script af­
ter the fourth century. Starting in the fifth century, it shows a faster ductus and acquires
some new characteristics, like a certain upward direction and a decided slant of the axis
to the right. The best examples are not from Egypt but from Italy, particularly Ravenna:
P.Ital. 1 (445/446), P.Ital. 35 (572), and P.Ital. 20 (590–602).

Chancery-type stylizations of the same new cursive are found from the fourth–fifth centu­
ry on in both the East (where Latin continues to be used in bureaucratic practice) and the
West.24 The two regions share certain habits like a rigorously vertical axis, a contrast in
size among letters, and a strong prolongation of some hastas even though everything sug­
gests that in other stylistic respects the two differed considerably. In eastern provincial
offices, the method of giving a recognizably bureaucratic appearance was making some
letters rounder and larger, a trait visible in P.Vindob. inv. L 31 (after 386) (figure 5.34), a
copy of an imperial rescript made in the office of the prefect of Egypt to be sent to the
praeses Arcadiae, or in P. Ryl. IV 615 (fifth century). In provincial and local offices in the
West, by contrast, a certain vertical direction in the writing marked official script, a trait
already found in third-century chancery style; to it were added the compression of eyelets
and the twisting of hastas, as ChLA I 5 (mid-sixth century) shows clearly. In the final
analysis, the provincial and municipal offices of the East, like those in the West, were dri­
ven to adopt the litterae communes, the current minuscule, and to develop new stylistic
variants of it in order to give their documents a distinctive graphic physiognomy in the
wake of the imperial constitution issued at Trier in 367 (CTh 9.19.3), forbidding the use of
the litterae caelestes, based on the old cursive and now reserved solely for the imperial
chancery.

In the stream of formal scripts during the fourth and following centuries, the semiuncial
and uncial were dominant. In the East, the semiuncial is attested several times in its vari­
ant with the axis leaning to the right. This is apparent in various legal texts, like
P.Bodl.Lat. class. G 1 (CLA II 248; fourth century), P.Vindob. inv. L 90 (CLA VIII 1042;
fourth–fifth century) (figure 5.35), and PRyl. III 474 and P.Rein. inv. 2219 (CLAV 700; both
sixth century). The same semiuncial also occurs with an upright axis, as normally in the
West, but it presents in general a rather crowded appearance, as in P.Oxy. 2401, or influ­
ences from Greek script as in P. Oxy. VI 884, both dated in the fifth—sixth century.

Page 31 of 37
Greek and Latin Writing in the Papyri

Figure 5.34. P.Vindob. inv. L 31. Photo courtesy of

the Papyrussammlung, Österreichische Nationalbib­
liothek.

Figure 5.35. P.Vindob. inv. L 90. Photo courtesy of

the Papyrussammlung, Österreichische Nationalbib­
liothek.

Uncial writing is attested in the papyri from the moment of its creation, but that remains
problematic.25 This script is a mixture of majuscule forms drawn from capitals, some re­
placed with minuscules (h, p, q, and u), and others from characteristic forms of A, D, E,
and M. Even these last forms, however, appear in the Latin graphic tradition, even if only
occasionally. The D and the A of the uncial can be observed respectively in P.Oxy. I 30
(perhaps second century) and P.Oxy. IV 668 (p. 143) (p. 144) (p. 145) (third—fourth centu­
ry), while E and M (the latter still in the process of formation) are attested in P.Berol. inv.
6757 (CLA VIII 1033) and P.Vindob. inv. L 103 (CLA X 1537), dated to the beginning of the
fourth century or even the end of the third; finally, A, E, and M appear in uncial form in P.
Oxy. XI 1379, perhaps also from the start of the fourth century. The uncial seems alto­
gether an artificial script, in which forms from majuscule and minuscule were combined
and exhibiting various other influences as well, including possibly even biblical majus­
cule. The final product is visible in PSI XII 1272 (perhaps fifth century) and in manu­

Page 32 of 37
Greek and Latin Writing in the Papyri

scripts preserved in libraries (see British Library Add. MS 40165 A.1 [CLA II 178; late
fourth century] or Marburg, Staatsarchiv Hr 1.1 [CLA suppl. 1728; fifth century]). In the
East and in papyrological finds, the uncial is attested above all in juristic fragments and
in a particular typology that E. A. Lowe has called the “B-R uncial” because of its most
characteristic letters: the high B and the R, with its hasta descending beneath the base­
line and its oblique stroke in a horizontal position. Besides various juristic texts one finds
it in PSI XI 1182 (Gaius), P.Strasb. I 6b (Ulpian), P.Oxy. XV 1813 (CTh), P.Oxy. XV 1814 and
PSI XIII 1347 (CJ, first and second editions, respectively), and P.Ryl. III 479 and
P.Heid.Lat. inv. 4 (CLA VIII 1221; Justinian's Digest). All of these are dated to the Justini­
anic era or close to it and are written in a very formal B-R uncial. At the same time, other
examples, although inspired by this uncial, display forms (e.g., m or d of minuscule type)
that diverge from it.

The rupture between the East and the West that developed after Justinian and was con­
cluded in the reign of Heraclius led to the disappearance of Latin writing in the eastern
provinces of the empire.

Bibliography
Bastianini, G., and C. Gallazzi. (2001). Posidippo di Pella: Epigrammi (P. Mil. Vogl. VIII
309). Milan: Edizioni Universitarie Lettere Economia Diritto.

Capasso, M. (2003). Il ritorno di Cornelio Gallo: Il papiro di Qaṣr Ibrîm venticinque anni
dopo. With a contribution by P. Radiciotti. Naples: Graus.

Casamassima, E., and E. Staraz. 1977. “Varianti e cambio grafico nella scrittura dei pa­
pyri latini. Note paleograWche.” Scrittura e civiltà 1: 9–110.

Cavallo, G. (1967). Ricerche sulla maiuscola biblica. Florence: Le Monnier.

———. (1983). Libri scritture scribi a Ercolano. Naples: Macchiaroli.

———. (2005). Il calamo e il papiro: La scrittura greca dallʼetià ellenistica ai primi secoli
di Bisanzio. Florence: Gonnelli.

———, E. Crisci, G. Messeri, and R. Pintaudi, eds. (1998). Scrivere libri e documenti nel
mondo antico. Florence: Gonnelli.

Cavallo, G., and H. Maehler. 1987. Greek Bookhands of the Early Byzantine Period, A.D.
300–800. London: University of London. Institute of Classical Studies.

Cencetti, G. 1993. Scritti di paleografia, ed. G. Nicolaj. Dietikon-Zurich: Urs Graf Verlag.

Cribiore, R. (1996). Writing, Teachers, and Students in Graeco-Roman Egypt. Atlanta:

Scholars Press.

Crisci, E. 1985. “La maiuscola ogivale diritta: Origine, tipologie, dislocazioni.” Scrittura e
civiltà 9: 103–145.

Page 33 of 37
Greek and Latin Writing in the Papyri

———. 1996. Scrivere greco fuori dʼEgitto: Ricerche sui manoscritti greco-orientali di
origine non egiziana dal IV secolo a. C. allʼVIII d. C. Florence: Gonnelli.

———. 1999. “I più antichi libri greci: Note bibliologiche e paleografiche su rotoli papira­
cei del IV–III secolo A.C.” Scrittura e civiltà 23: 29–62.

De Gregorio, G. 2000. “Materiali vecchi e nuovi per uno studio della minuscola
(p. 147)

greca fra VII e IX secolo.” In I manoscritti greci tra riflessione e dibattito: Atti del V Collo­
quio Internazionale di paleografia greca (Cremona, 4–10 ottobre 1998), ed. G. Prato, vol.
1, 83–151. Florence: Gonnelli.

Kern, O. 1913. Inscriptiones graecae. Bonn: Marcus and Weber.

Kresten, O. 1966. “Diplomatische Auszeichnungsschriften in Spätantike und Frühmittelal­

ter.” Mitteilungen des Instituts für österreichische Geschichtsforschung 74: 9–40.

Mallon, J. (1952). Paleographie romaine. Madrid: Consejo Superior de Investigaciones Ci­

entificas.

———. (1982). De lʼécriture: Recueil dʼétudes publiées de 1937 à 1981. Paris: Éditions du
Centre National de la Recherche Scientifique.

Mallon, J., R. Marichal, and C. Perrat. (1939). Lʼécriture latine de la capitale romaine à la
minuscule. Paris: Arts et Métiers Graphiques.

Marichal, R. 1948. “De la capitale romaine a la minuscule.” In Somme typographique, ed.

M. Audin, vol. 1, Les origines, 61–111. Paris: Audin Editeur.

———. 1950. “Lʼécriture latine et lʼécriture grecque du Ier au VIe siècle.” Lʼantiquité clas­
sique 19: 113–144.

———. 1956. “Lʼécriture du Paul de Leyde.” In Pauli Sententiarum Fragmentum Leidense

(Cod. Leid. B. P. L. 2589), ed. G. G. Archi, M. David, E. Levy, R. Marichal, and H. L. V. Nel­
son, 24–57. Leiden: Brill.

———. (1988). Les graffites de La Graufesenque. Paris: Éditions du Centre National de la

Recherche Scientifique.

———. (1992). Les ostraka de Bu Njem. Tripoli: Grande Jamahira Arabe, Libyenne, Popu­
laire et Socialiste, Département des Antiquités.

Messeri, G. 2005. “Relazioni fra papiri documentari e papiri letterari.” Nea Rhome 2: 5–
23.

Messeri, G., and R. Pintaudi. 1998. “Documenti e scritture.” In Scrivere libri: 39–53.

———. 2000. “I papiri greci dʼEgitto e la minuscola libraria.” In I manoscritti greci tra ri­
lessione e dibattito: Atti del V Colloquio Internazionale di paleografia greca (Cremona, 4–
10 ottobre 1998), ed. G. Prato, vol. 1, 67–82. Florence: Gonnelli.

Page 34 of 37
Greek and Latin Writing in the Papyri

Norsa, M. (1929). Papiri greci delle collezioni italiane: Scritture documentarie dal III sec­
olo a. C. al secolo VIII d. C. Rome: Istituto di Filologia Classica.

———. (1939). La scrittura letteraria greca dal secolo IVa. C. allʼVIII d. C. Florence: Ti­
pografia E. Ariani.

Orsini, P. (2005). Manoscritti in maiuscola biblica: materiali per un aggiornamento. Cassi­

no: Edizioni dellʼUniversità degli Studi di Cassino.

Pack, R. A. (1965). The Greek and Latin Literary Texts from Greco-Roman Egypt, 2nd ed.
Ann Arbor: University of Michigan Press.

Petronio Nicolai, G. 1973. “Osservazioni sul canone della capitale libraria romana fra I e
III secolo.” In Miscellanea in memoria di Giorgio Cencetti, 3–28. Turin: Bottega dʼErasmo.

Petrucci, A. 1962. “Per la storia della scrittura romana: i graffiti di Condatomagos.” Bul­
lettino dellʼArchivio Paleografico Italiano 3(1): 85–132.

———. 1991. “Paleografia greca e paleografia latina: significato e limiti di un confronto.”

In Paleografia e codicologia greca: Atti del II Colloquio internazionale (Berlino-Wolfenbüt­
tel, 17–21 ottobre 1983), ed. D. Harlfinger and G. Prato, 463–484. Alessandria: Edizioni
dellʼOrso.

Porro, A. 1985. “Manoscritti in maiuscola alessandrina di contenuto profano: Aspetti

grafici codicologici filologici.” Scrittura e civiltà 9: 169–215.

Radiciotti, P. 1998. “Osservazioni paleografiche sui papiri latini di Ercolano.”

(p. 148)

Scrittura e civiltà 22: 353–370.

Roberts, C. H. 1956. Greek Literary Hands, 350 B.C.–A.D. 400, 2d ed. Oxford: Clarendon.

Schubart, W. (1911). Papyri Graecae Berolinenses. Bonn: Marcus & Weber.

———. (1925). Griechische Palaeographie. Munich: Beck.

Seider, R. (1967). Paläographie der griechischen Papyri, vol. 1, Tafeln, Part 1, Urkunden.
Stuttgart: Hiersemann.

———. (1970). Paläographie der griechischen Papyri, vol. 2, Tafeln, Part 2, Literarische
Papyri. Stuttgart: Hiersemann.

———. (1972). Paläographie der lateinischen Papyri, vol. 1, Tafeln, Part 1, Urkunden.
Stuttgart: Hiersemann.

———. 1975. “Zur Paläographie der frühen lateinischen Papyri.” In Proceedings of the
XIV International Congress of Papyrologists, 277–284. London: Egypt Exploration Society.

———. 1978. Paläographie der lateinischen Papyri, vol. 2.1, Tafeln, Part 2.1: Literarische
Papyri. Stuttgart: Hiersemann.

Page 35 of 37
Greek and Latin Writing in the Papyri

———. 1981. Paläographie der lateinischen Papyri, vol. 2.2, Tafeln, Part 2.2: Literarische
Papyri. Stuttgart: Hiersemann.

———. 1990. Paläographie der griechischen Papyri, vol. 3.1, Text, Part 1: Urkundenschrift
I. Mit einer Vorgeschichte zur Paläographie der griechischen Papyri. Stuttgart: Hierse­
mann.

Tjäder, J.-O. 1955. Die nichtliterarische lateinischen Papyri Italiens aus der Zeit 445–700,
vol. 1. Lund: Gleerup.

———. 1979. “Considerazioni e proposte sulla scrittura latina nellʼetà romana.” In Paleo­
graphica Diplomatica et Archivistica: Studi in onore di Giulio Battelli, vol. 1, 31–62. Rome:
Edizioni di Storia e Letteratura.

———. (1982). Die nichtliterarische lateinischen Papyri Italiens aus der Zeit 445–700, vol.
2. Stockholm: Aströms Förlag.

Turner, E. G. (1978). The Terms Recto and Verso: The Anatomy of the Papyrus Roll. Brus­
sels: Fondation Égyptologique Reine Élisabeth.

———. 1980. “Ptolemaic Bookhands and Lille Stesichorus.” Scrittura e civiltà 4: 19–40.

———. (1987). Greek Manuscripts of the Ancient World, 2d ed., ed. P. J. Parsons. London:
University of London, Institute of Classical Studies.

Van Haelst, J. (1976). Catalogue des papyrus littéraires juifs et chrétiens. (Papyrologie, 1).
Paris: Publications de la Sorbonne.

Notes:

(1) For further discussion of Greek writing see Schubart 1911, 1925; Norsa 1929, 1931;
Roberts 1956; Cavallo 1967, 1983, 2005; Seider 1967, 1970, 1972, 1975, 1978, 1981,
1990; Turner 1987; Cavallo and Maehler 1987; Crisci 1996; Cavallo, Crisci, Messeri, and
Pintaudi, eds. 1998; for Latin writing see CLA I–XI and suppl.; ChLA I–XLIX; Mallon,
Marichal, and Perrat 1939; Marichal 1948; Mallon 1952, 1982; Tjäder 1955, 1982;
Cencetti 1993. Printed reproductions of the papyri cited can easily be found by consulting
the Leuven Database of Ancient Books (LDAB) and the Heidelberger Gesamtverzeichnis
der griechischen Papyrusurkunden Ägyptens (HGV), both available online.

(2.) Cribiore (1996, especially 97–118).

(3.) Besides the classic work of Turner (1978), see most recently Messeri (2005).

(4.) On the oldest Greek writing on papyrus, see particularly Turner (1980) and Crisci
(1999).

(5.) Messeri and Pintaudi (1998, 41–43).

(6.) Crisci (1996, 17–18).

Page 36 of 37
Greek and Latin Writing in the Papyri

(7.) Bastianini and Gallazzi (2001, 17).

(8.) On the evolution of these letters see Petrucci (1991, 472–477).

(9.) Crisci (1996, 44–49, 137–141).

(10.) Kern (1913, 48).

(11.) See Cavallo (1967); an update on biblical majuscule appears in Orsini (2005).

(12.) On the Alexandrian majuscule, besides Cavallo (2005) see Porro (1985).

(13.) Crisci (1996, 84–85).

(14.) On upright ogival majuscule see Crisci (1985).

(15.) Crisci (1996, 79–84).

(16.) Morelli (CPR XXII, 7) has called attention to these differences.

(17.) On the writing of these documents see De Gregorio (2000, 93–94).

(18.) Petronio Nicolai (1973) and Radiciotti (1998) have made studies of this type of capi­
tal.

(19.) See most recently Capasso (2003).

(20.) Besides Cencetti (1993), see Casamassima and Staraz (1977) and Tjäder (1979).

(21.) Marichal (1992, 182–188).

(22.) Petrucci (1962); Marichal (1988).

(23.) Marichal (1956).

(24.) On chancery stylization, besides Mallon (1982), Cencetti (1993), and Cavallo (2005),
see Kresten (1966).

(25.) See at least Tjäder (1974).

Guglielmo Cavallo

Guglielmo Cavallo is Professor of Palaeography, University of Rome—La Sapienza.

Page 37 of 37
The Greek and Latin Languages in the Papyri

The Greek and Latin Languages in the Papyri

Eleanor Dickey
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0006

Abstract and Keywords

For the first thousand years after the end of the classical period, documentary papyri con­
stitute the most important source of information on the development of the Greek lan­
guage. Meanwhile, Greek writing tablets rarely survive, so that the vast majority of the
extant corpus of subliterary Hellenistic and Roman Greek comes from documents written
on papyri or ostraca in Egypt. The Greek in which these documents were written is con­
sidered to be a form of koinê. Postclassical literature in both Greek and Latin tended to
take the classical model as a goal to be imitated linguistically. The languages, as they
were learned by children and used in ordinary conversation, were constantly evolving.
This article describes Greek phonology, Greek morphology and syntax, Greek vocabulary,
and the Latin of papyri and ostraca.

Keywords: Greek language, documentary papyri, Greek writing tablets, Egypt, ostraca, koinê, Latin papyri

Introduction
For the first thousand years after the end of the classical period, documentary papyri con­
stitute our most important source of information on the development of the Greek lan­
guage.1 Papyri are also significant for our understanding of the development of Latin but
less so, owing to the much smaller number of surviving Latin papyri and to the preserva­
tion of considerable amounts of subliterary Latin in other forms such as the Vindolanda
writing tablets. By contrast, Greek writing tablets rarely survive, so that the vast majority
of our extant corpus of subliterary Hellenistic and Roman Greek comes from documents
written on papyri or ostraca in Egypt.

The Greek in which these documents were written is considered to be a form of koinê
Greek. Koinê was created when the Macedonians exported an Ionicized form of the Attic
dialect to their newly conquered empire and used it as the language of government and
upper-class society. Most Hellenistic and some Roman-period Greek literature is also in
koinê; thus, literary texts as disparate as Polybius's histories and the New Testament fall
Page 1 of 22
The Greek and Latin Languages in the Papyri

into this category. The Greek of the documentary papyri has some features in common
with these literary texts, but in many respects it is considerably different from them.

Postclassical literature in both Greek and Latin tended to take the classical model
(p. 150)

as a goal to be imitated linguistically. The languages, as they were learned by children

and used in ordinary conversation, were constantly evolving; this is a natural and in­
evitable process common to all languages. But authors of literary texts did not attempt to
write in the language of everyday conversation; rather, they aimed at stylistic norms that
in many cases had been established centuries earlier. Some Roman-period authors wrote
in nearly flawless imitations of classical Attic, and others used a language that made
more concessions to subsequent changes, but none of them reflected the contemporary
spoken idiom. Writers of private documents such as letters, on the other hand, often used
a language much closer to that of ordinary conversation and so reveal to us the changes
that their language had undergone. Of course, such was not their intent; writing that sim­
ply reflected speech was stigmatized as a mark of lack of education, and the writers of
papyrus documents usually tried hard to use classical spelling and grammar. It is their
failures in this respect that have the most to teach us.

Literary papyri offer a different kind of insight into language. Some contain glossaries or
linguistic commentary on particular texts, while others have been annotated with accents
or other marks that enable us to know how their writers understood particular words and
phrases. Such marks may also tell us how their writers pronounced words or thought they
should be pronounced in classical usage.

Greek Phonology
Since the fifth century BCE profound changes have occurred in the way the Greek lan­
guage is pronounced.2 Ancient Greek was on the whole written phonetically at the time
the alphabet was originally introduced, and for a few centuries thereafter spelling tended
to change to keep up with pronunciation, as is evident from surviving archaic inscrip­
tions. After the classical period, however, spelling was largely standardized, and much of
that standardization has persisted even to the present day despite major changes in pro­
nunciation. (Something similar has occurred in English, where spellings standardized
centuries ago are still the norm, but the English spelling system differs from the Greek
both in having been standardized for a much shorter period of time and in having been
less phonetic to begin with.) Thus, the letters ι, η, and υ, which originally represented
very different sounds, are all pronounced the same way in the modern language, and the
original diphthongs η, ει, oι, and υι are also pronounced like ι, η, and υ.

Before the discovery of the papyri there was considerable debate about the history of the
Greek sound system. When Western scholars first rediscovered (p. 151) ancient Greek,
they initially shared the assumption of their Greek contemporaries that the ancient lan­
guage must have been pronounced like modern (i.e., Renaissance) Greek. Soon, however,
examination of ancient onomatopoeic words and of transliterations between Greek and
languages written in other scripts (such as Latin and Aramaic) made it clear that Greek is
Page 2 of 22
The Greek and Latin Languages in the Papyri

no exception to the general rule that the pronunciation of languages changes over time.3
Progress in determining the starting point of the changes, the sound system of the fifth
century BCE, was rapid, and the stage they had reached by the early modern period (i.e.,
the Greek of Erasmus's contemporaries) had long been known. But until significant num­
bers of papyrus documents became available for linguistic analysis, scholars had very lit­
tle understanding of how and when the different changes took place.

The writers of papyrus documents were not trying to revise the spelling system or to
write phonetically; they attempted to use classical spellings, just as writers of modern
English attempt to use our standardized spelling system. But when the same sound is
spelled differently in different words (as is usually the case in a language where sound
change has led to the merging of two originally distinct sounds), poor spellers have a ten­
dency to confuse them and write one where they intend the other. The misspellings thus
produced can provide valuable clues about homophony to someone without direct access
to the spoken language. Thus a linguist who had never heard English spoken but pos­
sessed a large corpus of letters written by a representative cross-section of our society
would quickly conclude from the errors they contained that the same sound may be rep­
resented in English by both c and s or by both ie and ei.

Fortunately for us, many Greek papyri were written by poor spellers, and their errors pro­
vide the vast majority of our evidence for the pronunciation changes and their chronolo­
gy. Unfortunately, the interpretation of the evidence is not always straightforward. When
a spelling confusion is widespread, it clearly shows that the sound change causing it had
taken place in the community concerned. But what do sporadic misspellings prove? They
could be the few clues provided by a generally well-educated group to changes that had
indeed taken place but were almost never visible in writing because of the writers' high
level of education. Or they could result from very different causes: a slip of the pen, a per­
sonal idiosyncrasy, a foreigner's imperfect command of Greek. Sporadic misspellings are
not uncommon in the papyri, and when they suggest sound changes that clearly have not
taken place in modern Greek, they are usually dismissed as not representative of the spo­
ken language. When rare misspellings are early examples of changes that later become
widespread, however, it is harder to know how seriously to take them. There are thus sev­
eral possible datings for most Greek sound changes, one the point at which the first mis­
spellings indicating that change are documented, the other the point at which such mis­
spellings become common; these points may differ by several centuries.4

(p. 152) Vowels

The greatest changes involved the vowel system. During the Ptolemaic period the long
diphthongs, ᾳ, ῃ, and ῳ came to be pronounced like the simple long vowels ᾱ, η, and ω.
The ι of these diphthongs, which is now written subscript following a Byzantine conven­
tion, was written as a full letter in antiquity. After it ceased to be pronounced it was often
omitted in writing, but sometimes it was written hyper-correctly after a long vowel that
had never been a diphthong (e.g., ἐρωτὣι for ἐpωτὣ, “I ask”). The fact that the mis­
spelling goes in both directions is the best proof of the complete merger of the sounds

Page 3 of 22
The Greek and Latin Languages in the Papyri

concerned. Also datable to the Ptolemaic period is the merger of ει and ι, which allowed
either one to be written for the other (e.g., εἵνα for ἵνα, ἰς, for εἰς).

The dating of the other vowel changes is disputed, as the papyrus evidence is not conclu­
sive (and occasionally conflicts with evidence from transliterations, particularly those
made elsewhere in the Greek-speaking world). A loss of distinctions of vowel length led to
the merger of ο and ω; the diphthong oι merged with υ, and the short diphthong at
merged with ε, so that, for example, the verb endings (infinitive) and - εσθε (second-per­
son plural) became confusable. These changes are normally dated to the second century
CE, when the relevant misspellings become common, but it has also been argued that
they started in the Ptolemaic period. Also traditionally dated to the Roman period is the
merger of η and ι, but this change could be dated to the Byzantine period instead. Some
Roman-period papyri also show confusion between υ and ι; it is, however, thought that
this change may have been a peculiarity of Egypt that was not generalized in other parts
of the Greek-speaking world until the Byzantine period.

Another vowel change of disputed date is the shift in the nature of the accent from a pitch
accent to a stress accent. It is generally believed that this change was related to the loss
of distinctive vowel length and the confusion between ο and ω, though there is disagree­
ment about which change brought about the other. It is therefore likely that the accent
shift occurred during the Roman period, but arguments in favor of an earlier dating have
also been advanced. The change of the accent is not directly reflected in writing: not only
are accents not normally written in documentary papyri, but when they were written, the
same accent marks were used before and for many centuries after the accent shift. Nev­
ertheless, the presence of a stress accent can sometimes be detected in late written texts
by the omission of unaccented vowels, a process that can be caused by stress accents but
not by pitch accents.

Consonants

Many of the consonants also changed their pronunciations radically, but because they re­
mained distinct from one another, these changes did not on the whole lead (p. 153) to
spelling confusions. Papyrus evidence is thus less important for our understanding of the
consonant system than of the vowel system. Evidence for pronunciation changes affecting
consonants can be found in transliterations, as when Latin or other foreign words are ren­
dered into Greek; occasionally entire passages of Latin are written in Greek script by
writers who knew how to speak Latin but not how to write it. Latin pronunciation, of
course, also changed over time, and sometimes the same spellings can serve as evidence
for either language or both: for example, when the Greek letter β is used to represent the
Latin consonantal u, it tells us either that the Greek letter was no longer pronounced like
our b or that the Latin letter was no longer pronounced like our w or both.5

The major consonant changes were the transformation of the voiced stops β, γ, and δ to
voiced fricatives (e.g., δ pronounced like the th in there rather than like our d) and of the
aspirated stops θ, φ, and χ to voiceless fricatives (e.g., φ pronounced like our f rather than

Page 4 of 22
The Greek and Latin Languages in the Papyri

our p + h). Double consonants ceased to be pronounced differently from single ones, so a
general confusion of single and double letters (e.g., λ for λλ and vice versa) occurred. All
these changes probably took place during the Roman or early Byzantine periods, but an­
other one, that of ζ from a double consonant zd to a single fricative like our z, was proba­
bly Ptolemaic.

One change that did affect spelling was the weakening of the pronunciation of final -v. In
modern Greek ν has tended to disappear at the ends of words, so that the nominative and
the accusative of second-declension neuter nouns ends in -o, as does (usually) the ac­
cusative of masculine second-declension nouns. This process was not completed until af­
ter the period represented by the papyri, but the weakened pronunciation is reflected in
the fact that final -v is sometimes omitted in the papyri.

Initial h, represented by a rough breathing in modern texts of ancient Greek and in many
literary papyri, was not normally written in documentary papyri. Certain final consonants,
however, were aspirated if the following word began with h; thus, one would expect οὐκ
αὐτός but οὐχ οὗτος (similarly κατ᾽ and καθ᾽, ἀπ᾽ and αφ᾽, etc.). During the Roman peri­
od the writing of such aspiration becomes erratic, with οὐκ and so on not infrequently ap­
pearing before words that originally started with h- and οὐχ and so on before those that
originally did not. Such spellings suggest that the writers no longer pronounced h at all
and that their aspirations were the result of training in spelling (or, in the case of incor­
rect aspirations, hypercorrection). Something similar happens with Greek words bor­
rowed into Coptic: hori, the Coptic equivalent of h, may be either used or omitted both
where classical Greek had an h and where it did not, but the correct use of hori is fre­
quent enough to indicate that, even as late as the third century CE, educated writers
were still aware of which words began with h in classical Greek.

(p. 154) Greek Morphology and Syntax

During the period for which we have papyrus evidence, the complex grammar of classical
Greek underwent a radical process of simplification.6 Some of that simplification was di­
rectly related to the phonological changes just described, for the vowel mergers some­
times caused different grammatical forms to be pronounced the same. For example, the
vowel changes caused frequent homophony between the aorist subjunctive and the future
indicative, and this in turn led to confusion of the two even in verbs where they had re­
mained phonologically distinct.

At the same time, none of the changes affecting Greek grammar were universal. Just as
the writers of some papyrus documents were sufficiently educated to avoid betraying
their pronunciation by misspelling words, so there were some, even at late periods, who
displayed a command of grammatical forms that had long since disappeared from the lan­
guage of the less educated.

Page 5 of 22
The Greek and Latin Languages in the Papyri

The morphology of nouns and adjectives was eventually simplified along two dimensions:
a reduction in the number of distinct cases commonly used and a decrease in the number
of different endings for each case. The case that was ultimately lost was the dative, which
had more or less died out by the Byzantine period. It is generally believed that the papyri
show early evidence of this loss, and sometimes (particularly in the later Roman period)
one finds other cases used where one would have expected a dative in classical Greek.
But for most of the period for which we have papyrus evidence, datives are in fact com­
mon in documentary papyri. The reduction of endings is more obvious in our documents;
it was efected partly by the replacement of morphologically difficult words with ones of
more regular formation (e.g., παȋς, παιóς was largely supplanted by παιδίον, παιδίου,
which lost its diminutive force) and partly by the reconfiguration of the paradigms of ex­
isting words.

Third-declension nouns and adjectives often borrowed endings from the first and second
declensions; this tendency eventually led to the almost complete loss of the third declen­
sion as a separate entity, though this final stage postdates the papyri. The most common
such change is the addition of -oυ to the -α of the third-declension accusative singular
(e.g., μητέραν as the accusative of μήτηρ). A genitive singular in -ου is also attested, par­
ticularly for words that originally had a genitive in -ους (e.g., ἔτov as the genitive of ἔτος).
Third-declension names in -ης were particularly likely to be declined like first-declension
names in -ης; thus, one finds Σωκράτου, Σωκράτηι, and Σωκράτην as oblique forms of
Σωκράτης, alongside the more traditional Σωκράτους, Σωκράτει, and Σωκράτη. In addi­
tion, the third-declension nominative plural ending -ες started to spread to the accusative
plural (e.g., τὰς γυναίκες), where it would eventually replace the classical -ας completely.

There was also a general tendency for unusual noun and adjective paradigms to be regu­
larized and for the different declensional types within each declension to merge with one
another. Thus, the Attic second declension (e.g., νεώς “temple”) largely (p. 155) disap­
peared in favor of the normal second-declension inflection (e.g., νᾱός), and in the first de­
clension, nouns and adjectives in -α may appear with stems in -η (e.g., μοίρης as genitive
of μοἳρα, μικρἣς as genitive of μικρά), or occasionally vice versa, while the distinctive -ου
genitive of the first-declension masculines sometimes disappears. Contracted nouns and
adjectives are occasionally found in their uncontracted forms (e.g., χάλκεος for χαλκοὓς).
In the third declension one finds a preference for more regular forms such as χάριτα
rather than χάριν as the accusative singular of χάρις and for regularizations such as
μέρος instead of μέρους as the genitive singular of μέρος.

Two-termination adjectives may have a separate feminine in the papyri. Thus where Attic
has masculine/feminine φρόνιμος, neuter φρόνιμον, papyri sometimes have φρόνιμος,
φρονίμη, φρόνιμον. On the other hand, the masculine forms of active participles are often
used for the feminine in papyri (e.g., ἡ ὁμολογὣν for η ὁμολογοὓσα7). Though these
changes appear to go in opposite directions, both are fundamentally regularizations, as
the first assimilates a minority group to the larger category of first- and second-declen­

Page 6 of 22
The Greek and Latin Languages in the Papyri

sion adjectives, and the latter does the same with a minority group of third-declension ad­
jectives.

Comparatives and especially superlatives are very common in the papyri, especially in the
Byzantine period, indicating that their marginalization in modern Greek was a later phe­
nomenon. But they are not always formed as they would have been in the classical period.
Irregular comparative and superlative forms are often replaced by formations in -τερος
and -τατος (e.g., ἀγαθώτατος instead of ἄριστος as the superlative of ἀγαθός), and it was
also possible to add these regular suffixes to the existing irregular forms (e.g.,
μειζότερος, μεγιστότατος instead of μείζων, μέγιστος as the comparative and superlative
of μέγας).

Most of the simplifications occurred in the verb system, however, as this was more com­
plex to begin with. Once again simplification involved both a reduction in the number of
different grammatical forms each verb could have and a decrease in the number of differ­
ent endings employed by various verbs to indicate those forms. The complexity of the ver­
bal system as a whole was diminished by the decline of the optative mood, middle voice,
and perfect tense. These losses were gradual, and none of them was completed within the
period of the papyri; even the optative, which started to disappear well before the end of
the Ptolemaic period, is sometimes used by well-educated writers of documentary papyri,
and middle forms are not infrequent (though they sometimes need a reflexive pronoun to
clarify their meaning). The loss of the perfect was particularly complex. It began in the
Ptolemaic period with a loss of the distinction in meaning between aorist and perfect, so
that either tense could be used in place of the other. For a while this merger resulted in
an increased use of the perfect, but eventually the aorist, which had always been more
common, prevailed, and the perfect largely disappeared from use. A periphrastic forma­
tion using a perfect or an aorist participle and the verb “be” came to fill the function orig­
inally performed by the old perfects, and the beginnings of that development are evident
in some papyri.

The morphological alterations of the moods and tenses that remained were exten­
(p. 156)

sive. Verbs with irregular principal parts often developed new, more predictable stems
(e.g., ἦξα and ἔλειψα instead of ἤγαγον and ἔλιπον as aorists of ἄγω and λείπω), a process
that largely eliminated types like contract futures and root aorists (i.e., the ἔβην type).
There was widespread amalgamation of endings, so that both first and second aorist end­
ings were used for both types of aorist and for the perfect and imperfect. Thus, one finds
forms like ήλθα and «λαβα (for second aorists ηλθον and ελαβον), έλεγαν (for imperfect
third plural ἔλεγον), ἔγραφες (for first aorist εγραφας), and δέδωκες and εἴληφαν (for per­
fects δέδωκας and εἰλήφασι). The equivalence in meaning between perfect and aorist led
to widespread morphological confusion, so that, in addition to the confusion of endings
with both the aorists, perfects could be formed with an augment instead of reduplication
(and sometimes aorists with reduplication instead of an augment), and so forth. The dif­
ferent types of contract verbs also showed a tendency to be conflated, particularly by the
spread of the -εω endings to the -αω verbs (e.g., ἀγαπoὓμεν for ἀγαπὣμεν), and the num­
ber of -οω contract verbs increased. In addition, the athematic (-μι) verbs started taking

Page 7 of 22
The Greek and Latin Languages in the Papyri

the same endings as the thematic (-ω) verbs, a tendency that eventually led to the com­
plete loss of their distinctive conjugation.

Augments are sometimes omitted; this tendency was no doubt fueled by phonological
changes such as the loss of distinction between ο and ω, but it extended into verbs for
which no phonological explanation is possible. Confusion over the use of the augment al­
so led to the occasional use of augments on forms where they did not belong (e.g., sub­
junctive ἠξήλθῃς for ἐξέλθῃς).

The future had particular difficulties. Phonological changes led to a homophony of future
forms such as πέμφεις, πέμφει, and πέμφομεν with aorist subjunctives such as πέμφγῃς,
πέμφῃ, and πέμφωμεν (the first person singular πέμφω had been homophonous from the
beginning). In some verbs the future and the aorist had different stems, but the tendency
to regularize principal parts reduced the number of those verbs. Because of ambiguity,
the old future forms became difficult to use and were eventually replaced by periphrases
with auxiliaries such as μελλω or, much later, θέλω ἵνα plus subjunctive.

Though the overall decline in the use of non-indicative forms meant that morphological
confusions were more immediately obvious in the indicative than elsewhere, the changes
in non-indicative forms were even greater than those in the indicative. Present or future
indicative forms were often used where we would expect subjunctives, seemingly indicat­
ing a genuine decline in the use of the subjunctive as well as phonological confusion. Im­
peratives underwent widespread alteration. Infinitive forms were particularly liable to
confusion, so that endings appropriate to one tense are frequently found on stems appro­
priate to another tense, and a number of different tenses started to form an infinitive in
-ει or -εν (e.g., πέμφεν as [aorist?] infinitive of πέμπω).

The morphological changes in the Greek of the papyri are often considered to be purely
simplifications of the classical language, and as we have seen this generalization (p. 157)
is largely true. Nonetheless, not every change was a simplification. One that went in the
other direction was the creation of a new class of nouns in -ις (masculine) or -ιν (neuter),
genitive -ιoυ, dative-ιῳ, and accusative -ιv. These were formed by dropping the ο from
second-declension nouns in -ιος and -ιov; some have survived into modern Greek as nouns
in -ι. Thus, the title κύριος can be found with the nominative κὓρις, accusative κὓριv, and
vocative Kὓρι, and ἀργύριον can become ἀργύριν.

Greek Vocabulary
The words that occur in documentary papyri are often unfamiliar to readers used to clas­
sical Attic. Some of the difference is an inevitable result of the subject matter: Egyptian
documents often refer to physical, social, or political matters that did not exist in classical
Athens and for which there were, consequently, no Attic words. Not all the new vocabu­
lary can be explained in this fashion, however. Some classical words (particularly those
that posed morphological difficulties) fell into disuse and were replaced by existing syn­
onyms or by newer, often morphologically more regular, terms. Other ancient words sur­
Page 8 of 22
The Greek and Latin Languages in the Papyri

vived but changed their meanings significantly. In many cases the vocabulary shifts visi­
ble in the papyri also turn up in low-register literature such as the New Testament, and
not infrequently their results have persisted into modern Greek.

Thus, in the papyri, classical βούλομαι (“want”) is marginalized by its synonym (ἐ)θέλω,
while αἰσχύνομαι (“be ashamed”) is starting to be replaced by the newer ὲντρέπομαι.8
The classical word for “one,” εἷς, can be used like an indefinite article (“a,” “an”), and the
definite article (ὁ, ἡ, τό) is often used for the relative pronoun (ὅς, ἥ, ὅ). The pronoun
αὐτός, which in classical Attic had an intensifying force in the nominative and so could be
used as a simple anaphoric pronoun (i.e., as an equivalent of “him,” “her,” “them,” etc.)
only in the oblique cases, came to be used anaphorically in the nominative as well. Some
distinctions were lost; for example, the pronoun ἕτερος, which originally meant “the other
of two,” became indistinguishable from ἄλλος, “other” of any number.

Prepositions are used more often in the papyri than in classical Greek and in new ways;
some constructions that were originally expressed without a preposition, such as the par­
titive genitive or the genitive of price, are likely to use a preposition in papyrus docu­
ments. Thus, ἀπό plus genitive can replace the partitive genitive; ἐκ plus genitive, the
genitive of price; διά plus genitive, the dative of means, and so on. There is also a loss of
distinction between certain prepositions, so that είς can be used for ἐν, ἀπό for ἐκ, διά for
περί, and so forth.

Many of these changes were the result of the language's natural development, which
would have resulted in vocabulary changes even if Greek had been entirely (p. 158) isolat­
ed, but there were also pressures on postclassical Greek from the other languages with
which it was in contact. The papyri show considerable evidence of influence from these
languages. One major source of new vocabulary, particularly in the Ptolemaic period, was
the local Egyptian language. Though the vast majority of preserved papyri are written in
Greek, and though it is clear that Greek was the dominant language of both administra­
tion and culture beginning with the Macedonian conquest of Egypt, many Egyptian terms
had to be taken into the language in order to describe preexisting physical and cultural
realities. In addition to Egyptian personal and place names, we find Egyptian months,
Egyptian deities, and so on.

An even larger, but somewhat later, source of new vocabulary was Latin. Latin loanwords
start appearing in the papyri in the second century BCE and become common after the
Roman conquest in the later first century BCE; at their peak in the fourth and sixth cen­
turies CE there is an average of one attestation of a loanword per preserved documentary
fragment. Although Greek remained the language of most provincial administration after
the Roman conquest, the presence of Roman officials and in particular of the Roman army
introduced a large body of new terminology. This included titles such as Αὔγουστος (Au­
gustus), military designations such as κεντυρίων (centurio), and eventually currency
terms such as δηνάριον (denarius). The writers of some papyri, particularly in the later
Roman period, used Latin loanwords even when a common Greek alternative was avail­
able: βέστη (vestis), πραἳδα (praeda), and ὅσπες (hospes).9

Page 9 of 22
The Greek and Latin Languages in the Papyri

In many cases, however, the external influence on Greek was more subtle. Many Roman
concepts were expressed either by the creation of new words from Greek roots and affix­
es or by the adaptation of older Greek words to new uses. Some of the most important Ro­
man concepts had two representations in Greek, one a direct borrowing and one a Greek
adaptation; thus we also find σεβαστός for Augustus and ἑκατόνταρχος for centurio. (For
many words the Greek forms appear earlier than the direct borrowings and are common­
er in the early imperial period, while the direct borrowings become more prevalent in lat­
er centuries, but this pattern is not always followed.) Other concepts were always repre­
sented primarily by an adapted Greek word, such as ὕπατος for consul or the address
κύριε for domine.10

A Sample Greek Document

This letter (P.Oxy. LXVII 4627) was written in the late third century CE and found at
Oxyrhynchus; we know nothing about either the writer or the recipient beyond what we
can deduce from the letter itself. Word divisions, capitalization, punctuation, accents, and
breathings are editorial additions.

(p. 159)

To my lord brother Hierakapollon, Serenus [sends] greetings. I am surprised how you have
stayed at your own place until today. For you threw me into not a little anguish by doing this. So
even now [i.e., late as it is] either come quickly to me or write to me what the delay is, and at
least before all [write] about your [pl.] health and about what you need [from?] here. I greet my
lady sister and my lady mother and all our people. I pray that you [pl.] be well for many years.
[Written on the] 20th [day of the month] Pachon.

Sarammon [writes] to my lord brother Hierakapollon.

The letter was written by a relatively well-educated man (note his use of particles such as
γε and γάρ, which are not common in papyrus letters); nevertheless, it shows several
spelling errors characteristic of papyrus documents. Iota adscript is usually missing
Page 10 of 22
The Greek and Latin Languages in the Papyri

(κυρίω(ι)and ἀδελφὣ(ι) in line 1, σεαυτὣ(ι) in lines 3–4, etc.), but it is written correctly in
one of the two datives in the address (ἀδελφὣι, but note the lack of iota in κυρίω(ι) two
words earlier). There is also an interchange of αι and ε (χρήζεται for χρήζετε in line 11)
and a morphological slip (ἐνέβαλας for second aorist ἐνέβαλες in line 5). The σ of σήμερον
is not an error but one of the Ionic elements characteristic of the koinê dialect; the Attic
form would be τήμερον.

But spelling and morphology are only a small part of what makes this letter immediately
recognizable as a papyrus document rather than a classical one. The biggest clue is found
in the vocabulary. The writer uses χρόνος to mean “year” rather than “time” (line 15),
καταλαμβάνω to mean “arrive at” rather than “seize” (line 7), the Roman-period
creation11 βράδος for “delay” (line 8), and the Egyptian month-name Παχών to give a date
according to the Egyptian calendar (line 16). The letter's recipient is given the title
κύριος in both the heading and the address, and the two women to whom greetings are
sent are both called κυρία (line 12), though in the classical period, titles would not nor­
mally have been used even to (p. 160) people of considerably elevated status, let alone to
ordinary citizens.12 Moreover, in the classical period κύριος had a few specific meanings,
such as the guardian of a woman, and its general use as a polite term is a phenomenon
characteristic of the Roman period (see Dickey 2004a).

The kinship terms used in this letter are likewise an indication of its date. In the classical
period kinship terms tend to be used alone; their combination with names (as in the head­
ing and address) and with titles (as in every instance in this letter) is a late feature. More­
over, there is a significant difference between the ways kinship terms are used in classi­
cal Greek and in papyrus letters: in the classical language ἀδελφός and αδελφή really
mean “brother” and “sister,” and μήτηρ means “mother,” but in papyrus letters it is not
uncommon for all these terms to be used in a more generalized sense for people with no
genetic relationship to the writer. In particular, ἀδελφή can be used to friends and even
distant acquaintances without conveying particular affection or intimacy. We know noth­
ing about the people named in this letter beyond what we can deduce from the letter it­
self, so in this case it is possible that all the kinship terms indicate blood relationship, but,
given the letter's date, such relationships cannot be assumed simply on the basis of the
kinship terms (see Dickey 2004b).

This letter also has a number of syntactic features that mark it as a papyrus document.
The dative is used instead of the accusative for extent of time (line 15). The κἄν in line 6
would in the classical period have been crasis for καὶ εάν, but in Hellenistic and later
Greek it can be used simply as an equivalent for καί. The writer has hesitated over the
construction to use after θαυμάζω in line 3. Originally he simply followed it by a sentence
in the indicative, without any conjunction, but then went back and added πὣς (“how”). In
the classical period the normal construction with θαυμάζω would be an indirect statement
introduced by εἰ (normally meaning “if” but in this construction best translated as “that”);
θαυμάζω ὅπως is possible but rare. The use of βράδος (“delay”) where a classical writer
would probably have preferred a verb is connected to a general tendency of the postclas­
sical language, both literary and subliterary, to use more nouns and in particular more ab­

Page 11 of 22
The Greek and Latin Languages in the Papyri

stract nouns than classical authors employed. The phrase μέχρι τήμερον is unattested in
classical Attic, though similar phrases such as μέχρι νυνί and εἰς τήμερον occur.13

The word order is also somewhat different from what one would expect to find in the clas­
sical period. The nonclassical particle combination γε δέ occurs in lines 8–9; a classical
author might have used δέ γε, but since that combination is not found in contexts like the
one here (see Denniston 1950, 152–156), most likely a classical writer would not have
used both these particles in this passage. Also, the heading begins with the addressee's
name, followed by the sender's, while the standard order for letter headings in the classi­
cal, Ptolemaic, and Roman periods puts the sender before the addressee. The reverse or­
der is a late antique phenomenon of which this letter is a relatively early example; the
change seems to have been motivated by considerations of politeness.

In addition to the features that unambiguously mark this letter as a product of its
(p. 161)

time and place, it contains a number of more doubtful elements, ones that were or might
have been used in classical Attic but that are nevertheless slightly surprising for one rea­
son or another. The particle γάρ would normally come in second position in its clause, but
its use here (line 5) as the fourth word is not unparalleled, especially in poetry (see Den­
niston 1950, 95–98). Similarly, ὑγιαίνειν is used for “to be well” far more often than
ἐρρὣσθαι in classical Attic, though the latter is also attested. The aorist participle
ποιήσας (line 6) seems odd for an action that is both continuous in aspect and simultane­
ous in time with that of the main verb, so one might have expected the present ποιὣν
instead, but the aorist participle could be defended as complementing the aorist main
verb.

The last two sentences of this five-sentence letter consist of courtesies: sending greetings
to two women and wishing the addressee continued health. Both these features are very
common in papyrus letters; indeed some letters consist almost entirely of good wishes
and greetings to long strings of friends and relatives. Such courtesies are considerably
less prominent in the classical period, and this rarity can only partially be ascribed to a
scarcity of preserved letters from that period. The Platonic epistles, those preserved with
the works of Demosthenes, Isocrates, and Hippocrates (not that all of these are authentic,
but in most cases they are probably not a great deal later than the period at which they
purport to have been written), the letters embedded in the works of various historians,14
and the few actual private documents that survive from an early period15 make it clear
that the culture of elaborate epistolary courtesy reflected in Roman-period letters did not
exist in the fifth and fourth centuries BCE. Most letters from that period, if they contain
any good wishes at all, have them in a one-word farewell such as ἔρρωσο (“be well”) or
εύτύχει (“be fortunate”); more elaborate courtesies and greetings to third parties are
rare.16

The sender's name is Serenus in the heading but Sarammon in the address. It is possible
that two different people are involved but more likely that the sender had several names.
Classical Greeks had only one name; if that did not suffice for precise identification, a
patronymic or demotic could be added. After the Roman conquest, however, Greek men

Page 12 of 22
The Greek and Latin Languages in the Papyri

who acquired Roman citizenship adopted the Roman system of using three names, and in
the imperial period even noncitizens often had multiple names.

This letter might have looked as follows had it been written by a well-educated fifth- or
fourth-century Athenian:

Serenus to Hierakapollon, greetings.

I am surprised that you have stayed at home until now, for you threw me into not a little
anguish by doing this. And now therefore either come quickly to me or (p. 162) write to
me why you are delaying, and before all [write] about your [pl.] health and about what
you need from us. Farewell.

The Latin of Papyri and Ostraca

The usual written language in Egypt was Greek, and therefore we have comparatively few
documents in Latin; we also have fewer document types in Latin than in Greek.17 Among
the more common types of Latin document are army communications and private letters;
we have a number of army documents and short letters on ostraca and a few longer let­
ters on papyrus. Six of the latter group were written by one person, a bilingual soldier
named Claudius Terentianus, who wrote to his family in the early second century CE. All
these letters provide evidence for Latin as spoken under the Roman Empire, often re­
ferred to as Vulgar Latin; other evidence for Vulgar Latin comes from curse tablets, graf­
fiti (especially at Pompeii), and numerous late Latin writings.

The Latin of papyrus letters contains misspellings that reveal a variety of phonetic
changes similar to but less extensive than those visible in Greek papyri. In postclassical
Latin, as in postclassical Greek, the phonemic distinction between long and short vowels
disappeared. Since Latin had never had separate letters to indicate long vowels, this shift
is not usually apparent in writing, but it was partially responsible for one visible spelling
error: confusion between long e and short i. This confusion produced spellings like nese
and nesi for nisi and dicet for dicit. Only one diphthong underwent substantial monoph­
thongization in the period of the papyri: ae, which is often written e, as in magne for mag­
nae or Alexandrie for the locative place name Alexandriae.

One can also find in the papyri spellings with short o where classical Latin had short u,
such as con for cum and nouom for nouum, and it has sometimes been suggested that
these spellings anticipate the late Latin confusion of o and u. But the words so spelled in
papyri are ones that were written o rather than u in both early and very late Latin, so
these papyrus spellings are more likely to be due to archaism than to anticipation of the
late change.18

Page 13 of 22
The Greek and Latin Languages in the Papyri

The Latin consonant changes also paralleled those of Greek to some extent. Final -m,
which had disappeared from educated speech by the second century BCE, is frequently
omitted in writing, producing forms like scriba for scribam, unu for unum, and minore for
minorem; sometimes it is added by hypercorrection where it does not belong, as factam
for facta or ducerem for ducere. Final -t may also be lost, as in uendedi for uendidit.
Intervocalic stop consonants may be voiced, as tridicum for triticum (Adams 1994, 108).
The letter h, which had early ceased to be pronounced in some varieties of Latin, is often
omitted in papyri, as for example in mi (p. 163) for mihi and abiturum for habiturum. The
sound traditionally written with b, which was originally pronounced like our b, become a
fricative similar (but not identical) to our v, while the consonant traditionally written with
u, which was originally pronounced like our w, changed to the same fricative. The result­
ing merged sound could be written with either letter, but b was more common: thus we
find bolt for uult, negabit for negauit, bia for uia, and benio for uenio.

Final -n was normally assimilated in pronunciation to the initial consonant of the follow­
ing word, even in educated speech (cf. Cicero, Fam. 9.22.2), but in standard Latin orthog­
raphy the assimilation was not reflected in writing. At all periods writers with ortho­
graphic preferences other than those that later became standard produced assimilated
spellings, for example, im perpetuo for in perpetuo; this tendency is particularly obvious
in papyrus letters because of the low educational level of their writers. Also reflecting a
widespread feature of standard pronunciation is the confusion between final -t and -d in
certain short words, leading to the use of each letter for the other. Thus we find not only
phrases like ed domino for et domino or aput te for apud te but also ones like ud continuo
for ut continuo or ed pater for et pater.

The Latin of the papyri, like the Greek, differs from the classical language in morphology
and syntax as well as in spelling. For example, the accusative is often found with preposi­
tions that would take the ablative in the classical language, such as con tirones for cum
tironibus. The comparative can be used where we would expect the positive, as in celerius
for celeriter. And writers show some difficulties with verb endings, particularly in the fu­
ture; many of these problems resulted from the sound changes that made endings like -bit
and -ēs homophonous with ones like -uit and -is. Writers also tended to use the indicative
instead of the subjunctive in certain constructions, such as indirect questions. But overall
the Latin of the papyri differs less from classical Latin than the Greek of the papyri does
from classical Greek in morphology and syntax as well as in phonology because of the
shorter chronological gap between the two phases of the language being compared. Of
course, one does not find Ciceronian oratory in soldiers' letters, but that is a question of
register rather than date: even Cicero's own letters use a markedly more colloquial style
than his speeches, and even in Cicero's day Romans with less education would have writ­
ten their letters in less perfect Latin than his.

Word order in papyrus letters also differs from that in classical Latin, particularly in the
later period. In the classical language the verb tends to come at the end of its clause, but

Page 14 of 22
The Greek and Latin Languages in the Papyri

in later Latin it normally comes earlier—before its object and any associated preposition­
al phrases.

The vocabulary of papyrus letters also diverges from that of classical Latin. As is only to
be expected, given Latin's minority status in Egypt, Greek loanwords are abundant; per­
sonal and place names, deities, and so on may come from either Greek or Egyptian. Na­
tive Latin words may be used differently in Vulgar Latin from in the classical language
(e.g., fortis for “healthy”). Moreover, writers of papyrus (p. 164) letters sometimes prefer
one word for an object where writers of literary texts preferred another (e.g., preferring
epistula to litterae). Sometimes words and usages found in the papyri or ostraca also oc­
cur in much earlier low-register texts and can be identified as persistent colloquialisms
rather than postclassical innovations (e.g., adiuto instead of adiuvo).

A Sample Latin Document

This letter (P.Oxy. XLIV 3208), which was found at Oxyrhynchus, was probably written in
the Augustan age.19 It is thus only slightly later than Cicero's letters and earlier than
Seneca's and Pliny's. Though it contains nonstandard spellings and other unusual fea­
tures, it is on the whole carefully written by someone with a certain amount of education.
Word divisions are indicated on the papyrus, but capitalization and punctuation are edito­
rial additions.

Suneros Chio suo plur(imam) sal(utem). s(i) u(ales) b(ene). Theo adduxsit ad me
Ohapim,

regium mensularium Oxsyrychitem, qui quidem mecum est locutus de inprobitate

Epaphraes. itaque nihil ultra loquor quam [[no]] “ne patiarus te propter illos
perire.” crede mihi, nimia bonitas

5 pernicies homin[i]bus est ‘uel maxsuma’. deinde ipse tibei de mostrabit qu[i]t rei
sit qum illum ad te uocareis. set perseruera: qui de tam pusilla summa tam mag­
num lucrum facit, dominum occidere uolt. deinde ego clamare debeo, siquod
uideo, “deuom atque hominum [[fidem.” si tu [.] ista non cuibis]]

10 tuum erit uindicare ne alio libeat facere.

[address] Chio Caesaris

Suneros [sends] very many greetings to his own Chios. If you are well, [that's]
fine. Theo brought to me Ohapim, the public banker of Oxyrhynchus, who spoke
with me about the wickedness of Epaphras. And so I say nothing beyond “Don't al­
low yourself to be ruined on account of them.” Believe me, excessive generosity is
a source of disaster for men, altogether the biggest [source]. Then he himself will
show you what it's about when you call him to you. But be persistent: someone
who makes such a big profit from such a trifling sum is willing to kill his master.

Page 15 of 22
The Greek and Latin Languages in the Papyri

Then I ought to shout, if I [correctly] perceive anything, “by the faith of gods and
men!” Punishment will be up to you, lest someone else should want to do it.

To Chios, slave of Caesar.

In many ways the language of this letter is close to classical Latin: it follows classical word or­
der; the verbs have their full range of inflection and (on the whole) the (p. 165) usual endings;
and much of the vocabulary is standard. Though many of the words look odd to readers used to
Latin literary texts, most of the unusual appearance is due simply to the presence of alternative
spellings (also common in inscriptions) that do not indicate a pronunciation different from the
standard one. In this category fall the learned spelling xs for x (adduxsit, line 1; Oxsyrychitem,
line 2; and maxsuma, line 5); archaic ei for long i (tibei for tibi, line 5; and probably also (with the
i omitted) deuom for diuum, line 9); n omitted before s in Oxsyry (n)-chitem (line 2) and demo(n)strabit
(line 5); o written instead of u after consonantal u in uolt (line 8) and deuom (line 9); the archaic
superlative in -umus rather than -imus (maxsuma, line 5); the unassimilated n in inprobitate (line
3); qum for cum (line 6); and the transliteration of the Greek upsilon in Suneros (line 1) with u
rather than the y used in Oxsyrychitem. The word break after the initial element of demostrabit
(line 5) and the lack of one between si and quod (line 8) do not fit with our Latin orthography but
would not have been unusual in antiquity; in any case they are matters of writing rather than of
pronunciation. The interchange of final d and t, which occurs in set for sed (line 8) and quit for
quid (line 9), probably does not indicate any particular pronunciation either, as it is likely that d
and t were not distinguished in pronunciation at the ends of such words.
This letter also contains some elements of nonstandard morphology. The dative of alius is
given as alio, which would be the normal ending for a second-declension adjective, rather
than the alii of standard Latin, which follows a special pronominal declension; this type of
inflection is thought to be a colloquialism (Cugusi 1992, vol. 2, 24). The -rus for -ris
ending of patiarus (line 4) is also found in graffiti at Pompeii and has been argued to be
rustic or otherwise low register.20 The use of quod where we would expect quid (line 8) is
part of a general tendency in early and colloquial Latin not to make a rigid distinction be­
tween the use of quis, quid and that of qui, quae, quod.21 The future perfect uocareis (=
uocaueris) in line 6 uses the widely attested syncopation of -ue- in the perfect system, as
well as the ei spelling for long i.22 The -aes ending on the Greek name Epaphraes (line 3)is
a partial Latinization of the Greek first-declension genitive ending -ης; such -aes genitive
endings are distinctly low register and not uncommon in Vulgar Latin contexts. Usually
they are found on the Latin gentilicia of people with Greek cognomina rather than on the
Greek name itself, as in this case (Adams 2003, 479–483). The ending has nothing to do
with the actual Greek genitive of this word, Ἐπαφρἃ.

Some of the vocabulary is revealing as well. Pusilla is classical but colloquial and occurs
in Cicero's letters. The phrase regius mensularius, which uses a rare word for “banker,” is
a translation of the Greek term βασιλικὸς τραπεζίτης (“public banker”) (Brown 1970,
141–142).

The letter contains numerous Greek names, all of which have been Latinized where possi­
ble. Theo has been given a Latin nominative ending (a straight transliteration of the
Greek would have been Theon); Oxsyrychitem has both a Latin suffix (p. 166) and a Latin
accusative ending; and the oddly Latinized ending of Epaphraes has already been noted.
Page 16 of 22
The Greek and Latin Languages in the Papyri

Suneros and Chio are not susceptible of Latinization.23 The language from which
Ohapim's name comes is uncertain—there have even been attempts to make him into a
Jewish banker named Joachim (cf. Brown 1970, 138)—but it is probably Egyptian and in
any case clearly not Latin in origin.

This letter might have looked as follows had it been written by someone with Cicero's ed­
ucational background:

Syneros Chio suo s(alutem) p(lurimam) d(icit). s(i) u(ales) b(ene) e(st). Theo ad­
duxit ad me Ohapim, argentarium publicum Oxyrhynchitem, qui quidem mecum
est locutus de improbitate Epaphrae. itaque nihil ultra moneo quam “ne patiaris te
propter illos perire.” crede mihi, nimia bonitas pernicies hominibus est uel maxi­
ma. deinde ipse tibi demonstrabit quid rei sit cum illum ad te uocaueris. sed
perseuera: qui de tam pusilla summa tantum lucrum facit, uel dominum occideret.
deinde ego clamare debeo, si quid intellego, “pro diuum atque hominum Wdem.”
tuum erit uindicare ne alii libeat facere.

Understanding the Latin and Greek of the Pa­

pyri
Because the language of documentary papyri is so often different from the classical lan­
guages that one traditionally learns, special resources are often needed to read them. For
Greek vocabulary not covered by Liddell, Scott et al. (1940), one can look in Arndt et al.
(2000), Moulton and Milligan (1930), Preisigke (1924–1993), and Sophokles (1870). Odd
spellings can usually be explained by reference to Mayser (1926–1938) and Gignac (1976
–1981); nonstandard morphology can sometimes be found in the same sources, but one
may also need to consult Mandilaras (1973). Syntax is inadequately covered by existing
reference works, though some information is given in the works listed above, particularly
Mandilaras (1973).

Because Latin papyri are so much rarer than Greek ones, we do not have the same kind of
specialized works on papyrological Latin as on papyrological Greek. Nevertheless some
works on Vulgar Latin, although they concentrate on other types of source, are helpful in
dealing with papyri. Latin words that do not appear in the Oxford Latin Dictionary (which
deliberately excludes all sources after 200 CE) can often be found in Lewis and Short
(1879). Common Vulgar Latin spellings, morphology, and so forth are discussed by Väänä­
nen (1963) and Herman (2000), and anything that happens to occur in Terentianus's let­
ters is covered by Adams (1977).

Bibliography
Adams, J. N. (1977). The Vulgar Latin of the Letters of Claudius Terentianus (P.Mich. VIII,
467–72). Manchester: Manchester University Press.

Page 17 of 22
The Greek and Latin Languages in the Papyri

———. 1994. “Latin and Punic in Contact? The Case of the Bu Njem Ostraca.” Journal of
Roman Studies 84: 87–112.

———. (2003). Bilingualism and the Latin Language. Cambridge: Cambridge University
Press.

———. 2007. The Regional Diversification of Latin, 200 BC–AD 600. Cambridge: Cam­
bridge University Press.

Allen, W. S. (1987). Vox graeca: A Guide to the Pronunciation of Ancient Greek, 3d ed.
Cambridge: Cambridge University Press.

Arndt, W. F., W. Bauer, F. W. Gingrich, and F. W. Danker. (2000). A Greek-English Lexicon

of the New Testament and Other Early Christian Literature, 3d ed. Chicago: University of
Chicago Press.

Brown, V. 1970. “A Latin Letter from Oxyrhynchus.” Bulletin of the Institute of Classical
Studies 17: 136–143.

Cervenka-Ehrenstrasser, I.-M., with J. Diethart. 1996–. Lexikon der lateinischen Lehn­

wörter. Vienna: Hollinek.

Cugusi, P. 1992–2002. Corpus epistularum latinarum: papyris tabulis ostracis servatarum.

3 vols. Florence: Gonnelli.

Daris, S. (1991). Il lessico latino nel greco dʼEgitto, 2d ed. Barcelona: Institut de Teologia
fonamental.

Denniston, J. D. (1950). The Greek Particles, 2d ed., rev. K. J. Dover. Oxford: Clarendon.

Dickey, E. 2003. “Latin Influence on the Greek of Documentary Papyri: An Analy­

(p. 169)

sis of Its Chronological Distribution.” Zeitschrift für Papyrologie und Epigraphik 145:
249–257.

———. 2004a. “The Greek Address System of the Roman Period and Its Relationship to
Latin.” Classical Quarterly, n.s., 54: 494–527.

———. 2004b. “Literal and Extended Use of Kinship Terms in Documentary Papyri.”
Mnemosyne 57: 131–176.

Exler, F. X. (1923). The Form of the Ancient Greek Letter of the Epistolary Papyri. Wash­
ington, D.C.: Catholic University of America.

Gignac, F. T. 1976–1981. A Grammar of the Greek Papyri of the Roman and Byzantine Pe­
riods. Vol. 1, Phonology, vol. 2, Morphology. Milan: Istituto editoriale Cisalpino–La Go­
liardica.

Herman, J. 1990. Du latin aux langues romanes. Tübingen: Niemeyer.

Page 18 of 22
The Greek and Latin Languages in the Papyri

———. (2000). Vulgar Latin, trans. R. Wright. University Park: Pennsylvania State Univer­
sity Press.

Horrocks, G. (1997). Greek: A History of the Language and Its Speakers. New York: Long­
man.

Jacoby, F., ed. 1923–. Fragmente der griechischen Historiker. Berlin: Weidmann.

Jordan, D. R. 2000. “A Personal Letter Found in the Athenian Agora.” Hesperia 69: 91–
103.

Kapsomenakis, S. G. (1938). Voruntersuchungen zu einer Grammatik der Papyri der

Nachchristlichen Zeit. Munich: Beck.

Lewis, C. T., and C. Short. (1879). A Latin Dictionary. Oxford: Clarendon.

Liddell, H. G., R. Scott, H. S. Jones, and R. McKenzie. (1940). A Greek-English Lexicon,

9th ed. Oxford: Clarendon.

Löfstedt, E. (1933). Syntactica: Studien und Beitröge zur historischen Syntax des Lateins.
vol. 2. Lund: C. W. K. Gleerup.

Mandilaras, B. G. (1973). The Verb in the Greek Non-literary Papyri. Athens: Hellenic
Ministry of Culture and Sciences.

Mason, H. J. (1974). Greek Terms for Roman Institutions: A Lexicon and Analysis. Toron­
to: Hakkert.

Mayser, E. 1926–1938. Grammatik der griechischen Papyri aus der Ptolemäerzeit. Berlin:
De Gruyter.

Meissner, T. (2006). S-stem Nouns and Adjectives in Greek and Proto-Indo-European: A

Diachronic Study in Word Formation. Oxford: Oxford University Press.

Moulton, J. H., and G. Milligan. (1930). Vocabulary of the Greek Testament, Illustrated
from the Papyri and Other Non-literary Sources. London: Hodder and Stoughton.

Palmer, L. R. (1945). A Grammar of the Post-Ptolemaic Papyri. Vol. 1, pt. i. London: Oxford
University Press, G. Cumberlege.

Preisigke, F. 1924–1993. Wörterbuch der griechische Papyrusurkunden. Wiesbaden: Har­

rassowitz.

Shipp, G. P. 1967. “Some Observations on the Distribution of Words in the New Testa­
ment.” In Essays in Honour of Griffithes Wheeler Thatcher, ed. E. C. B. MacLaurin, 127–
138. Sydney: Sydney University Press.

Sophokles, E. A. 1870. Greek Lexicon of the Roman and Byzantine Periods from B.C. 146
to A.D. 1100. Boston: Little, Brown.

Page 19 of 22
The Greek and Latin Languages in the Papyri

Teodorsson, S. T. (1977). The Phonology of Ptolemaic Koine. Gothenburg: Acta Universi­

tatis Gothoburgensis.

Väänänen, V. (1963). Introduction au latin vulgaire. Paris: Klincksieck.

———. (1966). Le latin vulgaire des inscriptions Pompéiennes, 3d ed. Berlin: Akademie-
Verlag.

Notes:

(1.) I am grateful for the assistance of J. N. Adams, P. Probert, and others with the prepa­
ration of this chapter.

(2.) For a more complete description of the phonetic changes in the Greek of the papyri,
with numerous examples of each, see Gignac (1976–1981, vol. 1).

(3.) Other types of evidence exist as well, including statements by grammarians; for a
fuller discussion see Allen (1987, esp. xiii).

(4.) Exponents of the later dating, which is more widely accepted, include Allen (1987)
and Gignac (1976–1981); the earlier dating is supported by Teodorsson (1977) and Hor­
rocks (1997).

(5.) The evidence of transliterations must, however, be used with some caution. One can­
not necessarily conclude from such spellings that the Greek and the Latin letters had ex­
actly the same pronunciation at the time a given transliteration occurred, let alone that
that common pronunciation was the same as the one their respective modern descen­
dants now share (i.e., like English v). When transliterating from one alphabet into anoth­
er, one often has to use the closest equivalent for sounds that have no exact match in the
other alphabet; for example, someone who wants to transliterate “cherry” into the Greek
alphabet might use τσ to represent the ch not because he thought the Greek consonant
cluster was identical to the English sound but because he could not find anything in
Greek closer to the English sound. Alternatively, he or she might use χ because that letter
is often transliterated ch in English. Therefore, the most a transliteration like ον (the orig­
inal Greek equivalent of Latin consonantal u)or β (the later equivalent) can tell us is that
the letters used for the transliteration were closer than anything else that alphabet of­
fered to the sound being transliterated.

(6.) For a more complete description of the morphological changes in the Greek of the pa­
pyri, with citations for the examples given here and many additional examples, see Gi­
gnac (1976–1981, vol. 2).

(7.) The motivation behind this example may be formulaic usage, as the writer probably
saw ὁ ὁμολογὣν as a fixed formula.

(8.) For more such pairs see Shipp (1967).

(9.) See Cervenka-Ehrenstrasser (1996–), Daris (1991), and Dickey (2003).

Page 20 of 22
The Greek and Latin Languages in the Papyri

(10.) See Mason (1974) and Dickey (2004a).

(11.) There are two possible attestations of this word before the Roman period, one classi­
cal and one Hellenistic, but see Meissner (2006, 100–104, 185, 224–225) for arguments
against accepting their authenticity.

(12.) For example, Plato's letters to various important figures, which may be spurious but
are linguistically classical enough that knowledgeable ancient writers like Cicero thought
them genuine, mostly have simple headings like Πλάτων Διοννσίῳ εὖ πράττειν.

(13.) The former, for example, at Aristophanes, Ran. 1256; the latter, for example, at Pla­
to, Symp. 174a. Demosthenes several times uses μέχρι τἣς τήμερον ἡμέρας (e.g., 19.297,
19.328), but this belongs to a more formal register than would be appropriate in a per­
sonal letter. The writer of this letter has used a perfect tense after μέχρι σήμερον, and
this is also the tense Demosthenes uses with μέχρι τής τήμερον ἡμέρας, but both μέχρι
νννί and the more common but slightly later μέχρι (τοὓ) νὓν tend to take a verb in the
present tense (e.g., Demosthenes 19.336, Hecataeus (Jacoby 1923–, vol. 1a, author 1, fr.
119, line 13), Ctesias (Jacoby 1923–, vol. 3c, author 688, fr. 1b, line 363), Aristotle, Poetics
1447b 9).

(14.) For example, Hdt. 1.124, 3.40; Thuc. 1.128.7, 1.129.3; Xen., Cyr. 4.5.27–33.

(15.) See Jordan (2000, esp. 91–92).

(16.) For example, ἔρρωσο at Xen., Cyr. 4.5.33; and εὐτύχεί at Plato, Epistle 4 (321c). Ex­
ceptions to this generalization include a greeting in Plato, Epistle 13 (τοὺς συσφαιριστάς
ἀσπάζου ὑπὲρ ἐμοὓ, 363d).

(17.) For a more detailed discussion of most of these phenomena, citations for the exam­
ples given here, and additional examples see Adams (1977).

(18.) See Väänänen (1966, 27), Adams (1977, 9–11), and Herman (1990, 138–139). The o
spellings are particularly frequent after consonantal u and in such circumstances seem to
be a graphic phenomenon that resulted from a reluctance to write uu.

(19.) For further information see the more detailed discussions in Brown (1970) and Cu­
gusi (1992, vol. 1, 93; vol. 2, 21–24).

(20.) See Cugusi (1992, vol. 2, 19, 22) and Adams (2007, chap. 7.6).

(21.) See Löfstedt (1933, 79–96).

(22.) The i of this ending, though we normally think of it as short, is attested in classical
poetry as both long and short; see Brown (1970, 140).

(23.) The -os ending of Suneros probably reflects Greek -ως rather than -ος, so one would
not expect it to be Latinized as - us in the manner of second-declension nominative end­
ings.

Page 21 of 22
The Greek and Latin Languages in the Papyri

Eleanor Dickey

Eleanor Dickey is Associate Professor of Classics, University of Exeter.

Page 22 of 22
Abbreviations and Symbols

Abbreviations and Symbols

Nikolaos Gonis
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0007

Abstract and Keywords

The first part of this article deals with abbreviations found in Greek documentary papyri
and ostraca. The documents in which abbreviation is rife are predominantly those pro­
duced on a massive scale and bound to repeat the same words, such as tax accounts and
receipts. The most common method of abbreviation is by suspension. to omit one or more
of the final letters of a word. Most symbols stem from abbreviations by suspension; these
may become reduced to monograms whose original constituents are sometimes no longer
discernible. This is the case with most symbols that represent weights and measures, as
well as, in the later period, money, which naturally occurred very frequently. In the early
days of papyrology, Verschleifungen were given the status of a particular subgroup of ab­
breviations. This practice is predominantly found with the names and titles of emperors
and the names of months in date clauses.

Keywords: Verschleifungen, Greek documentary papyri, abbreviations, tax accounts, monograms

Abbreviations
A full collection of abbreviations in the papyri is not generally available (Blanchard's 1969
Sorbonne dissertation has not been published). The most comprehensive list one may
consult is that offered by Bilabel (1923, 2296–2303), while the indexes to P.Lond. II–V and
to the text volumes of SPP remain useful (but a good deal of caution is advisable; only
P.Lond. IV–V are sufficiently reliable). Likewise, no detailed discussion exists, and the
present chapter does not offer one (this would have required a monograph-length study).
In setting out the material, I have consciously followed to a large extent Wilcken's treat­
ment of the subject (1912, xxxix–xlvii), the Wrst systematic and perhaps also the most lu­
cid attempt to describe the phenomenon. Blanchard (1974) is fundamental, although it is
not a systematic survey but two “essays in interpretative palaeography” (Parsons 1976,
265).

Page 1 of 8
Abbreviations and Symbols

Here we are concerned with abbreviations found in Greek documentary papyri and ostra­
ca (and occasionally wooden tablets); those in literary papyri are treated in McNamee
(1981) (cf. also McNamee 1985; Bastianini 1992). Shorthand, of which we still understand
very little, is not discussed (see the bibliography in Torallas Tovar and Worp 2006).

In writing, one “abbreviates,” or shortens a word, to save effort, time, or even space. No
rules need apply, but when the writing is meant to be read by another person, (p. 171) one
has to ensure that what is abbreviated is also understood. Arbitrary choices cannot be ab­
sent, but everything usually falls under the umbrella of convention, part of the common
writing ground that is taught or otherwise disseminated from generation to generation.
We find abbreviations in writing of all kinds and ages: literary or documentary and on
stone, wood, papyrus, or leather. The documents in which abbreviation is rife are predom­
inantly those produced on a massive scale and bound to repeat the same words, such as
tax accounts and receipts. We also find plentiful use of abbreviations in documents of in­
formal character or where compactness was desired: private accounts and memoranda,
drafts, notes, subscriptions, dockets, summaries, and so on.

Like letter forms, abbreviations change across the centuries; matters of taste and criteria
of convenience do not remain the same, while the writing conventions of other languages
also play an influential role. One may wish to see an organic development of the abbrevia­
tion system (cf. Blanchard [1974] with Parsons [1976, 266]). To be sure, one starts with a
few simple abbreviations in the third century BCE, which increase in proportion to the
growing cursive character of everyday script, while ten centuries later the system has ac­
quired a daunting complexity. Nonetheless, evolution does not progress along the same
single line; a practice may disappear, only to resurface, perhaps distorted but still recog­
nizable, some centuries later.

The following is a list of practices and principles that apply to the great majority of abbre­
viations. Some of them are more specific to certain periods than others.

(1) The commonest method of abbreviation is by suspension, that is, to omit one or
more of the final letters of a word, even all the letters after the first. Cases in which a
suspension is effected but not signaled in one way or another (“unmarked abbrevia­
tions”) are not unknown and occur at all times. But the reader should normally be
alerted to the presence of an abbreviation: Suspension may be indicated by having
the last remaining letter written directly above or to the upper right of the penulti­
mate letter or by adding some sort of marker above or after the last remaining letter.
These are the two commonest practices.
(2) Interpretation of an abbreviation depends on the context; more than one word
will share the same beginning and be abbreviated identically or be represented by a
single “symbol,” which may lead to confusion. For example, in O.Strasb. 654.1, αργ
was first resolved as ἀργ(νρικῶν) but should be read as ἀργ(ίας); or, in Ptolemaic
texts, μεμέτρηκεν and other words starting with με- have more or less the same
shortened graphic representation as πέπτωκεν, τέτακται, and other words beginning
with πε-or τε- (Blanchard 1974, 4–5).

Page 2 of 8
Abbreviations and Symbols

Since the abbreviated words are often reduced to their bare essentials, little atten­
tion is paid to their declension; again, the context will make it (p. 172) clear whether,
for instance, a genitive or an accusative is meant. We first witness departures from
this practice in the later period (see item 10).
(3) In the simple forms of suspension, superscription predominates, but in the case
of several two-letter abbreviations, chiefly in the Ptolemaic period, the last remain­
ing letter may be written “inside” or under the penultimate one (but not under the
baseline).
(4) Supralinear letters are often deformed. A leftward-facing curve, either big or
small, usually represents pi in documents of the Roman period. In the later Ptolemaic
and early Roman periods, alpha is often written as an acute angle (Hakenalpha). In
the Roman period, alpha, epsilon, mu, pi, and tau often end up as small supralinear
horizontals. In the seventh and eighth centuries, overwritten alpha, epsilon, and tau
“tend to become a mere line; the intermediate stage is a line slightly thickened at the
beginning” (Bell 1910, xliv). (See also items 12 and 16).
(5) Two letters may be combined into a “monogram.” These are most often the first
two, and the only two remaining, letters of the word. Thus, for example, the crossbar
of E may be placed between the arms of K to form an abbreviation transcribed as
κε(ράμιον); or, certain words beginning with ΓP or ΠP are represented by Γ or Π
intersected by P. This practice is predominantly Ptolemaic and is attested as early as
the third century BCE.
(6) A word may be reduced by abbreviation to a single letter (the first), often accom­
panied by a stroke that acts as a marker. Single-letter abbreviations occur at all
times (they were earlier associated mainly with the Ptolemaic period, but this view
no longer holds). Here belong the various representations of αὐτός, made of a sinu­
soid (originally a Hakenalpha), either capped by a short horizontal or curve or fol­
lowed by a small, accentlike oblique.
It is worth adding that, unlike Latin, Greek did not use abbreviations consisting of
strings of initials; there is nothing comparable, for instance, to Latin SPQR.
(7) A letter extended rightward or downward (or even upward, if it is iota) may indi­
cate an abbreviation. In the later periods, this practice is usually in evidence when
the scribe intends to abbreviate at the end of a line; the letter is followed “by a single
stroke, curved or straight, with or without an overwritten letter” (Bell 1951, 427).
(8) Groups of two words may be abbreviated in such a way that what remains of the
second word is its irst letter, which is superscript. As expected, we find it with pre-
and postpositives (usually articles and prepositions); ὁ κ(αí), a collocation mostly
used for aliases, written as οκ, is a classic (p. 173) example. A more drastic abbrevia­
tion of this sort is τοκ, representing τò κ(ατʼ ἄνδρα); on the face of it, one could also
resolve τ(ò καí), but the context guards against misunderstandings. Indicative of the
extent of the practice, though of limited occurrence, is the abbreviation δηγ = δη
(μόσιος) γ(εωργός). The remaining letters of two words may also be merged into a
monogram: In a common abbreviation of γἣ κ(aτoικική), we have Γ with a line
through the top stroke.

Page 3 of 8
Abbreviations and Symbols

(9) In compound words, abbreviation may affect both components (“double suspen­
sion”); for example, τοπογραμματεύς may be rendered as τγρ or τοπγρ (the more con­
ventional τοπογρ also occurs). One of the commonest abbreviations of this kind is κοι
(or κol) for κ(άτ)οι(κος), κ(ατ)οι(κικός), and the like. This practice is attested as early
as the Ptolemaic period but becomes particularly common in late antiquity, when the
system acquires a new profile (see item 10).
In compounds that contain a numeral, the latter may be represented by an ordinal:
Thus, Πεντακωμία will appear as εκωμια and ἀκτάμηνος as ημηνος (Youtie 1973–
1975, i 153). One of the components may be represented by a symbol, as in com­
pounds that include weights, measures, or money: ς∫ = ἑξαδραχμία.
(10) From the fourth century on (and especially after the sixth), we witness the in­
creasing use of abbreviations à thème discontinu (“of discontinuous theme”; Blan­
chard 1974, 12),1 a more fortunate term than the earlier “abbreviation by contrac­
tion.” This type of abbreviation has affinities to practices described in item 9 but is
not limited to compounds and seems to be due to Latin influence. Here, the supralin­
ear letter may not be the last letter of the unabbreviated part of the word, but one or
more letters may have been omitted between the letter on the line and that over it.
The supralinear letter is virtually always a consonant, most often the next consonant
after the last letter on the line, though in extreme abbreviations a scribe may choose
to write the consonant considered most significant, as, for instance, ζτ =
ζ(υγοσ)τ(άτης). Sometimes we find two letters superscript, as in ναυπγ =
ναυπ(η)γ(ός). Naturally, not everything is straightforward; for example, we may infer
how κωμοκάτοικος came to be reduced to κωμοικ, but the modern reader will stum­
ble at first sight.
(p. 174) Overwritten nonsequential vowels also occur, but they come from the end of

the words and serve to indicate inflexions (Bell 1951, 431); ινδ° = ἰνδ(ικτίων)ο(ς) is
the commonest such example.
(11) Abbreviation may be indicated by means of a separate stroke added at the end
of the shortened word. This is the preferred practice in the later periods, though su­
perscriptions never disappear. We mostly find short horizontals, sinusoids (or “dou­
ble curves”), and obliques. The use of the single dot, a Latin influence, is very spo­
radic.
(12) The horizontals are largely simplified versions of supralinear letters, which pro­
gressively lost their defining characteristics and were reduced to single strokes (see
item 3). Editors often waver between recognizing in such strokes the original letters
and taking them just as abbreviation markers. Here much depends on personal taste,
but it would be preferable to interpret such strokes as letters unless they have more
than one function in the same text.
(13) The use of the sinusoid in abbreviations becomes common from the Roman peri­
od on. It is possible that the sinusoids written at the end of words initially represent­
ed the ligature alpha plus iota written increasingly cursively (with alpha in the form
of Hakenalpha), so that in the end the original components are not distinguishable.
The sinusoid may also stand in its own right as a symbol with a plurality of meanings
(see item 21).

Page 4 of 8
Abbreviations and Symbols

(14) A short oblique stroke of varying angles may follow or intersect the final letter.
Letters with long descenders (e.g., mu, rho, phi) or with elements that can be extend­
ed below the line (e.g., kappa) are prime targets for such intersections. This is by
and large a feature of the later documentation.
Obliques may sometimes be very short and could be added high in the line (like
acute accents). They sometimes occur with doubled letters, indicating plural forms
(Islamic period only; see item 16).
Abbreviating obliques may occasionally be doubled (commonly after ινδ =
ἰνδ(ικτίωνος) and μ = μ(όνον), but this entails no difference in meaning.
(15) From the late fifth century on, oblique strokes and sinusoids were often used in
conjunction with superscription, even though this would seem to be redundant.
(16) From the sixth century on, the duplication of the final letter(s) of an abbreviated
word signals the plural; this replicates a Latin convention. The abbreviation proper is
marked by either a superscription or some other indicator. Thus κάραβοι may be
shortened as καρ́ρ́ or κα (P.Lond. IV 1414.56, 1416.41).

(p. 175) Symbols

Most symbols stem from abbreviations by suspension; these may become reduced to
monograms whose original constituents are sometimes no longer discernible (ingenuity in
explaining some of them as suspensions of one sort or another has not been in short sup­
ply). This is the case with most symbols that represent weights and measures, as well as,
in the later period, money, which naturally occurred very frequently. Blanchard has made
a perceptive classification of the commonest symbols, which I follow (figure 7.1).

(17) A number of symbols go back to earlier Greek conventions, including those for
the monetary values imported into Egypt (drachma, obol, and fractions), to which
one may add the symbol for the year.
The origin of the L-shaped symbol representing the year (ἔτovs) remains unclear de­
spite attempts to explain it, for example, as a demotic derivation (Wilcken 1912, xlv;
contrast Blanchard 1974, 42) or as a conflation of E + T (Bell 1951, 425). It always
precedes the numeral. We find it from the third century BCE to the fourth century
CE, when dating by regnal years progressively disappears; after that, it is used only
with datings in the Oxyrhynchite era, but there it is placed high in the line, approxi­
mating the shape of the documentary paragraphos at that time.
(18) Some other very familiar symbols that we find early enough seem to go back to
demotic (analysis in Blanchard 1974, 30–31):
(i) the symbol used for the total (γίνεται) in Ptolemaic and Roman papyri, usual­
ly an oblique stroke (/). In later times, γίνεται is commonly abbreviated as γι:
Whether this relates to the increased use of the oblique as an abbreviation indi­
cator, which might have undermined the original function of the freestanding
oblique, we do not know.

Page 5 of 8
Abbreviations and Symbols

(ii) the symbol for “what remains” (λοιπόν), which looks like lambda with a tiny
omicron under it (it is conceivable that its origin was not always known and that
it was occasionally also considered as the shortening of a Greek word; it is inter­
esting that in papyri of the Islamic period the abbreviation for the same word
usually consists of lambda with pi underneath).
(iii) the L-shaped symbol indicating a “minus” out of a larger total (ἀφʼ ὧv); in
money accounts, it also introduces the uses of funds received.

Figure 7.1. List of symbols.

(19) The Ptolemaic period sees the rise of certain symbols, originally straightforward
monograms, which stand for some essential quantities: the talent, wheat (πνροῦ) (see
further Blanchard 1974, 44n18, 45n21), (p. 176) (p. 177) and the artaba (on the origins
and history of the sign see Blanchard 1974, 34–38).
(20) The Roman period introduces new abbreviations and symbols for certain frac­
tions, for the aroura (analysis in Blanchard 1974, 38–40), and for the conjunction καί.
In late antiquity, the repertoire increases: there are symbols for new monetary values
(myriads, monads, solidi, carats), for fractions, and for common words in fiscal con­
texts (e.g., ὑπέρ).
(21) Meriting a note apart is a very common symbol in documents of the Roman and
later periods: the sinusoid (or “double curve”). It has a plurality of meanings, of
which the commonest are the following (Youtie 1974, 50):
(i) the drachma (on the development of the sign see Blanchard 1974, 32–34)
(ii) the one-half fraction (ἣμισυ), an evolution of the earlier angular form
(iii) the year (but only after the numeral; note that many of the sinusoids writ­
ten after ordinal numbers, usually regnal, indictional, and consular iteration
years, function only as markers)
(iv) καί.

Page 6 of 8
Abbreviations and Symbols

The direction of the curves may be reversed (Youtie 1973, vol. 1,132; Blanchard 1974,
33). This form occurs only sporadically in the Roman period but is particularly common in
the eighth century, when the symbol usually stands for the concept one-half.

Verschleifungen (“Slurs”)
In the early days of papyrology, Verschleifungen were given the status of a particular sub­
group of abbreviations (Wilcken 1912, xlii–xliii). The scribe “slurs” or pens the initial let­
ters of a word and continues with a wavy or ribbonlike line in which no individual letters
are distinguishable; the initial letters and the context suffice to identify the word meant.
This practice is predominantly found with the names and titles of emperors and the
names of months in date clauses, most often in tax receipts, but occasionally also in dec­
larations to state authorities. Also in tax receipts, especially on ostraca, some set words
such as διέγραψεν (“paid”), γενήματος (“produce”), and so on may be written in a similar
fashion (Préaux 1954). However, Verschleifungen are no longer considered as abbrevia­
tions insofar as no letter was intentionally omitted (Bilabel 1923, 2281; Blanchard 1974,
1, 17). Thus, in transcribing Verschleifungen, no brackets should be used.

Bibliography
Bastianini, G. 1992. “Le abbreviazioni.” In Corpus dei papiri filosofici greci e latini (CPF):
Testi e lessico nei papiri di cultura greca e latina. Parte I: Autori noti, vol. 1⋆⋆: 276–281.
Florence: Leo S. Olschki.

Bell, H. I. 1910. “§6. Palaeography, Diplomatic, etc.” P.Lond. IV, pp. xlii–xlv. London. (On
abbreviations in documents of the early Islamic period.)

———. 1951. “Abbreviations in Documentary Papyri.” In Studies Presented to David

Moore Robinson on his Seventieth Birthday, vol. 1, ed. G. E. Mylonas, 424–433. St. Louis:
Washington University.

Bilabel, F. 1923. “Siglae.” In Paulys Realencyclopädie der classischen Altertumswis­

senschaft, vol. 2A: 2279–2308. Stuttgart.

Blanchard, A. 1974. Sigles et abréviations dans les papyrus documentaires grecs:

Recherches de paléographie. Bulletin of the Institute of Classical Studies, Suppl. 30. Lon­
don: Institute of Classical Studies.

McNamee, K. 1981. Abbreviations in Greek Literary Papyri and Ostraca. Bulletin of the
American Society of Papyrologists, Suppl. 3. Chico, Calif.: Scholars Press.

———. 1985. “Abbreviations in Greek Literary Papyri and Ostraca: Supplement, with List
of Ghost Abbreviations.” Bulletin of the American Society of Papyrologists 22: 205–225.

Parsons, P. J. 1976. Review of Blanchard 1974. Journal of Hellenic Studies 96: 265–266.

Page 7 of 8
Abbreviations and Symbols

Préaux, C. 1954. “Sur lʼécriture des ostraca thébains dʼépoque romaine.” Journal of
Egyptian Archaeology 40: 83–87.

Torallas Tovar, S., and K. A. Worp. 2006. To the Origins of Greek Stenography (P. Monts.
Roca I). Barcelona: Publicacions de lʼAbbadia de Montserrat.

Wilcken, U. 1912. Grundzüge und Chrestomathie der Papyruskunde. Vol 1., Historischer
Teil. Part 1, Grundzüge. Leipzig: B. G. Teubner.

Youtie, H. C. 1973–1975. Scriptiunculae, vols. 1–2. Amsterdam: A. M. Hakkert.

———. 1974. The Textual Criticism of Documentary Papyri: Prolegomena, 2d ed. Bulletin
of the Institute of Classical Studies, Suppl. 33. London: Institute of Classical Studies.

Notes:

(1.) Whether contractions of the kind known from Latin writing (i.e., a word is reduced to
its initial and final letters) were in use in the papyri has been a controversial issue. Wilck­
en (1912, xliii–xlv) regarded contractions as a special subgroup, but his evidence relates
mainly to nomina sacra, which are a case apart (see chapter 25). Blanchard dismisses this
“subgroup” altogether (1974, 2, 18–19).

Nikolaos Gonis

Nikolaos Gonis, Reader in Papyrology, University College London.

Page 8 of 8
Practical Help: Chronology, Geography, Measures, Currency, Names, Proso­
pography, and Technical Vocabulary

Practical Help: Chronology, Geography, Measures, Cur­

rency, Names, Prosopography, and Technical Vocabu­
lary
Roger S. Bagnall
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0008

Abstract and Keywords

Both editors and readers of papyri, even if they know well the language in which their
texts are written, constantly encounter reminders that these artifacts come from another
society. These come in the form of a host of technical details. To most readers who are not
professional papyrologists, they are simply obstacles to understanding the transaction or
some aspect of it that is of interest to them. This article explains the most common and
most important of these matters and provides some guidance to works where more detail
can be found. Preferring variation in diction, literary Greek was on the whole averse to
the use of technical vocabulary. The Greek of the documents, however, has a rich reperto­
ry of words used with specialized meanings for particular institutions, offices, taxes, legal
acts, and other administrative and legal purposes.

Keywords: Papyri readers, papyri editors, technical details, technical vocabulary, literary Greek, documentary
Greek

BOTH editors and readers of papyri, even if they know well the language in which their
texts are written, constantly encounter reminders that these artifacts come from another
society. These come in the form of a host of technical details. If a lease of land tells us
that it was written in the fifth year of Nero, on Thoth 2, in Theadelphia in the Arsinoite
nome; that the land measures five arouras and will have an annual rent of twenty-five
artabas of wheat and fifty drachmas; and that the lessee is named Peteesis, son of
Haryothes—how is the reader not only to make (p. 180) basic sense of all of this informa­
tion but also to extract from it the maximum contribution to understanding the meaning
of the text?

Most of these matters are not of much inherent interest to the average papyrologist; they
are puzzles to be decoded, part of the specialized knowledge one acquires in the course
of learning the discipline. To most readers who are not professional papyrologists, they
are simply obstacles to understanding the transaction or some aspect of it that is of inter­

Page 1 of 19
Practical Help: Chronology, Geography, Measures, Currency, Names, Proso­
pography, and Technical Vocabulary
est to them. When exactly was the lease concluded? How many hectares or acres of land
were at stake? How much was the rent in terms that mean something to us? Who was the
lessee, and why do his names sound so strange to our ears? This chapter explains the
commonest and most important of these matters and provides some guidance to works
where more detail can be found. These subjects are often unavoidably complicated, and
many details must be left aside here. Even if we restrict ourselves to the documents from
Egypt, we find Egyptian, Persian, Greek, Roman, and Semitic origins for the names,
terms, and measures found in the papyri accumulating in layers over the centuries.

Chronology: Measuring Time

The Year

The Egyptians used a solar year of 365 days, divided into twelve months of 30 days each
and an additional 5-day festival period at the end of the year, which in Greek is called the
(hêmerai) epagomenai (“added” days). Because it was about a quarter-day shorter than
the true length of the solar year, the Egyptian calendar lost a day about every four years,
thereby causing the starting date of the year, Thoth 1, to move backward against the ac­
tual solar year, returning to its starting point in a cycle of 1,460 years, the “Sothic” cycle.
In Egyptian documents, the months are designated as the first through the fourth of the
three successive seasons, harvest (3ḫ.t), winter (pr.t), and summer (šmw). (For the Egypt­
ian calendar see Parker 1950.) The Greeks, accustomed to calling each month by a dis­
tinctive name usually connected with a festival that took place during it, adopted a set of
twelve such names for these Egyptian months. These names survive today in the Coptic
calendar.

Greek calendars, by contrast, were lunar, using twelve months alternately of 29 and 30
days, thus totaling 354 days. Each city or state had its own set of month names and peri­
odically declared additional (“intercalary”) months in order to keep the months in a
roughly constant relationship to the solar year. Because Alexander the Great and Ptolemy,
son of Lagos, were Macedonians, it was the Macedonian calendar that came to Egypt
with the conquest of 332 BCE. Although it coexisted (p. 181) with the Egyptian calendar,
few people found it easy to keep track of their relationship, which was constantly shifting
because of the different bases and separate sources of instability. Early papyri show an
attempt to use both calendars independently, but inconsistencies are common. Even un­
der Ptolemy II, some well-informed individuals used an informal equation in which the
count of days was maintained independently for only one of the calendars (for the Ptole­
maic calendar see Samuel 1962 and Grzybek 1990). From the later third century BCE on,
as a result of the difficulty of accurate maintenance of the dual calendars, people attempt­
ed various “assimilations” of the calendars (in which the lunar character of the Macedon­
ian calendar—with its names equated to Egyptian months—disappeared), but the early
history of these attempts is still a matter of controversy. The last system adopted, which
remained in use in the Roman period, equated the Macedonian month Dios to the Egypt­

Page 2 of 19
Practical Help: Chronology, Geography, Measures, Currency, Names, Proso­
pography, and Technical Vocabulary
ian month Thoth. The much simpler Egyptian calendar was thus the survivor of this com­
petition.

This method, however, remained unstable against the solar year. An attempt in 238 BCE
to reform the Egyptian calendar to eliminate this drift against the solar year was unsuc­
cessful. It was not until the first decade of Roman rule, under Octavian (probably in 26/25
BCE), that a lasting reform was introduced. This approach created leap years by adding a
sixth epagomenê every four years (in the year before that in which February acquired a
twenty-ninth day according to the Roman calendar; thus, for example, in August 103, not
February 104 [Skeat 1993]). From that point on, the relationship of Thoth 1 to the Roman
calendar was fixed at August 29 in normal years and August 30 in leap years. The old
reckoning (p. 182) without a leap year remained in use in some contexts, particularly reli­
gious, and dates “according to the ancients” or “according to the Egyptians” are found
for some three centuries after Octavian's reform (Hagedorn and Worp 1994). Macedonian
months are occasionally found in documents of the Roman imperial period, as are alter­
nate names for the months named in honor of various members of the imperial household
(Scott 1932). The Roman months and system of reckoning days themselves are rarely
used in the papyri except for a short period in the late third and early fourth century
(Bagnall and Worp 2004, 3) (table 8.1).

Table 8.1. Months

Egyptian Macedon­ Roman Honorific Julian Equivalent in Normal

ian Year

Thôth Dios Sebastos, Ger­ August 29–September 27

manikos

Phaôphi Apellaios September 28–October 27

Hathyr Audnaios Neos Sebastos October 28–November 26

Choiak Peritios Ioulieus, Neroneios November 27–December 26

Sebastos, Hadri­
anos

Tybi Dystros December 27–January 25

Mecheir Xandikos January 26–February 24

Phamenôth Artemisios February 25–March 26

Page 3 of 19
Practical Help: Chronology, Geography, Measures, Currency, Names, Proso­
pography, and Technical Vocabulary

Phar­ Daisios March 27–April 25

mouthi

Pachôn Panemos Germanikeios April 26–May 25

Pauni Lôios Sotêrios May 26–June 24

Epeiph Gorpiaios Drousieus June 25–July 24

Mesorê Hyperbere­ Kaisareios July 25–August 23

taios

Epagome­ August 24–August 28

nai

Counting Years

The Egyptians followed a system of counting the years of pharaohs' reigns within the
framework of the civil year, which began on Thoth 1. At a king's accession, the time until
the end of the next epagomenai was counted as the king's year 1, with year 2 starting on
Thoth 1. A king who came to the throne near the end of one year and died soon after the
start of another might thus have a “highest regnal year” that is higher by two than the
number of complete years he had actually ruled. The regnal years of the Ptolemies were
counted in the Egyptian calendar according to this system. In addition, in some third-cen­
tury-BCE documents, a financial year began halfway through the Egyptian year. (For the
Ptolemies' Egyptian regnal years, see Pestman 1967; to compute precise dates, use the
tables in Skeat 1969.)

Macedonian practice, however, reckoned reigns from the actual date of accession to the
throne until its anniversary. Depending on the date at which a king came to power, there­
fore, the Macedonian and Egyptian regnal years would be different for some part of each
year. (On Macedonian years see Samuel 1962; for the Macedonian calendar from 260 to
221, see Pap.Lugd.Bat. XXI 220–263.) Documents from the Ptolemaic period often have
prescripts giving the names of the ruling sovereigns and those of several holders of
priesthoods connected to the royal cult (Clarysse and Van der Veken 1983).

The regnal years of Roman emperors through the fourth century are reckoned in the pa­
pyri according to the Egyptian system, although elsewhere in the empire people used the
count of years for which they had held the official titles symbolizing their imperial power:
imperator, consul, and holder of tribunician power. The papyri commonly, but not univer­
sally, give the names and titles of the ruling emperors. Until the later third century, coem­
perors used the regnal year count of the senior emperor; after that, separate counts were
kept for each ruler. (For these regnal formulas and the beginning years of reigns, see Bu­

Page 4 of 19
Practical Help: Chronology, Geography, Measures, Currency, Names, Proso­
pography, and Technical Vocabulary
reth 1964; Pestman 1967 has a list of regnal years and a table for converting months and
days to their Julian equivalents. Grenier 1989 collects Roman titulature in Egyptian-lan­
guage texts. Kienast 1990 is also useful for Roman imperial reckoning.)

After the beginning of the Tetrarchy in 293, in Diocletian's ninth year, dating by
(p. 183)

the Roman consuls, which had been rare except in documents in Latin or translated from
Latin, started to be common, sometimes side by side with regnal years. Because the con­
sulate ran with the Julian year (January 1), it changed at a different time from the regnal
year. Consulates were used mostly for the official date at the beginning or end of a legal
or official document. Because the names of consuls changed each year, they required dis­
semination from the court. As a result, scribes sometimes did not know the names of the
new consuls and used “after the consulate” with the names from the previous year (Bag­
nall et al. 1987).

A further change came with the introduction of the indiction, initially a fiscal year, reck­
oned in fifteen-year cycles, with the first cycle beginning in 312. Indiction dates are thus
not sufficient in themselves to identify a year uniquely for the historian; other dating cri­
teria, internal or external, are also needed. Unfortunately for the papyrologist and histori­
an, the indiction eventually supplanted the regnal year entirely except in Oxyrhynchus
and perhaps a few other places, and even there the regnal year vanished before the end
of the fourth century. The imperial indiction year coincided in Egypt with the old civil
year, but in practice in the Thebaid and some other parts of the country it soon came to
be reckoned from Pachon 1, or four months before the civil year. There are a number of
local complexities in its use. (See generally Bagnall and Worp 2004, 22–35.)

The regnal year was reintroduced (and required in legal documents) by the Emperor Jus­
tinian in 537 CE but now reckoned from the date of accession. It thus became possible to
have a regnal year, a consulate, and an indiction date all in the same document and each
with a different starting date in the course of the year. In Oxyrhynchus there was in addi­
tion a local era year, based on the regnal years of Constantius and Julian and using a
Thoth 1 year. Scribes in Oxyrhynchus tried various methods of simplifying this informa­
tion overload; these are still not fully understood. Documents of the Byzantine period con­
tain a considerable number of scribal inconsistencies between the different reckoning
systems; the indiction year is usually the most reliable marker (see Bagnall and Worp
2004, index s.v. “blunders”). After the Arab conquest, an era dated by the accession of
Diocletian (284), which is found for horoscopes and some private inscriptions under Ro­
man rule, begins to appear in the documentary papyri (at least from the Arsinoite nome);
still later the Hijra year (reckoned from 622) begins to be used, sometimes with the
Egyptian months and occasionally with the Arabic calendar. The latter is lunar but does
not use intercalary months as the Macedonians did, thus moving backward against the so­
lar calendar by about eleven days each year. (For all aspects of year reckoning and the
calendar in the period from Diocletian to after the Arab conquest see Bagnall and Worp
2004.)

Page 5 of 19
Practical Help: Chronology, Geography, Measures, Currency, Names, Proso­
pography, and Technical Vocabulary

(p. 184) Geography

Egypt was historically divided into forty nomes (Greek nomoi), half in Lower Egypt (the
delta of the Nile) and half in Upper Egypt (the Nile valley from the head of the delta to
the first cataract of the Nile at Aswan). The balanced duality of Upper and Lower Egypt
was deeply rooted in the Egyptian consciousness. The Persians, Alexander, and the
Ptolemies all in turn took over this basic structure, which reflected not only administra­
tive convenience but also regional and local religious identity. The Greeks called the capi­
tals of the nomes “cities” even though they were not autonomous, and they began to ac­
quire standardized sets of institutions already in the Ptolemaic period. The Romans accel­
erated the civic development of these “metropoleis,” culminating in Septimius Severus's
grant of the right to have a city council (boulê) in 200 CE. As the delta gradually devel­
oped into the dominant part of the country, with perhaps two-thirds of Egypt's arable land
and population, the number of nomes there grew.

Most of the nomes were subdivided into units called toparchiai (toparchies), representing
a group of villages with their land. We know the full extent and number of toparchies for
only a small number of nomes, but the average nome seems to have had six to ten of them
referred to by their names. They were replaced in 307/308 with a system of pagi, in which
nomes were typically divided into a larger number of subunits than before. These pagi
were generally numbered. The Arsinoite nome, uniquely as far as we know, was divided in
the third century BCE into three large subdivisions, the merides of Herakleides, Themis­
tos, and Polemon, to which toparchies were subsidiary. During some parts of the Roman
period these units had their own governors as if they were separate nomes. The large size
of the Fayyum (perhaps about fifteen hundred square km at its peak) was presumably re­
sponsible for this unique structure.

The village (Greek kômê) was the next level of administration and geography; the villages
varied greatly in population and amount of land within their control. Smaller units, often
called epoikion, also existed. The character of these evidently varied and included both
large private farmsteads and small hamlets with multiple households. Some evolved into
villages, but others remained part of the territory of a larger village.

At the other end of the spectrum, the nomes were from time to time grouped into larger
regional units with some form of government. Only the Thebaid in Upper Egypt plays an
important regional role in the Ptolemaic documents, but, under the Romans, three or four
units, each headed by an epistrategos, are found. After Diocletian, the province of Egypt
was divided into smaller provinces headed by praesides, whose number and boundaries
changed repeatedly. The subdivision of the delta marks its increasing importance. The im­
perial authority was in this way brought closer to the population.

Individuals are often identified in the papyri by their place of legal residence (idia). This
description usually includes the city of residence or an adjective derived (p. 185) from it
(“from Hermopolis” or “Hermopolite”) in the case of urban residents. Villagers are gener­
ally identified as coming from a particular village of a specified nome, sometimes with

Page 6 of 19
Practical Help: Chronology, Geography, Measures, Currency, Names, Proso­
pography, and Technical Vocabulary
toparchy or (from 307/308 onward) the pagus specified as well. On locally used docu­
ments, the nome name is sometimes omitted, and only the toparchy or pagus is given.

Most place names in the papyri are, as one would expect, of Egyptian origin. Sometimes
equivalences between Greek and demotic or Coptic place names are known; in other cas­
es, the Egyptian derivation of a Greek transcription may be apparent. Still others are so
far unexplained. Many times the same name appears in more than one part of Egypt. The
Arsinoite nome forms a special situation in which the large-scale expansion of settlement
in the early Ptolemaic period produced a wave of new place names. Some of these were
drawn from the ideological program also visible in the naming of the demes of Alexan­
dria, encompassing Ptolemaic dynastic names and cultic names; others (Memphis,
Boubastos, and the like) refer to the places from which Egyptian settlers had come to the
Arsinoite.

Geographical names in Egypt are inventoried in Calderini and Daris (1935–). There are
useful repertories specific to particular nomes, most notably, the Hermopolite (Drew-Bear
1979); Oxyrhynchite (Pruneti 1981); Herakleopolite (Falivene 1998); and Arsinoite (Leu­
ven Fayum Villages Project). For topographical maps that show locatable places, see the
Barrington Atlas (Talbert 2000).

Measures
Nowhere do we see the succession of influences on Egypt better than in weights and
measures. Pharaonic, Persian, Greek, and Roman measures form cultural strata, with new
measures being connected to old and local variants common and numerous. Only a selec­
tion can be treated here, but these account for the overwhelming majority of instances
encountered in texts. (In the discussion, the figures given are approximate.)

The Land

Linear and square measurements are usually given by a system that dates back to
pharaonic times, in which the basic unit of length was the cubit (Eg. mḥ, Gk. pêchus,
Copt. maje), of .525 m. (Other cubits are also known, but “cubit” without description nor­
mally refers to this measure.) The cubit was divided into hands and fingers and served in
turn as the basis for the principal unit of surface measurement or area, the aroura (Eg.
st3, Copt. sôt), which was 100 cubits, or a schoinion (52.5 m) on a side, thus 2,756 square
meters (1 square km = 363 arouras). Subdivisions of the (p. 186) aroura are usually given
by fractions (1/2, 1/4, 1/8, 1/16, 1/32, 1/64), the smallest of which, 1/64, is called a hamma.
However, units based on a decimal division were also known and used a notional unit that
measured 1 × 100 cubits (the “ground cubit,” or mḥ-itn in Demotic). These were probably
reconciled by use of the geometrikon, or surveyor's measure, which was based on 96
rather than 100 cubits. An obscure measure called the bikos may have been 1½ hammata,
but its equation to the mḥ-itn is not excluded. (See T.Varie, pp. 156–158.) The aroura and
its subdivisions were the principal basis for the computation of taxes on land.

Page 7 of 19
Practical Help: Chronology, Geography, Measures, Currency, Names, Proso­
pography, and Technical Vocabulary

Table 8.2. Length and Land

1 sacred cubit (mḥ) = 7 hands = 28 fingers = .525 m

1 kalamos = 6 cubits (as square measure, 36 square cubits)

1 ground cubit (mḥ-itn) = 96 square cubits = 1/100 aroura

1 hamma = 144 square cubits = 1/64 aroura

10,000 square cubits = 1 aroura (st3, sôt) = 2756.25 square met

A volume of earth was measured by the naubion, which in the Ptolemaic period referred
to a mass of 2 cubits in each dimension, or 1.158 cubic meters. In the Roman period, the
naubion meant a volume of 3 cubits in each dimension, or 3.9 cubic meters (table 8.2).

Dry Measure

For wheat, barley, and other dry commodities the standard unit of reference was the arta­
ba, a Persian measure that replaced the pharaonic ẖ3r (sack) of about 80 liters. The men­
tions in documents of many different measures and the varying information found in an­
cient metrological treatises have generated much controversy about the degree to which
the artaba varied from place to place and time to time. It may initially have corresponded
to about 60 Egyptian hin (30 liters) and thus to 30 Greek choinikes and perhaps equated
to the Persian hofen (Vleeming 1980, 1981). In the Greek papyri of the Roman period, the
artaba is consistently composed of 40 choinikes, which was the unit recognized by the
government; it was also the commonest measure in the Ptolemaic period and perhaps
corresponded roughly to half of the pharaonic sack. A unit called simply the “measure,”
short for “four-choinix measure,” is often found; it is thus a tenth of an artaba. The Egypt­
ian-derived term mation (maje in Coptic) was also used in the Roman and Byzantine peri­
ods for a tenth of an artaba. The artaba is also, however, measured in fractions down to
1/48. How the two different systems of subdivision were handled in practice is unclear,
but the parallel of coexisting divisional systems for land is instructive. Although the suc­
cessive governments aimed to enforce standardization in the measure used in tax pay­
ments (the “receiving measure,” or metron dochikon) (p. 187) and therefore checked ship­
ments for conformity, there was no central control of the measures used in private trans­
actions. Contracts often specify which measure to use in order to avoid possible doubt.
One common measure, the Athenian, is at times equated to the “receiving
measure” (Clarysse 1985).

Table 8.3. Dry Measure

1 artaba = 10 metra = 10 matia = 40 choinikes = 72 sextarii (xestai) = 38.808 liters

Page 8 of 19
Practical Help: Chronology, Geography, Measures, Currency, Names, Proso­
pography, and Technical Vocabulary

1 mation or metron = 3.88 liters

1 choinix = .97 liter

1 sextarius = .539 liter

1 artaba = 3 1/3 modii xystoi = 4.5 modii Italici

1 modius xystos (castrensis) = 21.6 sextarii = 11.64 liters

1 modius Italicus = 16 sextarii = 8.62 liters

Part of the controversy over the artaba and choinix arises from metrological tables of the
Roman period and from equations between the artaba and Roman dry measures, which
sometimes appear in the papyri. The Romans measured grain with the modius, but this
too needed to be specified. The modius Italicus (Italian modius) equated to 16 sextarii
(xestai in Greek), or 8.62 liters. The artaba generally equated to 4.5 modii Italici, or 72
sextarii. The other common modius, the modius castrensis, is usually called xystos in
Greek, and ten of them equaled 3 artabas; its size was thus 21.6 sextarii. A slightly larger
variant of the modius castrensis (at 22 sextarii) is also attested.

Because the sextarius held .539 liter, these Roman equivalences allow a calculation of the
capacity of the artaba, which was thus 38.8 liters, and the standard choinix .97 liter. This
capacity for the artaba has also been identified in the Rosetta Stone on the basis of the in­
dication there that it was 4/15 of a cubic cubit. If Pliny's indication that the modius Itali­
cus of Egyptian wheat weighed 20 5/6 Roman pounds is correct, the artaba of 4.5 modii
must have weighed about 30.28 kg. It is impossible to be sure how far this standard arta­
ba prevailed throughout Egypt at all times, but there is no good evidence for any normal
use of a different artaba (Duncan-Jones 1976a, 1976b; Shelton 1977, 1981; Rathbone
1983) (table 8.3).

A very large number of other measures appear at different times and places for a variety
of other goods; examples are the môion and sarganê for chaff, the desmê for hay, and the
load, or gomos, for various goods. In most instances we neither know the size of these
measures nor have a clear sense of how localized their capacity may have been.

Liquid Measure

The Greek system of liquid measurement was based on the kotylê, but this was itself not
of fixed size. The Attic kotylê, the measure most often meant when another is (p. 188) not
specified, was about .27 liter. The next larger unit was the chous, which had the following
equations:

12 kotylai = 1 chous (3.24 liters)

Page 9 of 19
Practical Help: Chronology, Geography, Measures, Currency, Names, Proso­
pography, and Technical Vocabulary
12 choes = 1 metrêtês (39 liters).

An alternative system in use in the Arsinoite nome in the Ptolemaic period had a larger kotylê (.
406 liter) and chous (4.875 liters), with a metrêtês of 6 choes, or 29.25 liters.

Table 8.4. Liquid Measures in the Great Oasis in the Fourth Century

1 kotylê = .27 liter

2 kotylai = 1 sextarius (.54 l)

12 kotylai = 6 sextarius = 1 chous (3.24 l)

9 sextarii = 1½ choes = 1 boxion (4.86 l)

18 sextarii = 3 choes = 1 keramion (9.72 l)

20 sextarii = 3 1/3 choes = 1 marion (10.8 l)

(1 metrêtês = 12 choes = 4 keramia [38.9 l])

The papyri of the Ptolemaic and Roman periods offer a rich array of names for containers
of wine and (to a lesser extent) oil (see Kruit and Worp 1999, 2000). Some of these are
formed from place names like Rhodes, Knidos, Chios, and Samos. Originally they referred
to jars of wine that originated in these places. The containers had varying capacities; for
instance, Thasian jars have been estimated at 21 liters, Chian jars at 23, Rhodian ones at
26, and Milesian and Samian jars at 52 (thus 1.5 Attic metrêtai). Some of these jar names
remained in use but apparently referred to locally made jars of distinctive capacities; the
knidion had the longest life and was equated to the diploun (“double jar”). The common­
est term, however, is keramion, which simply means “pottery jar” and does not of itself in­
dicate any particular capacity. Within any household or enterprise, everyone would nor­
mally know what size keramion was in use. Where doubt could arise, adjectives referring
to capacity were added. The commonest keramia corresponded to 3 (9.72 liters) and 4
choes (12.96 liters).

The Romans used a liquid and dry measure called the sextarius, which the Greeks trans­
lated as xestês. This was about .54 liter. It makes its appearance in Egypt well into the pe­
riod of Roman rule; by the late fourth century it had become dominant and ousted the old­
er Greek system entirely. table 8.4 shows the set of equivalences of measures in use in a
fourth-century context that continued to employ both choes and sextarii.

Weight
The Greek weight system was based on the drachmê, or drachma, a term also used for
coins of this weight. In the Roman period, the drachma was treated as 1/96 of the Roman
pound (about 323 g), or about 3.36 g (table 8.5). But different Greek (p. 189) cities and
other regions of the Greek world had their own weight standards, and one always needs
Page 10 of 19
Practical Help: Chronology, Geography, Measures, Currency, Names, Proso­
pography, and Technical Vocabulary
to know what standard is in use in order to interpret figures in the texts. In general in the
papyri one can presume that the drachma as a weight refers to the Ptolemaic standard
(see the following section), which is close to but a bit higher than the figure given above.
In Egyptian texts the drachma was reckoned as half of a qd (Copt. kite) and thus 1/20 of
the dbn, which equated to 20 drachmas. In the Greek system of weights (described in the
next section), the drachma was a subdivision in a system that included the mna and the
talent, both also accounting terms for money. (For weights in jewelry see Ogden 1996.)

Table 8.5. Weights and Currency

8 chalkoi = 1 obol

6 obols = 1 drachma

2 drachmas = 1 qd (kite)

4 drachmas = 2 qd = 1 stater (tetradrachm) = 1 denarius

20 drachmas = 10 qd = 1 dbn

100 drachmas = 50 qd = 5 dbn = 1 mna

6000 drachmas = 3000 qd = 300 dbn = 1500 denarii = 60 mnai = 1 talent (Eg. krkr)

1 gold “quarter” (tetartê) = ½ drachma (a quarter of a didrachm or qd)

1 gold mnaieion = 8 drachmas (weight) = 16 “quarters”

1 Roman pound (litra) = 12 ounces (ounkiai) = 288 Roman grams (grammata) = 323 g

1 solidus (Diocletian to Constantine) = 1/60 pound = 4.8 grammata

1 solidus (post-Constantine) = 1/72 pound = 4 grammata = 24 carats

The Roman pound also had its own system of subdivisions and consisted of 12 unciae (Gk.
ounkiai), or ounces, and grams (Gk. grammata), at 24 grams to the ounce. Thus 288
grams constituted a pound.

Currency
Egypt used coinage only to a limited extent before the Macedonian conquest, but Egypt­
ian documents both before and after the conquest readily express monetary value using a
weight-based system, the deben and the kite.

Page 11 of 19
Practical Help: Chronology, Geography, Measures, Currency, Names, Proso­
pography, and Technical Vocabulary
Greek monetary systems were based on the weight unit called the drachma, with the fol­
lowing equivalences:

100 drachmas = 1 mna

60 mnai = 6,000 drachmas = 1 talent

The talent and the mna, however, were accounting units rather than minted coins. The
(p. 190)

drachma, as indicated earlier, varied according to local standards. During the last decade of the
fourth century, Ptolemy I abandoned the heavier Athenian weight standard used by Macedonian
kings up to this time in favor of a lighter Phoenician drachma of about 3.55 metric grams of sil­
ver, with the tetradrachm (four-drachma coin) weighing about 14.2 g. This weight standard re­
mained in use through the remainder of Ptolemaic rule. Gold coins were also minted, with the
eight-drachma coin valued thesameasa mina of silver (the value ratio between the metals was 12
1/2:1).
Beginning in the reign of Ptolemy II, the role of bronze coinage became steadily more im­
portant. The stages of this development are controversial (Maresch 1996; Cadell and Le
Rider 1997; von Reden 2007). At an early period, under Ptolemies II and III, the bronze
coinage had a substantial fiduciary element, but the relationship of silver and bronze be­
gan to change from the reign of Ptolemy IV on in tandem with an increase in nominal
prices expressed in bronze. The bronze-to-silver ratio eventually reached a level of 1:480,
and additional charges were levied for exchange of bronze for silver. Most of the prices in
later Ptolemaic papyri are expressed in bronze drachmas and talents. This development
was not the result of the debasement of the silver tetradrachms, however, which did not
begin until 149/148 and became substantial only under Ptolemy XII (Hazzard 1995).
When working with prices in drachmas or talents in Ptolemaic texts, one must determine
whether the figures are given in silver or bronze drachmas. If they are in silver, they can
be compared across the Ptolemaic period with little risk of distortion; if in bronze, howev­
er, one must try to establish the period and the ratio of bronze to silver prevalent at that
time (see Maresch 1996; Cadell and Le Rider 1997).

After the Romans acquired Egypt, Augustus minted bronze coins on the same denomina­
tional standard as the Ptolemies; from his year 28 on, these bear year dates, as do those
of his successors. Tiberius introduced a tetradrachm of bronze with a small admixture of
silver (“billon”), nominally equated to the Roman denarius, and Nero a bronze drachma
equated to the sestertius. The drachma, as under the Ptolemies, was divided into six
obols, and each obol into eight chalkoi, although the smallest fraction actually minted was
usually the two-chalkoi coin. Tax payments were supposed to be made in billon
tetradrachms, and a surcharge was levied for payment in the bronze drachmas and frac­
tions. This policy gave rise to a number of complex accounting practices visible in the pa­
pyri, including the charging of supplements for exchange, discounting of the smaller
coins, and the use of a tetradrachm with more than twenty-four obols in accounting. No
gold was minted in Roman Egypt. This monetary system was separate from that of the
rest of the Roman East and thereby kept Egypt in a kind of monetary isolation (West and
Johnson 1944; Gara 1976).

Page 12 of 19
Practical Help: Chronology, Geography, Measures, Currency, Names, Proso­
pography, and Technical Vocabulary
That isolation was ended in 296 by Diocletian, who introduced into Egypt standard Ro­
man coinage of his day; from this point on we begin to find references to denarii in the
papyri. The old terminology, however, with drachmas and talents, remained in use for
quite some time. A wave of price increases had occurred in the (p. 191) period after the
Antonine plague in the later 160s, but prices then remained stable until about 275, when
the papyri indicate an increase of nearly an order of magnitude. The successive waves of
reduction in silver content in the imperial coinage after 296, coupled with the redenomi­
nation of coins, led to what looks like a massive price inflation between Diocletian and the
third quarter of the fourth century. The nature, causes, and effects of this phenomenon
are controversial, but there are good reasons to regard it essentially as a monetary phe­
nomenon, the result of episodic pressures on imperial finances that led to debasement of
the currency (Bagnall 1985). Just as the obol disappears in the papyri in the second half
of the third century, the drachma vanishes in the fourth. Accounts come to be kept largely
in talents or in myriads of denarii (a myriad, or ten thousand denarii, was equivalent to 6
2/3 talents).

The most decisive monetary phenomenon of the fourth century, however, was the rise of
the gold coin called the solidus (mostly called nomisma, nomismation, or holokottinos in
Greek), first issued by Diocletian at 1/60 of a Roman pound, subsequently set at four Ro­
man grammata, or 1/72 of a pound, by Constantine and stable at that weight for the re­
mainder of Roman rule in Egypt. Because we have no consistent series of data, price
movements in gold are harder to track than those in bronze currency, although wheat
seems to have become cheaper against gold between the fourth and the sixth century
(see Banaji 2002). Most major transactions in the papyri of the fifth to the seventh cen­
turies were denominated in solidi, and tax obligations in money were calculated in gold.
The solidus had subdivisions in imperial coinage; the only one commonly referred to in
the papyri is the tremissis, a third of a solidus. We find also the carat (keration), at 1/24 of
a solidus, as an accounting unit rather than an actual coin. Most taxes were actually paid
in bronze, however, thus giving rise to extensive opportunities for exchange transactions
and, presumably, proits.

The papyri of the fifth to the seventh centuries quote many figures in the form of so many
solidi minus (Gk. para) so many carats. The number subtracted varies from place to place
and time to time, and the meaning of these sums remains controversial (see generally
Maresch 1994). In part they seem to reflect local standards, which also led to accounting
practices in which a total number of carats was divided by a figure lower than twenty-
four to get the number of whole solidi. The standard of twenty-four, however, remained
the “full-weight” standard.

Names
Stretching over nearly fifteen hundred years, the papyri record many names of individu­
als. Most of these are Egyptian, as one would expect, but many are Greek, Latin, and
Semitic, along with fewer names of other origins. A number of volumes collect these

Page 13 of 19
Practical Help: Chronology, Geography, Measures, Currency, Names, Proso­
pography, and Technical Vocabulary
names, but because these have been created according to the languages (p. 192) of the
documents, none gives a comprehensive picture or is even current. For the Greek papyri,
searching the Duke Data Bank of Documentary Papyri (DDbDP) is an indispensable sup­
plement to the printed volumes of Preisigke (1922)and Foraboschi (1967–1971). For texts
in other languages there is at present no such easy solution, but demotic texts are well
covered by Lüddeckens and Thissen (1980–2000). A list of names in Coptic documentary
texts by M. Hasitzka (2007) is available on the Internet.

Ancient names almost always have meanings that their etymologies reveal. Because many
Egyptian names are known to us only in Greek transcription, the meanings of many are
not yet known, but many are either known with certainty or conjectured with some likeli­
hood. The collaboration of Greek papyrologists and Egyptologists has enabled much
progress in this area (the index of names in Pestman 1981 is a good example), but a great
deal remains to be done.

Most Egyptian names are nominal phrases, very often constructed with the name of a
god, such as “he who has been given by Isis” (Peteesis) or “the daughter of
Amoun” (Senamounis). Some are short sentences, and others are just the Egyptian defi­
nite article (masculine p-, feminine t-) with an adjective, noun, or ethnic (e.g., Pekysis,
“the Nubian”). Some of these names (e.g., Horos) are common throughout Egypt, but
most are local, limited to one or several places. This localism is mainly the product of
variations in Egyptian religion; children were named after the gods particularly venerat­
ed in their home village, city, or nome, whom parents thanked for the gift of the child.
Some also come from dialect differences among the regions of Egypt, as, for example,
with the alternation between /a/ and /o/ in the vocalizations of the name of the god Shaï
and of persons given names derived from it (Quaegebeur 1975).

Egyptian onomastic practice also changes over time; names come into or go out of fash­
ion sometimes reflecting religious change as well. The result of this situation is that,
when papyri from a provenance not previously well known come to light, they tend to
yield a large number of previously unknown names. If no demotic texts are available to
provide etymologies, Greek transcriptions may remain opaque. Editors of Greek texts
have often been lulled by these difficulties into an attitude of “anything goes” with re­
spect to Egyptian names, but the result of this approach is usually the reading of phan­
tom names. At the same time, it is true that a large proportion of personal names occurs
only once or twice in the documentation; rarity is thus not necessarily an indication of
misreading.

The Greek settlers in Egypt brought with them their own names, some of them common
throughout the Greek world but others particular to a given region or city. In particular,
many Macedonian names occur in the papyri because of the importance of the Macedon­
ian contingent among the earliest settlers. Ptolemaios is only the commonest of these.
Among them were many formed from the names of Greek gods, for the Egyptians were
hardly alone in constructing personal names in this manner; thus, we find names like
Diodoros (“gift of Zeus”). Alongside this (p. 193) repertory developed a body of Greek

Page 14 of 19
Practical Help: Chronology, Geography, Measures, Currency, Names, Proso­
pography, and Technical Vocabulary
names analogously formed from the appellations of Egyptian gods, like Besodoros (“gift
of Bes”) or Horion.

The papyri also contain numerous names from the Semitic languages, mostly brought by
settlers from regions like Judaea and Idumaea, whether recruited for the Ptolemaic mili­
tary or coming to Egypt for other economic reasons. The large Jewish population of Egypt
used some relatively colorless Greek names alongside some of more distinctively Jewish
character. After the large-scale slaughter of Egypt's Jewish population in the wake of the
revolt under Trajan and Hadrian, distinctively Jewish names become rare in the papyri.
Names found in the Hebrew scriptures, however, come back into prominence in late an­
tiquity with the Christianization of Egypt and the rebuilding of a Jewish community in
Alexandria and other Egyptian cities. With particular names, it is at times difficult to be
certain whether an individual is a Christian or a Jew.

Christianity also brought a range of other names that provide specific indication of reli­
gious allegiance. Some of these are the names of apostles and martyrs in the New Testa­
ment (Peter and Paul, in particular); others refer to qualities (Adelphios), to theological
doctrines or abstractions (Anastasia), or to saints and martyrs venerated throughout
Egypt or more locally. It is mainly through this last category that some theophoric pagan
names survive in late antiquity by means of their association with one or more saints
(Apollos and Phoibammon are characteristic examples).

Papyri written after the Arab conquest (cf. chapter 19) also attest Arabic names. At first
these are mainly those of high-ranking officials in the Arab administration, but as the pop­
ulation of Egypt began to convert to Islam, Arabic names are increasingly found among
the Egyptian population in general. In the Coptic papyri, this phenomenon is evident es­
pecially from the ninth century on.

Prosopography
The papyri mention many thousands of individuals bearing these varied names. The over­
whelming majority of these (mostly ordinary taxpayers or other inhabitants but some of
much higher status) are known from only one appearance. A smaller percentage are at­
tested more often, and a relative handful appear in numerous documents, usually be­
cause these belong to archives constituted by these individuals. Papyrologists and histori­
ans have created some tools aimed at inventorying the identifiable people of various
times and places, a scholarly subdiscipline called prosopography (the study of individu­
als), much used for Roman history and to a lesser extent for Greek history, but the sheer
mass of (p. 194) material has limited such efforts. The most extensive is the Proso­
pographia Ptolemaica, now most fully available on the Internet.

Prosopographies of the Roman and the Byzantine periods have been less viable enterpris­
es because the quantity of data is so much larger. They exist for particular periods for the
Arsinoite and Oxyrhynchite nomes, as well as for the localities of Antinoopolis and
Aphrodito; another prosopography is limited to the Coptic documentation from Thebes
Page 15 of 19
Practical Help: Chronology, Geography, Measures, Currency, Names, Proso­
pography, and Technical Vocabulary
(mainly the West Bank) at the end of Byzantine rule and in the first century of Arab rule.
No such works based on papyrological documents have been produced in the last twenty-
five years except for the digitization of the Prosopographia Ptolemaica (print version:
Peremans and Van't Dack 1950–). (For other printed prosopographies see Diethart 1980;
Jones and Whitehorne 1980; and Till 1954.) High-ranking officials in the papyri of the pe­
riod from Diocletian to Heraclius can often be found in the Prosopography of the Later
Roman Empire (Jones, Martindale, and Morris 1971–1992).

Technical Vocabulary
Preferring variation in diction, literary Greek was on the whole averse to the use of tech­
nical vocabulary. The Greek of the documents, however, has a rich repertory of words
used with specialized meanings for particular institutions, offices, taxes, legal acts, and
other administrative and legal purposes. Many of these were adopted in Egyptian and ap­
pear in the Coptic texts as well. There has never been a full study of the technical vocabu­
lary, and the only attempt to compile it is Preisigke (1915), now long out of date and per­
haps the least successful of his research tools in that it requires considerable experience
for profitable use. The specialized indexes of the Wörterbuch (Preisigke et al. 1925–) pro­
vide a kind of categorized listing of such vocabulary by domain but not always with defini­
tions. Pestman (1994, 283–304) presents a handy glossary with English definitions. For
Greek terms that represent Roman concepts or institutions, see Mason (1974), which, al­
though flawed, is often useful. Frequently, however, the simplest way to understand the
usage and meaning of technical terms is to search in the DDbDP for an attestation in a re­
cent volume that is well edited and has translations.

Bibliography
Bagnall, R. S. 1985. Currency and Inflation in Fourth-century Egypt. Atlanta: Scholars
Press.

———, A. Cameron, S. Schwartz, and K. A. Worp. 1987. Consuls of the Late Roman Em­
pire. Atlanta.

Bagnall, R. S., and K. A. Worp. 2004. Chronological Systems of Byzantine Egypt,

(p. 195)

2d ed. Leiden: Brill.

Banaji, J. 2002. Agrarian Change in Late Antiquity: Gold, Labour, and Aristocratic Domi­
nance. New York: Oxford University Press.

Bureth, P. 1964. Les titulatures impériales dans les papyrus, les ostraca, et les inscrip­
tions dʼÉgypte (30 a.c–284 p.c.). Brussels: Fondation égyptologique Reine Elisabeth.

Cadell, H., and G. Le Rider. 1997. Prix du blé et numéraire dans lʼÉgypte lagide de 305 à
173. Brussels.

Page 16 of 19
Practical Help: Chronology, Geography, Measures, Currency, Names, Proso­
pography, and Technical Vocabulary
Calderini, A., and S. Daris. 1935–. Dizionario dei nomi geographici e topografici
dellʼEgitto greco-romano. 5 vols. and 3 supplements to date. Milan.

Clarysse, W. 1985. “The Athenian Measure at Hermopolis.” ZPE 60: 232–236.

———, and G. van der Veken. 1983. The Eponymous Priests of Ptolemaic Egypt (P.L. Bat.
24): Chronological Lists of the Priests of Alexandria and Ptolemais with a Study of the De­
motic Transcriptions of Their Names. Leiden: Brill.

Diethart, J. M. 1980. Prosopographia Arsinoitica. Vol. 1. Vienna: In Kommission bei Verlag

Brüder Hollinek.

Drew-Bear, M. 1979. Le nome Hermopolite: Toponymes et sites. Missoula, Mont.: Schol­

ars Press.

Duncan-Jones, R. P. 1976a. “The Choenix, the Artaba, and the Modius.” ZPE 21: 43–52.

———. 1976b. “The Size of the Modius Castrensis.” ZPE 21: 53–62.

Falivene, M. R. 1998. The Herakleopolite Nome. Atlanta: Scholars Press.

Foraboschi, D. 1967–1971. Onomasticon alterum papyrologicum. Milan.

Gara, A. 1976. Prosdiagraphomena e circolazione monetaria. Milan.

Grenier, J.-C. 1989. Les titulatures des empereurs romains dans les documents en langue
égyptienne. Brussels.

Grzybek, E. 1990. Du calendrier macédonien au calendrier ptolémaïque. Basel.

Hagedorn, D., and K. A. Worp. 1994. “Das Wandeljahr im römischen Ägypten.” ZPE 104:
243–255.

Hazzard, R. A. 1995. Ptolemaic Coins: An Introduction for Collectors. Toronto: Kirk &
Bentley.

Jones, A. H. M., J. R. Martindale, and J. Morris. 1972–1992. The Prosopography of the Lat­
er Roman Empire. 3 vols. Cambridge: Cambridge University Press.

Jones, B., and J. Whitehorne. 1980. Register of Oxyrhynchites, 30 B.C.–A.D. 96. Chico,
Calif.: Scholars Press.

Kienast, D. 1990. Römische Kaisertabelle. Darmstadt.

Kruit, N., and K. A. Worp. 1999. “Metrological Notes on Measures and Containers of Liq­
uids in Graeco-Roman and Byzantine Egypt.” APF 45: 96–127.

Kruit, N., and K. A. Worp. 2000. “Geographical Jar Names.” APF 46: 65–146.

Page 17 of 19
Practical Help: Chronology, Geography, Measures, Currency, Names, Proso­
pography, and Technical Vocabulary
Lüddeckens, E., and H. J. Thissen. 1980–2000. Demotisches Namenbuch. Wiesbaden: Re­
ichert.

Maresch, K. 1994. Nomisma und Nomismatia: Beiträge zur Geldgeschichte Ägyptens im

6. Jahrhundert n. Chr. Opladen.

———. 1996. Bronze und Silber: Papyrologische Beiträge zur Geschichte der Währung im
ptolemäischen und römischen Ägypten bis zum 2. Jahrhundert n. Chr. Opladen.

Mason, H. J. 1974. Greek Terms for Roman Institutions: A Lexicon and Analysis. Toronto:
Hakkert.

Ogden, J. 1996. “Weight Units of Romano-Egyptian Gold.” In Archaeological Research in

Roman Egypt: The Proceedings of the Seventeenth Classical Colloquium of the Depart­
ment (p. 196) of Greek and Roman Antiquities, British Museum, held on 1–4 December,
1993, ed. Donald M. Bailey, 191–196. Ann Arbor, Mich.: Journal of Roman Archaeology.

Parker, R. A. 1950. The Calendars of Ancient Egypt. Chicago: University of Chicago Press.

Peremans, W., and E. van't Dack. 1950–. Prosopographia Ptolemaica. 10 vols. to date,
arranged systematically with a name index. Leuven: Bibliotheca Universitatis. http://
prosptol.arts.kuleuven.ac.be/.

Pestman, P. W. 1967. Chronologie égyptienne dʼaprès les textes démotiques (332 av. J.-C.–
453 ap. J.-C.). Leiden: Brill.

———. 1981. A Guide to the Zenon Archive. Vol. 1. Leiden: Brill.

———. 1994. The New Papyrological Primer, 2d ed. Leiden: Brill.

Preisigke, F. 1915. Fachwörter des öffentlichen Berwaltungsdienstes Ägyptens in der

griechischen Papyrusurkunden der ptolemäisch-römischen Zeit. Göttingen.

———. 1922. Namenbuch. Heidelberg.

———, et al. 1925–. Wörterbuch der griechischen Papyrusurkunden, mit Einschluss der
griechischen Inschriften, Aufschriften, Ostraka, Mumienschilder, usw. aus Ägypten.
Berlin.

Pruneti, P. 1981. I centri abitati dellʼOssirinchite: Repertorio toponomastico. Florence.

Quaegebeur, J. 1975. Le dieu égyptien Shaï dans la religion et lʼonomastique. Leuven.

Rathbone, D. W. 1983. “The Weight and Measurement of Egyptian Grains.” ZPE 53: 265–
275.

Samuel, A. E. 1962. Ptolemaic Chronology. Munich: Beck.

Scott, K. 1932. “Greek and Roman Honorific Months.” Yale Classical Studies 2: 201–278.

Page 18 of 19
Practical Help: Chronology, Geography, Measures, Currency, Names, Proso­
pography, and Technical Vocabulary
Shelton, J. C. 1977. “Artabs and Choenices.” ZPE 24: 55–67.

———. 1981. “Two Notes on the Artab.” ZPE 42: 99–106.

Skeat, T. C. 1969. The Reigns of the Ptolemies, 2d ed. Munich: Beck.

———. 1993. The Reign of Augustus in Egypt: Conversion Tables for the Egyptian and Ju­
lian Calendars, 30 B.C.–14 A.D. Munich: Beck.

Talbert, R., ed. 2000. Barrington Atlas of the Greek and Roman World. Princeton, N.J.:
Princeton University Press.

Till, W. 1954. Datierung und Prosopographie der koptischen Urkunden aus Theben. Vien­
na.

Vleeming, S. P. 1980. “Masse und Gewichte.” Lexikon der Ägyptologie 3: 1209–1214.

———. 1981. “The Artaba and Egyptian Grain-Measures.” In Proceedings of the Sixteenth
International Congress of Papyrology, ed. R. S. Bagnall, 537–545. Chico, Calif.: Scholars
Press.

Von Reden, S. 2007. Moneyin Ptolemaic Egypt: From the Macedonian Conquest to the
End of the Third Century BC. Cambridge: Cambridge University Press.

West, L. C., and A. C. Johnson. 1944. Currency in Roman and Byzantine Egypt. Princeton,
N.J.: Princeton University Press.

Roger S. Bagnall

Roger S. Bagnall is Professor of Ancient History and Director of the Institute for the
Study of the Ancient World at New York University. He is a papyrologist and historian
of Hellenistic, Roman, and Late Antique Egypt.

Page 19 of 19
Editing A Papyrus

Editing A Papyrus
Paul Schubert
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0009

Abstract and Keywords

The process of editing a papyrus is undeniably a central aspect in the field of papyrology.
This article notes that the task of a scholar who undertakes the edition of a papyrus re­
sembles that of a detective. Following some basic methodological principles, adding a cer­
tain amount of experience gained through contact with many texts, and using state-of-the-
art tools to find their way around an increasingly vast corpus of primary sources, papyrol­
ogists must fit together various pieces of a puzzle. The advent of electronic tools has
made possible the quick handling of a huge mass of data, thus changing substantially
many aspects of the way in which papyrologists edit their texts. Electronic tools will even­
tually supersede the main papyrological reference books. Supplements to Friedrich
Preisigke's Wórterbuch are becoming redundant now that scholars can search the same
data on the Internet.

Keywords: papyrus editing, textual criticism, papyrology, text editing, electronic editing tools, Friedrich Preisigke,
Wórterbuch

Introduction: The Tools of the Trade

The process of editing a papyrus is undeniably a central aspect in the field of papyrology.
Starting often with little more help than the document itself, which is seldom completely
preserved, a papyrologist must tackle many forms of writing. The edition of a papyrus
should give readers from other fields of specialization access to this material in its origi­
nal language, offer a translation in a modern language, and, most important, place the
new text into its broader environment, be it historical or literary.

Although papyri are an abundant source of information for our knowledge of daily life in
the ancient world, one could easily object that nobody needs yet another contract for the
sale of a donkey or another scrap from the first book of the Iliad. On the other hand, one
can argue that, when placed in the right setting, very few papyri lack any new informa­
tion for the reader. It is therefore all the more desirable that the edition of a new papyrus

Page 1 of 19
Editing A Papyrus

not only provide readers with a sound text and internal commentary but also inform them
of the text's addition to our knowledge about a particular topic.

As this chapter explains, the task of a scholar who undertakes the edition of a papyrus re­
sembles that of a detective. Following some basic methodological principles, adding a cer­
tain amount of experience gained through contact with many texts, and using state-of-the-
art tools to find their way around an increasingly vast corpus of primary sources, papyrol­
ogists must it together various pieces of a puzzle. The emerging picture then needs to be
interpreted in the light of findings made in related fields.

Several excellent introductions to the process of editing papyri have been written
(p. 198)

by some eminent specialists (Turner 1973, 1980; Youtie 1963, 1966, 1974). In the last
three decades, however, the advent of electronic tools has made possible—and necessary
—the quick handling of a huge mass of data, thus changing substantially many aspects of
the way in which papyrologists edit their texts.

Until the last quarter of the twentieth century, scholars handling a literary papyrus re­
sorted primarily to their general knowledge of ancient literature, thereby following in the
path of the pioneers of papyrology, such as Bernard Grenfell and Arthur Hunt. Many had
started learning Latin and Greek at secondary school and, by the time they were con­
fronted with the task of deciphering a papyrus, displayed an intimate familiarity with the
masterpieces of classical civilization. Excellent dictionaries were available, notably Lid­
dell and Scott's (1940). Lexica and concordances pertaining to the major authors con­
tributed to the identification of known texts; some had been produced by hand, while lat­
er the first computers helped to compile others. The now ubiquitous Thesaurus Linguae
Graecae (TLG) created at the University of California—Irvine did not exist, and when it
first came into being, it required the use of a computer dedicated almost exclusively to
the task of reading Greek texts.

Papyrologists who were dealing with Greek documentary texts were better equipped,
thanks mostly to the foresight of a retired German post office official, Friedrich Preisigke.
He not only created the Wörterbuch, a specialized dictionary devoted to Greek documen­
tary papyri, but also started a number of other useful tools (see chapter 3: the Namen­
buch for personal names; the Sammelbuch for papyri not published in an indexed volume;
the Berichtigungsliste for corrections to texts; and the Fachwörter for technical vocabu­
lary).1 To the list should be added Aristide Calderini's Dizionario dei nomi geografici
(1935–2007), a dictionary of all place names in Egypt. Scholars in related Welds (e.g.,
epigraphy) have often envied papyrologists for the diversity and thoroughness of their in­
struments. In Preisigke's day, papyrology was a relatively young discipline, and the bulk
of material was of manageable proportion; thus, he could claim to have taken into ac­
count nearly every papyrus that was available to him.

Such was the setting in which previous introductions to the process of editing papyri
were developed. Much of what was said then remains valid today. Nevertheless, it is now
much easier for a papyrologist to sort through a vast number of papyri—tens of thou­
sands of texts—at high speed, thus making it possible to look for parallels to a new item
Page 2 of 19
Editing A Papyrus

within a few minutes, where the same process would have required several hours not
long ago. On the literary side, the TLG can be easily consulted either with a CD-ROM or
on the Internet. Working on a small fragment, a papyrologist can now compare readings
of even a short series of characters with nearly every text preserved from ancient Greek
literature.

Another important aspect of the recent transition toward electronic tools is the increas­
ing integration of these tools: The Duke Databank of Documentary Papyri (DDbDP), which
contains the text of nearly every documentary text on papyrus in an (p. 199) easily read­
able form, is now linked to the Heidelberger Gesamtverzeichnis der Papyrusurkunden
Ägyptens (HGV), where the user will find an exhaustive catalogue of Greek and Latin doc­
umentary papyri, including date, place, and contents. Similar databases are now available
for Coptic documents and for demotic texts. Texts in the DDbDP also partially record cor­
rections to the original edition. In the United States, major papyrus collections are
presently catalogued in a standardized form through the Advanced Papyrological Infor­
mation System (APIS). Every papyrus can be examined from anywhere in the world
through the Internet; once a papyrus has been published, digital images are provided, as
well as a translation and commentary. Since papyrus collections and archives tend to be
scattered among many locations, such catalogues enable scholars from every country to
compare these fragments and make sense of the apparent disorder.

Electronic tools will eventually supersede the main papyrological reference books. Sup­
plements to Preisigke's Wörterbuch are becoming redundant now that scholars can
search the same data on the Internet. The same can be said of R. A. Pack's Greek and
Latin Literary Texts from Greco-Roman Egypt, which can be consulted in updated form as
an online database from the Centre de Documentation de Papyrologie Littéraire, along
with the Leuven Database of Ancient Books.

Decipherment: to and Fro between the Eye and

the Mind
The first phase in the process of editing a papyrus, however, does not require many so­
phisticated tools but rather a pair of sharp eyes and a certain amount of common sense.
Before even starting to decipher a text, a papyrologist will look at the papyrus from a dis­
tance. Its general state of preservation will offer a first idea of the gaps that may have to
be filled in later. The format of the sheet is also of importance: Obviously, there will be a
difference between a tax receipt written on a small slip and a register for a whole village,
which would require a roll of some length. In the case of literary papyri, the height of a
roll or the layout of a codex page will give the editor a first impression of what they may
contain. Margins are also worth taking into account; for instance, elaborate copies of
books tend to have more generous margins (cf. chapter 11).

Page 3 of 19
Editing A Papyrus

Papyri display a great variety of hands in both literary and documentary texts. Every level
of proficiency appears, from the skilled professional scribe producing a book or working
at the higher levels of administration in Egypt, to poorly trained individuals who can
hardly write their own name (Youtie 1971). It can therefore be of great help to assess the
quality of a hand before deciphering even a single word: (p. 200) Does the scribe know
how to write fluently? Is his hand fast? How easy would it be for his contemporary read­
ers to reclaim the meaning of the written text?

A first glance at the writing will also allow a papyrologist to make a rough estimate of the
date of a papyrus before starting to look for the possibility of an explicit date to confirm
it. Styles vary considerably throughout the Ptolemaic and Roman periods, and, on that ba­
sis alone, a trained reader should in most cases be able to narrow the dating of a papyrus
to within a century.2

Decipherment does not necessarily follow a straight path: Instead of starting at the upper
left corner of a sheet, one may find it more helpful to look first for the most easily read­
able sections. As in a puzzle, every piece added to the picture contributes to the reading
of more difficult parts. In a documentary text, the heading often provides crucial informa­
tion on the type of document under scrutiny. For instance, the mere presence of the word
παρά (para) at the beginning of the second line, followed by a name in the genitive (mean­
ing “from X”), will suggest that this is a document that follows the form of a hypomnema,
a memorandum, most often a request submitted to a person of higher standing by an indi­
vidual in a subordinate position. In other cases, the presence of a telltale χαίρειν
(chairein), or only the abbreviated form X— in the first lines of a document may signal the
format of a letter or perhaps a contract drafted in the form of a letter. Again, the pres­
ence of a date, accompanied by a titulature and a place name at the beginning of a docu­
ment, may mean that a contract was written by a scribe working for an official notary,
while private contracts usually show a date at the end of the document.

The same first quick assessment applies also to literary papyri. For instance, the pres­
ence of short horizontal strokes between the beginnings of two consecutive lines (para­
graphos) may indicate that the text is a dialogue, possibly a drama, or a lemmatized com­
mentary. Nomina sacra, the abbreviated forms of names like
with a horizontal stroke above the letters, immediately indicate a Christian context (Paap
1959). When the whole width of a column is preserved, the length of lines may suggest a
metrical unit: Hexameters are longer than iambic trimeters, and elegiacs alternate be­
tween longer and shorter lines. No single rule applies to each individual case, so common
sense clearly plays a crucial role.

Much research has been done on the process of reading modern languages, notably Eng­
lish. Although not every result from such studies can indiscriminately be applied to the
technique of deciphering ancient cursive scripts, certain findings can serve as useful
guidelines to papyrologists. I therefore summarize the most relevant elements here and
compare them with the technique of decipherment, which involves mostly trial and error.3

Page 4 of 19
Editing A Papyrus

Starting with the shape of the letters, the “absolute legibility of a single letter is not what
matters; it is relative confusability with other letters that is important” (Gibson and Levin
1975, 195). This is an experience that every papyrologist has had many times. In cursive
scripts, confusion between Η and Ν, B and K, or N and T I, for example, is very frequent.
Beginners often start deciphering a papyrus one letter (p. 201) at a time, whereas in gen­
eral one should concentrate on whole words. It is easier to read entire words than each
letter sequentially. With modern languages, adults typically read in units of words, al­
though this does not mean that we necessarily recognize a word only by its shape. Ac­
cording to Gibson and Levin (ibid., 197):

Research on the effectiveness of global form or contour as a basis for recognition

of the word has consistently tended to refute the idea. Overall word shape is not a
good enough differentiator, and children (as well as adults) do not use it. Not only
is differentiation poor without internal analysis, but without such analysis there
would be no basis for transfer to new words.

In fact, many factors make a word recognizable. Graphic cues such as dominant letters (in mod­
ern languages, capital letters; in a papyrus, a ϕ, for instance) play an important role. The begin­
nings (and to a lesser degree the endings) of words seem to be more helpful than the middle in
rendering a word recognizable. In reality, we seldom read single words but rather connected
text, which implies that the meaning of the words affects our capacity to read the text as a
whole.
When deciphering a papyrus, one is confronted not only with the straightforward process
of reading but also with the additional difficulty of having to restore the text where the
papyrus has been damaged. In this, the psychology of modern reading allows us to draw a
parallel with the decipherment of ancient texts. For historical reasons, the Latin alphabet
used in many European languages bears a resemblance with Greek cursive script. Gibson
and Levin (ibid., 170) state that “if the text is mutilated so that only the top half or the
bottom half is visible, it is reasonably easy to read the top and very difficult to read the
bottom. The preponderance of distinctive features exists on the top of the letters.”

Coming back to the decipherment of papyri, if we leave aside very neat capital letters
(where each letter is separated from the next and has an easily recognizable shape), it
makes little sense to try to decipher words one letter at a time. Words are most often
barely separated, and scribes use punctuation only rarely. Typically, in some documents
from the end of the Ptolemaic period, a scribe often lifted his pen not between two letters
but within a letter itself. He then linked it to the next without lifting his pen.4

The shape of letters, however, does matter. Although the same letter may have two differ­
ent shapes in the same papyrus, one should look for the distinctive features of the most
typical letters. For instance, if the scribe systematically ends the descending stroke of ρ
with a hook but does not do so when writing a φ, this will become useful information
when only the lower end of a descending stroke is preserved.

Page 5 of 19
Editing A Papyrus

Moving away from the individual letters, a reader of papyri will try to grasp whole words,
sometimes sequences of words. The process of deciphering then depends heavily on a
movement from the eye to the mind and back again: Papyrologists incessantly check their
readings against their prior knowledge of Greek (or whatever ancient language is in­
volved) to match the signal caught by (p. 202) their eyes in a sequence of letters that
makes sense in their memories, at least at the basic level of vocabulary and morphology.
In ancient as in modern languages, “only certain combinations of letters or words can be
a word. Knowledge of these rules even in the skilled reader is tacit rather than
explicit” (Gibson and Levin 1975, 224). A sequence of strokes and curves will suggest to a
papyrologist perhaps not a full word but a root or an ending. One then compares this first
result with a number of possibilities and reverts to the papyrus, testing those hypotheses
iteratively against the written material.

Because isolated words are seldom of much use, the next step obviously consists of trying
to link several words into a sentence. Documentary papyri tend to use standard expres­
sions that recur in many instances. A basic experience of the usual phraseology is there­
fore a great help to a papyrologist while reading new texts. Beginners will improve their
deciphering skills by getting acquainted with a generous selection of texts in one of sev­
eral available sourcebooks.5 Even in the case of a perfectly preserved papyrus, a sea­
soned scholar may at first fail to comprehend certain passages because the cursive writ­
ing of particular scribes can be very confusing. For instance, imperial titulatures are of­
ten written in a fast scribble in which the individual letters are hardly recognizable. The
most extreme cases show little more than a sharply undulating wave for words such as
εὐτυχο ῦ ς, εὐσεβοῦς, or σεβαστοῦ (corresponding to Latin Felix, Pius, and Augustus).
With a basic knowledge of the expressions that frequently appear in such passages, the
reader can hope to identify some words from the shape of one or two letters. For in­
stance, the χ of εὐτυχο ῦ ς usually stands out from the illegible scribble.

Deciphering a papyrus is of course only part of the process of editing; the next step is to
transcribe the text into a form that other scholars will understand. For this purpose, pa­
pyrologists have agreed on a system of dots and brackets that indicate the state of the
original papyrus and the level of confidence with regard to a particular word or phrase.
Those rules were adopted in 1931 and are widely known as the “Leiden system” (table
9.1).6

Arguably the single most important item in this list is the dot, the purpose of which is to
warn the reader of an uncertain reading. “Van Groningen [originator of the Leiden sys­
tem] used to say with regard to the rule that uncertain letters should be dotted: ‘The dot
is a papyrologist's conscience’.”7 Papyrologists, as well as nonspecialists, who use papyro­
logical editions must therefore tread carefully when using a passage that has been heavi­
ly dotted—or bracketed. Simultaneous use of different types of brackets is also to be con­
sidered with caution.

When a text is badly corrupted, this is indicated in the apparatus, a section below the text
where the reader will find relevant information on the state of the original papyrus. In the

Page 6 of 19
Editing A Papyrus

case of documentary papyri, editors do not correct spelling mistakes in the edited text but
give the standard spelling in the apparatus. A single transcript is sufficient for most prac­
tical purposes. Words are separated, punctuation is supplied, (p. 203) breathings and ac­
cents are added, and capital letters used for proper names. Accents, diaereses, or punctu­
ation found in the papyrus are indicated in the apparatus.

Table 9.1. Editorial Conventions of the Leiden System

1. letters about the reading of which there is a genuine doubt α̣ β ̣ γ ̣ δ ̣

or that are so mutilated that without the context they might
be read in more than one way

2. illegible letters, the approximate number of which is Known … or −10− or

± 10

3. missing letters, the approximate number of which is Known […]or

[−10−]or[ ±
20]

4. missing letters, the number of which is unknown ] or [ ] or [

5. letters restored by the editor of the text [αβγδ]

6. lacunae in the text (omissions of the scribe) <> or ***

7. additions made by the editor in order to fill such lacunae < αβγδ >

8. resolutions of abbreviations (e.g., γρ− = γρ(αμματεύς)) (αβγδ)

9. interpolations (i.e., letters or words wrongly added by the {αβγδ}or{…}

scribe and canceled by the editor of the text) or{αβγδ}

10. erasures by the scribe [[ αβγδ]]

11. interlinear additions that would be impractical to print be­ ὰβγδ́

tween the lines

With a literary papyrus, if this is a new text, it is advisable to produce two transcripts.
The first (“diplomatic transcript”) will provide the reader with the text precisely as it is
read on the papyrus; it should retain the scribe's spelling, including errors. The second
(“full transcript”) will correct these errors and supply the reader with word breaks, punc­
tuation, breathings, and accents. In the case of a well-known literary text (e.g., a frag­

Page 7 of 19
Editing A Papyrus

ment from the Odyssey), however, one transcript is enough; the spelling found on the pa­
pyrus is usually given in the text, with a note in the apparatus where necessary.

Restoration: The Search for Parallels

Cursive writing in which the shape of every letter is not self-evident can by itself be cause
for trouble, but it can also be aggravated by another frequent problem: holes of various
sizes in the papyrus. Therefore, in deciphering a new text, a papyrologist often cannot re­
ly solely on shape recognition and approximate feeling of the language to produce a de­
cent transcript but must use parallels from previously published texts.

Looking for parallels has always been a tedious task that requires access to a
(p. 204)

well-furnished library. Until recently, references to possible parallels had to be located in

standard dictionaries; after this, the papyrologist would go on to check the edition of the
texts. Few university libraries possess a nearly complete set of classical Greek literature,
let alone a generous selection of editions of documentary papyri. New electronic tools,
however, have opened the field of papyrology to scholars working in universities of more
modest means, where they now have access to texts from around the world. Electronic
databases enable one to search not only for a specific word but also for a few letters in a
word or for several words in the same context. A search can be further refined by specify­
ing a time span (this is when a rough palaeographical estimate is most useful) or, in the
case of Graeco-Roman Egypt, a particular village or nome. With literary fragments, one
can also search by genre, date, or place. Recourse to those new tools, however, does not
obviate the need to check the original editions later on, when the time comes to add some
flesh to the skeleton: Reliance on electronic media alone too often proves treacherous
and can lead to serious mistakes. Moreover, sound scholarship always relies on extensive
reading of the original sources, regardless of technical progress.

At this point in the process of editing a papyrus, a scholar may follow different paths de­
pending on the type of text under study. Literary and documentary papyri can present the
editor with various situations, where the text under scrutiny either conforms to a previ­
ously known model or displays no immediate parallel.

A literary papyrus consisting of a fragment from a major work that was independently
copied down to the Renaissance is probably the easiest case to handle. A search for paral­
lels might show, for example, that a small scrap of papyrus dating from the second centu­
ry ce contains a few broken lines from one of the speeches of Demosthenes, the orator
most commonly found in Greek papyri. The papyrologist's task would then consist of
restoring (with the help of a modern edition of Demosthenes) the appearance of the origi­
nal papyrus roll; the reading of the papyrus is compared with the modern edition (i.e.,
with the testimony of Byzantine manuscripts); and variations in the transmitted text are
identified. This is quite often a relatively straightforward process, which helps us mainly
to evaluate the state of the textual transmission at a given period in relation to our Byzan­
tine manuscripts.

Page 8 of 19
Editing A Papyrus

Figure 9.1. P.Gen. inv. 500.

Matters can become more difficult when a search for parallels with a literary fragment
yields no suitable match. Complete rolls or codices were a rarity even when papyrology
was in its infancy. Nowadays the typical unpublished literary fragment rarely contains a
whole column of text. Faced with a previously unknown source, the editor focuses on
smaller details in the search for parallels, even though this does not always lead to a com­
plete text. figure 9.1 is an example of (p. 205) a potentially interesting literary fragment
(P.Gen. inv. 500) that still awaits a full interpretation.

Since P.Gen. inv. 500 is a previously unknown text, both a “diplomatic” and a “full” tran­
script are provided: The restorations made in the full transcript suggest that we are deal­
ing with a shrine of Hermes (line 4: ί]ε ρὀυ ‘E ρμοῦ). It seems that someone has been in­
structed (p. 206) to set up a statue (line 5: ἀ]ṿaθεĩυαι ξό[ανον). There is also a mention of
either building or rebuilding (line 9: άνοικοδομη[). Those elements point to a mythologi­
cal narrative, perhaps explaining the erecting of a statue. The most promising parallel ap­
pears in the story of Hermes Perpheraios.8 In a prose summary of one of Callimachus's
Iambi, we learn that some fishermen from Ainos in Thracia found a statue in their net.
Recognizing this as an image of a god, they set it up in their city. This is an interesting
parallel, which could help a papyrologist in interpreting the new papyrus fragment, al­
though the argument in favor of linking it to the Callimachean narrative remains rather
weak. A firm conclusion can sometimes elude a papyrologist for many years until another
clue shows up in a new papyrus.

1 (1st hand) ]‥[․]κ̣τηρ̣̣ [ ]‥[․]․κ̣τηρ̣[

2 ]․[․]․α̣ π̣ οτ.[ ]․[․]․α ̣ π ̣ οτ ․[

3 ]. γωκατασιτ ̣[ ]․γω κατ ὰ σιτ̣[

Page 9 of 19
Editing A Papyrus

4 ]ερονερμουεị.[ ἱ]ερòν ‘E ρμοῦ εị.[

5 ]․α θειυαιξο[ ả] ṿ αθεîυαι ξó[ανον

6 ].υπο.[ ].υπο.[

7 ] ῳιεp.[ ] ῳιεp.[

8 ]ουθησεοτησκ[ ἐπηκολ]οὑθησε ὁ τῆς κ[

9 ]ουαυοικοδομη[ ]ου ἀυοικοδομη[

10 (2nd hand) ].εịωμελ.[ ].εί ῳ μελ[

11 (1st hand) ] ῳ v.[ ] ῳv.[

Let us now return to documentary papyri. Not all texts are equally likely to conform to a
given model. Scribes producing property returns, for instance, usually stick to a standard­
ized phraseology that allows papyrologists to fill in considerable gaps in badly damaged
papyri; the same can be said of petitions and many other types of documents. Contracts
from the Ptolemaic period often begin with a long list of so-called eponymous priests who
had charge of various dynastic cults in honor of the ruling family. These lists follow a
standard form, and it is a relatively straightforward process to restore them to the full
breadth of a column even when much is actually missing. Once a few lines are confidently
filled in and one can estimate the width of the sheet, it is easier to work on the rest of the
text, where the formulation may vary depending on the object of the contract.

The contents of mutilated private letters, on the other hand, are notoriously difficult to re­
store because of several factors that conspire to make a papyrologist's task more ardu­
ous: After the usual greetings that the sender shares with the addressee, the sender often
continues by alluding to matters familiar to both parties but of which we have no direct
knowledge. Because the contents of private letters frequently revolve around the daily
life of Egypt's rural population, a myriad of scenarios may confront the modern reader,
and a writer's inadequate knowledge of morphology and syntax can contribute to making
the text still less intelligible.

Whatever the document's level of standardization and regardless of the quantity of paral­
lels found in other documents, a few simple rules can help one avoid editorial errors:

• The scribes were no fools. When faced with phrasing that makes no sense, a papyrol­
ogist should consider what he has not understood before assuming that the scribe was
under the influence of strong beer while writing the document. One must also remem­
ber that, at the receiving end, scribes expected their readers to be able to understand
what they were writing. This general principle should, however, be mitigated by the

Page 10 of 19
Editing A Papyrus

fact that the level of instruction among scribes varied considerably. A particu­
(p. 207)

lar scribe might have been incompetent, but a papyrologist should resort to such an
explanation only after exhausting other hypotheses more respectful of the scribe's
skills.
• The closer an editor is to a gap in the papyrus, the more careful he should be. The so-
called Lex Youtie was formalized by Reinhold Merkelbach and neatly summarized by
the Latin motto iuxtau lacunam ne mutaveris (“next to a gap, thou shalt not alter [the
text]”).9 In other words, assuming scribal error is most perilous when one does not
know what was written next.
• A sheet of papyrus had its cost, and one can therefore expect scribes to use the full
width of their page. Left and right margins in a document are not necessarily perfectly
vertical, but one should nevertheless aim for a relatively even line length in the
process of restoring a partially mutilated column. If, in a sequence of four lines, the
text produces lines of 23, 25, 13, and 24 characters, the odds are good that the miss­
ing part of the text in line 3 has not been properly restored. This last rule should, of
course, not be followed with excessive rigidity since exceptions are not infrequent.10
For example, the layout of a heading sometimes requires that an element start on a
new line, thus causing an indentation or a different spacing of the letters in the pre­
ceding line.

Translation: When It All Makes Sense

The early editors of many papyri did not bother to translate the texts they had deci­
phered, assuming that any decent scholar would know Greek (or any other ancient lan­
guage) well enough to make this step unnecessary. It is all the more remarkable that
Bernard Grenfell and Arthur Hunt, the founders of the Oxyrhynchus Papyri series, decid­
ed to include a translation of all of the texts they were editing. Did they believe that
British readers had less command of the classical languages than their continental coun­
terparts? It would seem that they had in fact two excellent reasons for translating their
texts into English.

First, Grenfell and Hunt depended heavily on funding from the Egypt Exploration Fund
(now known as the Egypt Exploration Society) for their excavations at Oxyrhynchus.
Therefore, they had to appeal to a wide readership so as to convince their benefactors of
the relevance of their work in Egypt. One can see how, in the first volumes of the series,
they carefully included some new Christian texts, presumably in order to attract the inter­
est of the British clergy. Numerous subscribers of the Graeco-Roman Memoirs, as the se­
ries (including the Oxyrhynchus (p. 208) Papyri) was called, would have known some
Greek and Latin without having pursued an academic career. Translations thus served the
general purpose of making the highly specialized contents of the papyri available to a
wide public of enlightened amateurs. It was a shrewd move on the part of the two Oxford
scholars and one that classical scholars should not forget.

Page 11 of 19
Editing A Papyrus

The second good reason was that a translation could be considered as the first step to­
ward a commentary. Some papyrologists, considering the translation to be the ultimate
result of editing a papyrus, place it at the end of the edition, after the commentary. No
clear consensus has emerged on this practice. Things have not changed since the days of
the pioneers: While editing a papyrus, one frequently finds that serious problems arise
just as one starts translating the text into one's own language. What seemed obvious sud­
denly becomes more obscure, and one's brilliant conjectures for filling in a gap turn out
to make little sense or even to contradict the rules of ordinary grammar. A translation
thus serves a double purpose: to make the editor verify that the text makes proper sense
and to convey the meaning to readers in a way that they can understand.

A good translation should fulfill several criteria that are not easily reconciled. Given the
fact that we are most often dealing with an editio princeps, precision should take prece­
dence over literary elegance. As much as possible, the translator should endeavor to keep
the structure of the original sentence, although this is not always possible. Greek parti­
cles are very important for this purpose since they play the role of our modern punctua­
tion and can quite often be interpreted as such in a modern language.

Technical terms provide the translator with a difficult challenge. What should we do with
words like epikrisis, dioikêtês, or embadikon? A specialist will no doubt understand them
in many cases without a proper translation, and some will claim that translating them can
only obscure their meaning. Having sometimes made what I now believe to be the wrong
choices, I would argue in favor of an appropriate translation and not a mere translitera­
tion. Thus, for those who have not been initiated into the intricacies of papyrology, epikri­
sis will be more understandable if it is called “examination of civic status,” and dioikêtês
can be translated as “finance minister” or “financial manager,” depending on the context.
As for embadikon, which literally means “tax for moving in,” one might consider using
“real estate purchase tax.”11

Commentary: The Details that Matter

Editors of papyri can differ greatly in the way in which they write a commentary, al­
though they will likely agree on the basic purpose of the task: It should first provide justi­
fication for the choices made in the edition of the original text, especially when (p. 209)
they rest upon conjecture; it should also explain to readers what is new in the papyrus
and then give them information on certain technical matters. One can assume that the
broad outline of interpretation has already been dealt with in an introduction.

Dating a papyrus is of course of the utmost importance. Documentary papyri often con­
tain a date accurate to the day, and tables provide papyrologists with convenient means
of translating various systems into our Julian calendar.12 Geography also significantly af­
fects the practice of scribes. The origin of a papyrus can be identified through not only
the explicit mention of place names but also the presence of a specific personal name, a
turn of phrase in an administrative document, or some phonological peculiarity. For in­
stance, when a man bearing the name Stotoetis shows up in a papyrus, there is a high
Page 12 of 19
Editing A Papyrus

probability that the document comes from the village of Soknopaiou Nesos, on the north­
western shore of Lake Moeris, where this name is very common; again, the names Antio­
chos and Theon for a father and a son suggest Oxyrhynchus (P.Louvre II 100). Confusion
between the sounds /r/ and /l/ (e.g., φόλετρον written for φόρετρον) is typical of the
Egyptian dialect spoken in the Arsinoite nome and can therefore be taken as a clue that
the scribe is a native Egyptian who lives in that particular area (Gignac 1976, 102–107).

When stumbling upon a technical term (i.e., a poorly attested office or a rare verb denot­
ing a specific action in a procedure), readers expect to find some explanation in the com­
mentary. The electronic version of the Brussels Bibliographie papyrologique (on CD-ROM
only), now containing material for years starting in 1932, has made it much easier to lo­
cate the relevant information on many topics, although by using this method one will also
miss much specific information lurking, for instance, within the commentary to previously
published papyri.

The search for parallels has already been described. In the commentary, they are useful
for explaining an unusual wording or for justifying the reconstruction of the text in a dam­
aged section of the papyrus. Some editors choose to leave no stone unturned and offer ex­
haustive lists of parallels to their readers, who thus know that, in the work under consid­
eration, no evidence has been overlooked; on the other hand, abundance of material can
distract one's attention from substantive discussion or the most useful parallels.

Illustrations: Unless I See, I shall not believe

Because the image of a papyrus enables readers to check the decipherment for them­
selves, it can be considered an indispensable part of any papyrus edition. Unfortunately,
plates are also the most expensive part of a book, which explains (p. 210) why some edi­
tors choose to display only the most important papyri in a volume. In the case of small
scraps of little interest, this attitude is justifiable on the grounds that the cost of produc­
ing a plate exceeds its usefulness; such bits usually end up being only described rather
than receiving a full edition.

The way in which papyri are illustrated has changed dramatically in the past two
decades. For about a century, the standard illustration was a printed image in black and
white. For most practical purposes, color mattered little because carbon ink, which is
most widespread in papyri, contrasts well with the background of the papyrus sheet. Only
recently has color been introduced on a regular basis in some papyrus editions.

The biggest change, however, lies not so much in the advent of color as in illustrations
supplied by means other than paper. A short-lived attempt to provide readers with images
on microfiche (e.g., BGU XV) met with limited success. Microforms are not always easy to
handle, and reading them requires machines that libraries are becoming reluctant to
maintain and that few individuals own.

Page 13 of 19
Editing A Papyrus

The revolution in image-display technology was thus triggered by the advent of databases
available on the Internet. Papyrus collections around the world are gradually being
scanned and catalogued in such a way that external users can access the images and dis­
play them on their screens wherever they are working. Images can be produced in many
different formats, depending on the intended use, and—with the proper software—can be
treated so as to enhance contrast or move fragments around. This recent development
will influence the working methods of papyrologists in several ways.

First, in the not-too-distant future, costly plates in papyrus editions will tend to be re­
placed by links to images accessible on the Internet. This will certainly make sense from a
financial point of view, although one may worry about the long-term stability of such im­
ages and their references. This question, however, belongs to a debate that far exceeds
the scope of the present discussion. Second, it will become increasingly easy to check al­
most instantly a reading on a papyrus stored anywhere in the world. Third, papyrologists
will be able to browse through whole collections—in some cases also unpublished texts—
and look for stray parts of a document or for papyri belonging to an archive.

Major Editions: Of Dwarves Standing on the

Shoulders of Giants
The image of “dwarves standing on the shoulders of giants” was originally applied by
Bernard de Chartres in the twelfth century to describe the debt that scholars of (p. 211)
his day owed to the great minds of antiquity. The same image can be applied—mutatis
mutandis—to present-day scholars in relation to the pioneers who produced the first edi­
tions of papyri more than a century ago.

In the vast field of papyrology, the hard core is made up of Greek—and to a lesser extent
Latin—documentary texts, and not surprisingly it is around such texts that the process of
editing papyri is most coherently organized. If one moves away from this center of gravity
toward literary papyri (and especially toward other languages such as demotic, Coptic or
Arabic), the editorial process has until recently been conducted in a less orderly and stan­
dardized manner. This situation, however, is evolving rapidly, as any user of the Checklist
of Editions (Oates et al. 2001) will have noticed: In its latest printed version, it includes
demotic and Coptic papyri; Arabic papyri are also registered in a separate checklist. The
following survey therefore inevitably reflects the disproportionate amount of attention
that was devoted to Greek documentary papyri for a little more than a century. The read­
er should nevertheless bear in mind that many things said about this particular category
of papyri also apply to areas considered peripheral until quite recently.

With the exception of a few items, the publication of papyri on a systematic scale began
only in the late nineteenth century. Not surprisingly, the institutions holding the largest
collections started the most prestigious series, among which the Oxyrhynchus Papyri is
probably the best known. Bernard Grenfell and Arthur Hunt edited both literary and doc­
umentary texts from the city of Oxyrhynchus in middle Egypt at the impressive rate of

Page 14 of 19
Editing A Papyrus

one volume a year without compromising on quality. Their very high standard of scholar­
ship has been kept alive for more than a century, and many papyrologists from around the
world contribute to keeping up with the yearly installments. The Oxyrhynchus Papyri have
also served as a model for other papyrus editions with regard to general layout, transla­
tion, commentary, and indexes. This has led to the adoption of a number of editorial
guidelines, together with the Leiden system of conventional signs (Bell 1932; Hunt 1932).

With less speed but comparable regularity and quality, the National Library in Vienna has
edited the Corpus Papyrorum Raineri for more than a century. Other major series have
met various destinies, following the fate of the institutions where the papyri were pre­
served. Thus the Berliner Klassikertexte (BKT) and Berliner griechische Urkunden (BGU)
both had remarkable beginnings, but the Second World War and the Cold War took their
toll in spite of the publication of several BGU volumes in recent years. The Strasbourg col­
lection was established at the time when Alsace was German, and the first two volumes
were produced by Friedrich Preisigke, the founder of most of the major papyrological in­
struments. After the Second World War, the output resumed under French authority. In
Egypt, publication of papyri was in the hands of scholars from the former colonial powers
—especially Britain and France—and virtually ceased after that time. In Italy, the Papiri
della Società Italiana also stalled shortly after the Second World War.

On the whole, one can argue that today the liveliest series are those where the pa­
(p. 212)

pyri are integrated in a center of higher learning. Whereas superb collections at the
British Library in London and at the Louvre Museum in Paris receive only sporadic atten­
tion from scholars, some of the best papyrus editions are being produced at a steady rate
by universities that possess large collections, as, for instance, those in Ann Arbor (Michi­
gan Papyri), Cologne, and Heidelberg.

There has never been a successful attempt to produce in print a unified corpus of all pa­
pyri. At an early period, Ulrich Wilcken gathered and reedited in his monumental Urkun­
den der Ptolemäerzeit many important Greek documentary papyri from the Ptolemaic pe­
riod that were published in the nineteenth century. Although the material from that peri­
od has increased enormously since Wilcken's time, his edition remains one of the greatest
achievements in papyrology. On an even more ambitious scale, a Corpus Papyrorum Grae­
carum started with two volumes in which the editors had assembled every text of a given
type (in this case, notifications of death and contracts for wet nurses), but the project was
then discontinued. The closest thing to a unified corpus of papyri is the electronic Duke
Databank of Documentary Papyri. In the not-too-distant future, a complete network of dig­
ital tools should make it possible for any scholar to create a custom-made corpus on virtu­
ally any selection principles by drawing metadata from various sources and texts extract­
ed from the Duke Databank.

One area where near exhaustiveness has been achieved is in the collecting of Greek docu­
mentary papyri scattered in various journals or more generally outside of standard text
editions. We owe the so-called Sammelbuch to the initiative of Friedrich Preisigke; this in­

Page 15 of 19
Editing A Papyrus

strument has been continued up to the present time in installments that cover several
years.

The relative rarity of Latin papyri enabled Robert Cavenaile to cover the material in his
Corpus Papyrorum Latinarum, a volume of fairly modest size. This book has, however, be­
come in part obsolete as many more Latin papyri have been published in the half-century
since its publication. A comprehensive survey of all Latin papyri and parchments that pre­
date 800 CE has resulted in the publication of the magnificent Chartae Latinae An­
tiquiores, a series that will include (when completed) an edition and a plate of every sin­
gle text.

Textual Criticism: No Text is Perfect

A first edition of a papyrus is seldom flawless, as one can see from consulting the—as yet
—eleven volumes of the Berichtigungsliste, where (in principle) all corrections to papyri
are recorded. The editing of papyri is therefore an evolving process. A practical example
taken from the revision of the first (p. 213) volume of the Geneva papyri illustrates how
one can first spot a misreading, then correct the text, and finally draw some conclusions
from the newly established interpretation. It also shows how the confrontation of two
wrong readings can produce a better text and contribute to refining the prosopography of
Roman Egypt.

Dating from ca. 87 CE, P.Gen. I 4 is a complaint sent to a high official, the iuridicus
Alexandreae, by a man who claims that he was registered in the wrong civic category. He
suggests that the iuridicus (judicial official) write to the strategos (governor) of the divi­
sion of Herakleides in the Arsinoite nome to clear up the matter. The name of the strate­
gos (lines 17–18) was first read as…ηλíῳ | ι̣κ̣ ι and soon corrected to… ηλí ῳ |
‘I[π]πoκpάτει (Wilcken 1906, 380); Wilcken excluded Ạ ύ̣ ρ̣ ηλί ῳ. A century later, closer
examination of the original produced the nomen ‘ ʼỊ ο̣ υ̣ λίωι. It is, however, the reading of
the cognomen that offers the most interesting insight into the process of correcting the
text.

After Wilcken's time, this Hippocrates was found to have a near homonymous colleague,
also a strategos of the same division, in a papyrus from Vienna (P.Vind. Bosw. 1.35), dated
shortly after 87: Σ ̣ ωκρ̣ ά̣ τ̣ ης. The occurrence of two strategoi in the same division at vir­
tually the same time, and bearing the names Socrates and Hippocrates, could only raise
the suspicion that at least one of the two was misread. This is where the dots used to
mark uncertain letters are most useful, despite the fact that their use often reflects an in­
dividual papyrologist's subjectivity. How can we reconcile ʼI[π]ποκράτει with Σ ̣ ωκρ̣ ά̣ τ̣
ης? A close examination of the Geneva papyrus shows that the gap at the beginning of the
word is too narrow for ʼI[π]πο-but that another name will fit: ʼI[σ]ο̣ κράτ̣ε̣[ι]. An examina­
tion of the Vienna papyrus has confirmed that Isocrates is also to be read there. Having
been given a false identity for more than a century, the strategos Iulius Isocrates is at last
saved from oblivion.

Page 16 of 19
Editing A Papyrus

In more general terms, documents on papyrus display a regularity that makes papyrolo­
gists beware of exceptions. If these occur, they should be justified as far as possible. Un­
paralleled personal names, grammatical oddities, and geographical or chronological in­
consistencies should alert a reader to the possibility of an erroneous reading. The process
of editing a papyrus therefore never ends.

Bibliography
Bell, H. I. 1932. “Notes on Methods of Publication.” CdÉ 7: 270–271.

Calderini, A., and S. Daris. 1935–2007. Dizionario dei nomi geografici e topografici
dellʼEgitto greco-romano. Cairo: Società Reale di GeograWa dʼEgitto; Madrid: Consejo su­
perior de investigaciones cientificas; Bonn: R. Habelt; Milano: Fabrizio Serra.

Fassino, M. 1998. “Sulla cosiddetta ‘lex Youtie.’ ” Rivista di filologia 126: 72–75.

Gibson, E. J., and H. Levin. 1975. The Psychology of Reading. Cambridge, Mass.: MIT
Press.

Gignac, F. T. 1976. A Grammar of the Greek Papyri of the Roman and Byzantine Periods.
Vol. 1. Milan: Istituto editoriale Cisalpino-La Goliardica.

Hengstl, J., ed. 1978. Griechische Papyri aus Ägypten als Zeugnisse des öffentlichen und
privaten Lebens. Munich: Heimeran.

Hunt, A. S. 1932. “A Note on the Transliteration of Papyri.” CdÉ 7: 272–274.

———, and C. C. Edgar, eds. 1932–1934. Select Papyri. Vols. 1–2. Cambridge, Mass.: Har­
vard University Press.

Kramer, B. 1997. “Der κτίστης Boethos, und die Einrichtung einer neuen Stadt I.” APF 43:
315–339.

Liddell, H. G., and R. Scott (eds.), H. S. Jones. (rev.). 1940. A Greek-English Lexicon, 9th
edition. Oxford: Oxford University Press.

(p. 215) Merkelbach, R. 1980. “Lex Youtie.” ZPE 38: 29.

———. 2003. “Iuxta lacunam ne mutaveris.” ZPE 142: 34.

Paap, A. H. R. E. 1959. Nomina sacra in the Greek Papyri of the First Five Centuries A. D.
Leiden: Brill.

Pestman, P. W. 1990. The New Papyrological Primer. Leiden: Brill.

Rossum-Steenbeek, M. van. 1998. Greek Readers' Digests?: Studies on a Selection of

Subliterary Papyri. Leiden: Brill.

Skeat, T. C. 1954. The Reigns of the Ptolemies. Munich: Beck.

Page 17 of 19
Editing A Papyrus

Turner, E. G. 1973. The Papyrologist at Work. Greek, Roman, and Byzantine Monographs
6. Durham, N.C.: Duke University.

———. 1980. Greek Papyri: An Introduction, 2d ed. Oxford: Clarendon Press.

Van Groningen, B. A. 1932. “Projet dʼunification des systèmes de signes critiques.” CdÉ 7:
262–269.

Wilcken, U. 1906. “Zu den Genfer Papyri.” APF 3: 369–404.

Youtie, H. C. 1963. “The Papyrologist: Artificer of Fact.” GRBS 4: 19–32 [= Scriptiunculae,

vol. 1, 9–23].

———. 1966. “Text and Context in Transcribing Papyri.” GRBS 7: 251–258 [= Scriptiuncu­
lae, vol. 1, 25–33].

———. 1971. “Bραδέως γράφων: Between Literacy and Illiteracy.” GRBS 12: 239–261 [ =
Scriptiunculae, vol. 2, 629–651].

———. 1974. The Textual Criticism of Documentary Papyri: Prolegomena, 2d ed. London:
Institute of Classical Studies.

Notes:

(1.) Reference to all of those standard reference books appears in Oates et al. 2001.

(2.) Useful sets of plates are listed in ibid. (74–75). The two first volumes of P.Mert. also
offer numerous plates presented mostly in chronological order.

(3.) Gibson and Levin (1975). Although not the most recent work on the subject, Gibson
and Levin's book focuses on many aspects that can—at least remotely—be connected with
the practice of reading papyri.

(4.) See, for example, P.Ryl.II 73.8 (pl. 3; 32–31 BCE): χαίρειν. ἀττέχομεν παρ[ά σοῦ κτλ.
Notice the ν of χαίρειν, which is written in three distinct strokes; the last is directly
linked to the following α of ἀ πέχομεν. Again, at the end of ἀ πέχομεν, the ν is written in
three strokes, the last of which is linked to the following παρ[ά.

(5.) Hunt and Edgar (1932–1934) with English translation; Pestman (1990) with no trans­
lation but a good commentary for beginners; Hengstl (1978) with German translation.

(6.) Van Groningen (1932). The English version presented here is from Turner (1980,
187–188).

(7.) Pestman (1990, 15). To avoid confusion with punctuation, the dot should always be
set below line level, even when there is no letter above it.

(8.) Callimachus fr. 197; Rossum-Steenbeek (1998, 268).

Page 18 of 19
Editing A Papyrus

(9.) Merkelbach (1980, 2003); for criticism of the principle underlying the Lex Youtie, see
Fassino (1998).

(10.) See, for instance, P.Hamb.I 15, a contract from the early third century CE, where the
width of the column varies considerably from one line to another. This can probably be ex­
plained by the fact that the papyrus is twice as wide as it is high (43 × 21 cm).

(11.) On ἐ μβαδόν = ἐ μβαδικόν, see Kramer (1997, 325).

(12.) For the Ptolemaic period, where leap years were not taken into account and the cal­
endar therefore drifted by one day every four years, see Skeat (1954). The precise dating
of documents during the reign of Augustus is also notoriously difficult.

Paul Schubert

Paul Schubert is Professor of Greek Language and Literature, University of Geneva.

Page 19 of 19
Archives and Dossiers

Archives and Dossiers

Katelijn Vandorpe
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0010

Abstract and Keywords

Documentary papyri describe ancient people. Where unrelated texts are like instant snap­
shots, archives present a coherent film of a person, a family, or a community and may
span several months, years, or decades. Bilingual archives show how some Egyptians
tried to become Hellenized, but their private accounts betray their native language. An
archive is bound to be of greater interest than isolated texts, and the possibilities of
archival research for any aspect of life in Graeco-Roman Egypt are practically unlimited.
This article offers a systematic approach to archival documents and explains what consti­
tutes an archive, how archives come to light, how we can reconstruct them, the type of
archives that may be discerned, and the types of documents in them. Such an approach to
archival documentation of the ancient world has in general been attracting increasing in­
terest and brings together scholars who are studying different regions.

Keywords: documentary papyri, archival research, Graeco-Roman Egypt, archival documents, archival documenta­
tion, ancient world

DOCUMENTARY papyri lead us into the living rooms of ancient people. Where unrelated
texts are like instant snapshots, archives present a coherent film of a person, a family, or
a community and may span several months, years, or decades.1 We may enjoy people's
professional successes or feel sorry when they suffer misfortune. Contracts and wills may
show how rich elite members were, and their business papers may give insight into the
way their estates were run. Bilingual archives show how some Egyptians tried to become
Hellenized, but their private accounts betray their native language. Private letters may
testify to affectionate relationships or reveal marital problems.

Page 1 of 38
Archives and Dossiers

Figure 10.1. Petaus practicing the phrase “I, Petaus,

village scribe, have handed in” (Πεταὓς
κωμογραμματεὺς ἐπιδέδωκα); after four lines he for­
gets the first letter of the verb and writes “I, Petaus,
village scribe, have anded in” (πιδέδωκα). Cologne,
Papyrussammlung, inv. 328; P.Petaus 121.

An archive is bound to be of greater interest than isolated texts, and the possibilities of
archival research for any aspect of life in Graeco-Roman Egypt (e.g., economy, institu­
tions, history, gender, ethnicity) are practically unlimited; they can be illustrated here,
however, only by way of numerous examples. This chapter offers a systematic approach to
archival documents and explains what constitutes an archive, how archives come to light,
how we can reconstruct them, what type of archives we may discern, and what types of
documents we may find in them. Such an approach to archival documentation of the an­
cient world has in general been attracting increasing interest and brings together schol­
ars who are studying different regions (Brosius 2003; Pantalacci 2008). (p. 217)

Terminological Issues
The Term Archive in Papyrological Jargon

Papyrologists use the term archive in a slightly different sense from modern historians,
who think first of official records. The word archive is indeed derived from Greek archeion
(“government house for official records”). Papers of a more private nature gained in sig­
nificance in the second half of the twentieth century, and modern historians now draw on
the archives of prominent individuals, politicians, aristocratic families, and businesses.

The ancient Greeks and Romans had extensive and well-organized public records, but
with few exceptions these are now lost. Most papyrological archives are of a private na­
ture; they belonged to various classes of society; an ordinary person, as well as an elite
member of society, may have kept an archive. Consequently, (p. 218) papyrologists have al­
ways used “archive” to designate both public and private records and have anticipated
the evolution in archival research (Martin 1994). By what other term than “family
Page 2 of 38
Archives and Dossiers

archive” could one stress the unity of papers that an individual maintained for his family
and kept hidden in a closed jar in the cellar?

In addition, modern historians and papyrologists differ significantly in their approaches

toward archival research. Unlike modern historians, who must ignore irrelevant docu­
ments from their archives in order to preserve focus, the rarity of ancient documents
means that each and every text should be preserved and be welcomed by historians. Even
the discovery of a new fragment may be front-page news in papyrology.

Archive versus Library, Archive versus Dossier

The term archive is clearly reserved for documentary groups of texts, whereas a public or
personal library is a collection of literary texts (Martin 1994). Works of literature may,
however, be included among documentary papers (discussed later).

There has been debate among papyrologists about the exact definition of the terms
archive and dossier. Several scholars consider an archive “a deliberate collection of pa­
pers in antiquity by a single person, family, community (e.g., of priests) or around an of­
fice.” An arbitrary collection, such as papers found together by accident on a rubbish
dump from antiquity, is not included in this definition (Pestman, in Prim.2: 51, and 1995,
91–92; Martin 1994; for a summary and discussion of the different views see Van Beek
2007).

Where an archive is a collection made in antiquity, a dossier is “a group of texts brought

together today concerning a particular person or family, or a particular subject.” A dossier
may include a person's archive, if available, as well as other texts that do not belong to
the original archive but provide additional information on the archive holder (Pestman, in
Prim.2: 51, and 1995, 91–92; Martin 1994).

To papyrologists, Zenon is a celebrity. He was born in Kaunos (in Asia Minor) and settled
in Egypt, where he ran a large estate for the finance minister Apollonios under the sec­
ond Ptolemy. Almost two thousand papyri from his personal archive have been unearthed
at Philadelphia in the Fayyum, the town he helped develop. The majority of his documents
are related to his management of the estate. When he started working for Apollonios, one
of his main tasks was the irrigation of almost 2, 750 hectares of land. Encountering diffi­
culties, Zenon sent the chief engineer of the Fayyum the following letter: “Zenon to
Kleon, greetings. The water in the canal has not risen more than a cubit, and so the land
cannot be irrigated from it. Please, open the sluice-gates [at the entrance of the Fayyum]
so that the land can be irrigated” (P.Zen.Pestm. B; see Van Beek 2005). As this letter was
found among Kleon's papers, it is part of the dossier on Zenon, not part of his archive.

A dossier may include texts other than papyri and ostraca. The study of the “war
(p. 219)

of scepters, ” a conflict between Jews, Syrians, and Egyptians in the years 103–101 BCE,
presents literary and epigraphical texts alongside the reedition of a small archive of
Greek and demotic letters sent by soldiers on campaign to their home town. The study (P.

Page 3 of 38
Archives and Dossiers

War of Sceptres) has the appropriate subtitle “a multilingual dossier concerning a ‘war of
scepters’ ”.

Because not all papyrologists have accepted the distinction between “archive” and
“dossier, ” it is often ignored in practice. Furthermore, the difference is not always easy
to make because information on the circumstances that surrounded the finding of docu­
ments is often lacking. In a well-argued article, Andrea Jördens proposes another, three­
fold classification that departs from the German word “Nachlass” and focuses on the find
circumstances of the papyri (Jördens 2001; Van Beek 2007).

Intact and Mixed Archives

Archives in Their Original Depository

The ancient Egyptians kept their papers in a safe place; otherwise, their title deeds prov­
ing ownership might be stolen, as happened to a woman named Tapentos: “To Polemon,
epistates of Kerkeosiris, from Tapentos daughter of Horos, of the same village. An attack
was made upon my dwelling by Arsinoe and her son Phatres, who went of with the con­
tract relating to my house and other business documents” (P.Tebt. I 52; ca 114 BCE).

Archives were kept in private houses, public buildings, temples, or monasteries. Within a
building, several spots might be appropriate to deposit someone's papers: They could be
hidden below a staircase or even under a doorstep (see the photograph of a house in
Karanis, figure 2.3), but papers were, by preference, kept in the cellar.⋆Milon, who con­
trolled the finances of the Egyptian temple of Edfu (225–222 BCE), took a number of busi­
ness papers with him to the island of Elephantine, where they were found in a “round-bel­
lied jar, which stood in a narrow cellar; the jar was badly damaged when the surrounding
mud brick walls collapsed upon it and could not be preserved” (P.Eleph., p. 34).

Page 4 of 38
Archives and Dossiers

Figure 10.2a. The demotic archive of the funerary

priest Teos and his wife, Thabis, was bundled in
three packets of papyri wrapped in cloth. The second
(10.2b) and third (10.2c) photographs show the un­
wrapping of one of the packets in the Egyptian De­
partment of the Musées Royaux dʼArt et dʼHistoire,
Brussels (327/326–306 BCE, Thebes; Depauw 2000,
pl. 1–3).

Choachytai, priests responsible for mummies after their embalming, and funerary priests
in general instead put away their papers along with their tools in one of the tombs of the
necropolis they worked in (for the Theban necropolis see P.Choach.Survey, pp. 10–13).
Further archives have been hidden in caches such as the famous “cave of letters” in the
Judean desert, where refugees of the Bar Kokhba revolt of 132 CE took shelter (see chap­
ter 2 and figures 2.6–2.7). (p. 220)

Page 5 of 38
Archives and Dossiers

Figure 10.2b.

Papyri belonging to an archive could either be bound together into packets with strips or
be wrapped in cloth (see figure 10.2a–c). Packed or not, archive papers might be put into
a jar or box (see figure 10.3). The private archive of the ⋆Melitian monks of Labia (511–
513 CE), discovered in 1889 by Petrie in the church northwest of the Hawara pyramid,
was carefully put away: “Each was rolled up separately; the rolls were then bound round,
along with slips of reed, to prevent their being bent or broken; then tied up in a linen
cloth; next in a large lump of old tattered woolen embroidery; and the bundle placed in a
big jar sunk in the ground” (McGing 1990). (p. 221)

Archives Mixed Up in Antiquity

Papyri were, however, not always kept in their original depository, and even in antiquity
archives sometimes fell into disarray. They might be thrown in the wastepaper basket and
end up on rubbish dumps. The papyri in the dumps of Oxyrhynchus were often thrown
away in groups; thus, the origin of a papyrus from a particular section may allow it to be
linked with other texts (cf. chapter 2).

Page 6 of 38
Archives and Dossiers

Figure 10.2c.

Further discarded papyri or archives, especially from government offices, were bought as
secondhand paper and were reused for the fabrication of mummy cases or for wrapping
and stuffing mummified crocodiles (chapter 2). During the excavations in the desert
cemeteries of the site of Tebtynis, Ptolemaic papyri were unexpectedly discovered in sev­
eral crocodile mummies. Among these reused texts, groups of interrelated documents
have been reconstructed. Some smaller groupings (p. 222) are evident, such as the corre­
spondence of ⋆Adamas, keeper of the public granary, spread over at least three different
crocodiles. These original groupings are somewhat obscured in the edition of P.Tebt. be­
cause Grenfell and Hunt classified the documents according to text types. The best-
known group is the so-called archive of ⋆Menches, village scribe of nearby Kerkeosiris,
which comprises almost two hundred published documents to date. As Verhoogt (1998b,
2005) has shown, we are actually dealing with documents discarded from the office of the
village scribe and then used as secondhand paper by other people before ending up in the
mummified crocodiles. Menches's papers were thus reused twice, once as paper and once
as cartonnage, a circumstance that seriously hampers the reconstruction of the original
archive.

Page 7 of 38
Archives and Dossiers

Figure 10.3. The bilingual archive of Totoes, priest

of Hathor, was discovered in two large jars inside a
house (probably his own house) near the temple of
Hathor at Deir el-Medina (Thebes). The jars, of which
one is shown, are well preserved and have a diame­
ter of 23–24 cm and a height of 39 cm. The papyrus
rolls, in general some 32 cm long, were placed up­
right in the container. The jars were closed by insert­
ing a cup into the opening and were kept in place
with a rope knotted around the small handles (189–
100 BCE; Turin, Museo Egizio; for further examples
of archives found in jars, see Depauw 2000, 3, 4n4).

Figure 10.4. One book, two archives. A special case

of reuse involves the manufacture of a book from the
secondhand paper of two long administrative rolls
that contained official correspondence by Apollinar­
ios, strategos of the Panopolite nome. Both rolls were
cut up and pasted so that the blank versos made up a
small codex whose pages were not sewn together.
The new book belonged to a family (the ⋆descendants
of Alopex) who kept twenty-six tax receipts in it
(Dublin, Chester Beatty Library; P.Panop. 19).

Page 8 of 38
Archives and Dossiers

These examples show that several archives that papyrologists cherish actually came out
of a wastebasket. They help explain John D. Ray's reservations about (p. 223) (p. 224) call­
ing such groupings archives: “Archive, may the word be pardoned” (O.Hor, p. XIV).

Archives Mixed Up in Recent Times

Numerous archives were unfortunately not recovered during scientific excavations con­
ducted by trained scholars but unearthed by sebbâkhîn and other native diggers who of­
fered the texts for sale on the antiquities market (chapter 2). The antiquities dealers may
have acquired a number of archives in one batch, but they rarely sold them intact. The
bilingual archive of the Ptolemaic soldier ⋆Horos, son of Nechouthes, for example, was
purchased by Lord Adler, who noted: “I think the Hadj was trustworthy when he assured
me that my potful had not been tampered with” (P.Adl., p. 3); there are, however, reasons
to believe that either the jar had actually been tampered with or that there was a second
jar.

Figure 10.5. These four fragments of a single peti­

tion by Dryton, sold separately on the antiquities
market and acquired by three different collections,
had to be puzzled together (John Rylands Library inv.
67 + British Library inv. 686f + 687b + Heidelberg
inv. G.1301; P.Dryton 31). Rylands fragment repro­
duced courtesy of the university librarian and direc­
tor at John Rylands University Library, University of
Manchester. Heidelberg fragment reproduced cour­
tesy of the Institut fur Papyrologie, Heidelberg.

Most archives, however, were split up when they were brought onto the antiquities mar­
ket. Even individual texts that belonged to an archive were, deliberately or through igno­
rance, torn into pieces (figure 10.5). The famous archive of (p. 225) (p. 226) ⋆Zenon, agent
to the finance minister Apollonios, is scattered over a series of collections. The first texts
came on the market in 1911, and their price rose as Zenon became a celebrity. Almost
two thousand texts are known to date, dispersed all over the world. The Museum of Cairo
acquired the largest collection, published by C. C. Edgar in the five volumes of
P.Cair.Zen.; other substantial parts of the archive have turned up in Europe (London, Flo­
rence, Manchester) and the United States (New York, Michigan). To this dispersion of the
texts and fragments corresponds the fragmentation of publications; consequently, the
“Guide to the Zenon Archive” (two volumes) has become an indispensable instrument
(Pap.Lugd.Bat. XXI).

Certain archives have had an active life even after their initial publication. The demotic
archive of the herdsman and soldier ⋆Pabachtis was probably intact when Georg
Hauswaldt (hence the name P.Hauswaldt) bought it for the Konigliche Museen in Berlin.

Page 9 of 38
Archives and Dossiers

But during the Second World War one papyrus was destroyed in a bombardment, and oth­
er texts were seized by Soviet soldiers as war booty; fortunately, the majority have been
returned to Berlin. A few texts, though, have turned up in Magdeburg and Warsaw (Man­
ning in P.Hausw., p. 1).

An additional problem is that, for a long time, purchasers were interested mainly in Greek
papyri. The demotic papers of a bilingual archive thus often ended up in a different col­
lection from the Greek documents. A special case in this respect is the archive of ⋆Dionys­
ios, son of Kephalas, of Egyptian descent and a priest of the local ibis cult. His family be­
came Hellenized, and Dionysios was recruited as an infantryman of the Ptolemaic garri­
son at Tenis-Akoris. Reinach acquired this bilingual archive in Egypt in the winter of
1901–1902, but after the initial edition it was split up: The Greek portion went to Paris,
and the Demotic batch to Munich (Pestman in Pap.Lugd.Bat. XXII, pp. V–VI).

Reconstruction of Archives
When archives were intermingled with other texts in ancient rubbish dumps, mummy cas­
es, or mummiied crocodiles, or were split up by clandestine diggers or dealers, scholars
may use several methods in reconstructing them (compare Martin 1994).

Relying on the Archaeological Context: A Solid Method

Information on the archaeological context, whenever available, is crucial. Van Minnen

(1994) has demonstrated how productive a “house-to-house search” at Karanis might be,
starting with the old (but for their time exemplary) excavation reports and (p. 227) taking
into account all found objects, including papyri. He was thus able to identify the house
and archive of the tax collector ⋆Socrates (see chapter 2). Verhoogt (1998a) reconstructed
a small Tebtynis archive starting from the editor's note in the introduction of P.Tebt. II
326: “This papyrus was found tied up in a bundle with 285, 319, 335, 378, 408 and 588.”
The papers appear to have belonged to ⋆Sarapias, member of a family of Antinoite
landowners in the Fayyum, who added some documents that belonged to her male rela­
tive Sarapammon (165–264/270 CE). Where information on the archaeological context is
lacking, however, other methods are available.

Studying Prosopographical Data: A Successful Approach

Proposopographical data—information about individuals—and insight into the family tree

are extremely helpful in sorting out which texts belong to which archive. As Bagnall right­
ly puts it, “Reconstructing the archive is itself an iterative process, in which clues allow
the historian to delineate enough of an individual's characteristics to permit attribution of
further documents to the archive” (Bagnall 1995, 40). It is ideal to dispose of a will that
mentions grandparents and children or a census declaration filed by the householder, in
which he lists himself and his family members, but this kind of information is scarce.

Page 10 of 38
Archives and Dossiers

The archive of the descendants of ⋆Alopex, kept at Dublin and Cologne, has been recon­
structed mainly on prosopographical grounds. Because the name Alopex is rare and a few
other names appear in both collections, Hagedorn (1970) concluded that these papers
constitute an archive that belonged to a family descended from Alopex. Several texts re­
fer to the family's olive yards, oil business, and taxes on oil production (308–355 CE).

In March 1934 Italian archaeologists made a spectacular discovery at the ancient site of
Tebtynis in the Fayyum: A cellar of one of the houses harbored several hundred Roman
papyri, thus earning the name cantina dei papiri (figure 10.6). Unfortunately, the find­
spots of the individual papyri were not described in detail; as a result, the archives of the
cantina have had to be reconstructed solely by means of prosopography. The family
archives known thus far belonged to (1) ⋆Kronion, (2) the family of the ⋆Patronids (former­
ly called the archive of Laches), and (3) the descendants of ⋆Pakebkis. As Smolders (2005)
proposes, Pakebkis's grandson Turbo can be identified as a manager or phrontistes of the
Patronids, one of the wealthiest and most influential families of the nome at that time. If
this identification is accepted, the second and third archives were probably in the hands
of a single person before being dumped in the cantina.

Figure 10.6. The “cantina dei papiri, ” shown here

after it was emptied, is considered a storage area for
discarded material (perhaps destined to be used as
fuel) rather than an archival repository (Gallazzi
1990, 284; see also Davoli 1998, 186n313). Its
archives are reconstructed on prosopographical
grounds. Photo by Gallazzi (1990, pl. 18).

This example shows the importance of detailed prosopographical research: A scholar not
only has to definitively establish someone's identity but analyze the person's professional
activities as well. Even the identification of the handwriting may be useful. (p. 228)

Identifying Real Estate: A Method Rarely Applied

Because they prove ownership, title deeds occupy a prominent place in family and per­
sonal archives of a private nature. When one person sold property to another, he turned
over the older title deeds for the property to the new owner. In these cases, it is impor­
tant to identify real estate through the remaining title deeds. Once the identification is

Page 11 of 38
Archives and Dossiers

completed and the last owner of the property is known, it is clear which archive the title
deeds belong to. Pestman (1965) has developed this method for the bilingual archive of
⋆
Peteharsemtheus, son of Panobchounis, where title deeds account for one third of the
documents.

The owner might bundle the papyri relating to his property. Around 83 CE ⋆Apynchis, son
of Tesenouphis, for instance, collected all of the papers concerning his house with a
courtyard in Soknopaiou Nesos in the Fayyum: the most recent and the older title deeds,
a copy of a bank receipt, and a declaration of property to the bibliophylakes of the me­
tropolis. All these documents were probably pasted together into a tomos synkollesimos
(see later description).

Museum Archaeology: A New Method

A more recent method is known as “museum archaeology”: by tracing the routes of pa­
pyri and the dates of their acquisition by museums and other collections, one may draw
several conclusions. For instance, if one and the same archive has been split up or if one
and the same papyrus has been torn to pieces and if those papyri (p. 229) or fragments
have turned up in different collections, there must be a link between the acquisitions of
the collections in question. For a number of family archives from the Upper Egyptian
town of Pathyris (modern Gebelein), this method has proven successful. The extension
through museum archaeology, of the archive of the Greek cavalryman ⋆Dryton and his
wife, which had been reconstructed on purely prosopographical grounds, is significant
and shows who inherited the papers on Dryton's death (Vandorpe 1994 and P.Dryton).
Clackson (2004) successfully applied this method to Coptic documents as well.

A scholar will, however, never be certain of having found all of the pieces of the puzzle.
Especially anonymous papers, like school exercises, accounts, or lists, are difficult to at­
tribute to a specific archive. Papyri or fragments may lie concealed somewhere in an en­
velope or in the closet of a collection, waiting to be discovered a second time.

Editions, Guides, and Lists of Archives

Editions

An archive uncovered intact during a legal excavation and preserved as such in a muse­
um is usually published in its entirety (e.g., P.Adl., P.Hausw. by Spiegelberg; P.Tor.Botti).
The documents of an archive that have been dispersed among several collections are, by
contrast, often published in separate editions. The early Ptolemaic demotic archive of a
woman named ⋆Taienteus has been reconstructed from papyri scattered among several
collections: The Manchester portion was published by Francis Griffith as P.Ryl.Dem. 10–
14, and the London batch, by Stephen Glanville in 1939 (P.Brit.Mus. I), whereas other
documents from Moscow still await publication (Pestman 1989, 14–24).

Page 12 of 38
Archives and Dossiers

As Pestman (1989, 10) observes in a Dutch publication on family archives, an archive

should be brought together and published as an entity before a thorough and reliable
study can be written. The Roman archive of ⋆Philosarapis was the first dispersed archive
published as a whole (by van Groningen in 1950, Pap.Lugd.Bat. VI). Next was the Roman
archive of ⋆(Eutychides, son of) Sarapion (by Schwartz in 1961, P.Sarap.) and the archive
of ⋆Flavius Abinnaeus, of the early Byzantine period (by Bell and others in 1962, P.Abinn.).
Those three archives are composed of Greek documents only.

Publications of bilingual (particularly Graeco-demotic) archives are not as numerous as

single language collections. The Greek and demotic papers of ⋆Horos, son of Nechouthes,
were published together in 1939 but with editorial (p. 230) work divided by language com­
petence. Pestman set an example by publishing the Greek and demotic documents of
“Lʼarchivio di Amenothes figlio di Horos” in 1981 (P.Tor.Amen.) and of “Les archives
privées de Dionysios fils de Kephalas” in 1982 (Pap.Lugd.Bat. XXII, together with
Boswinkel).

Guides

These publications have undoubtedly shown the value of a (re)edition of dispersed pa­
pyrus archives. When, however, an archive is composed of a large number of texts, an edi­
tion would constitute a lifetime project. It is then often more convenient to compose a
guide to or survey of the archive, as was done for the ⋆Zenon archive (Pap.Lugd.Bat. XXI)
and that of the Theban ⋆Choachytai Osoroeris and Panas (P.Choach.Survey). Such guides
or surveys can be accompanied by a reedition of a part of the archive. The guide to the
Zenon archive was preceded by a volume of the same series that incorporated, among
other documents, editions of the bilingual papyri of Zenon's archive (Pap.Lugd.Bat. XX).
The survey of the choachytes' archive appeared one year after Pestman's “Il processo di
Hermias e altri documenti dellʼarchivio dei choachiti, ” a reedition of the archive's Greek
and demotic texts preserved in Italian collections (P.Tor.Choach.). Many texts of these two
archives still await publication.

Descriptions, Lists, and Studies of Archives

Several lists of archives, some accompanied by a brief description, are available. A long
chronological list of 135 Greek archives, though certainly not pretending to be exhaus­
tive, has been compiled by Montevecchi (1988, 248–261, 575–578). A similar initiative
was undertaken for demotic and bilingual (Graeco-demotic and Graeco-Coptic) archives
in 1986 by Lüddeckens (Lexikon der Ägyptologie, vol. 7, 876–886, s.v. Urkundenarchive,
with sections for the Ptolemaic, Roman, and Byzantine periods) and in 1997 by Depauw,
who included early demotic archives from the pre-Ptolemaic period (Depauw 1997, 153–
162). Erwin Seidl presented a selection and a substantial description of Greek, demotic,
and bilingual archives from Graeco-Roman Egypt (Seidl 1962, 15–49, for Hellenistic
Egypt; Seidl 1973, 55–71, for Roman Egypt).

Page 13 of 38
Archives and Dossiers

The Leuven project, “Leuven Homepage of Papyrus Collections, ” which began in 2002, is
compiling an exhaustive list of all Greek, demotic, Coptic, and bilingual archives from
Graeco-Roman and Byzantine Egypt. The results are presented in a database-driven web­
site that offers several search options (directed by W. Clarysse and K. Vandorpe, see
http://www.trismegistos.org/arch/index.php) and provides a systematic description of
each archive (see also Van Beek 2007).

A list of archival studies in journals or in chapters of books would be long indeed, regard­
less of whether they concentrated on small archives or on particular (p. 231) aspects of
large ones. Lewis (1986) has bundled several short portraits of individual archive keepers
into his book Greeks in Ptolemaic Egypt: Case Studies in the Social History of the Hel­
lenistic World. Among the book-length synthetic studies, Orrieux's two monographs on
Zenon (1983, 1985) and Rathbone's Economic Rationalism and Rural Society in Third-
century AD Egypt (1991) take a prominent place. In the latter study Rathbone deals with
the extensive archive of ⋆Heroninus, manager of a unit of the private estate of Appianus in
the Fayyum. In it he “brings to the attention of historians of the ancient world a unique
and so far neglected treasury of social and economic information whose significance, es­
pecially as regards the social structure of the estate and its system of management and
accounting, is not confined to the history of Roman Egypt” (Rathbone 1991, 3).

Types of Archives and Archive Keepers

The jurist Erwin Seidl (1962) was the first to propose a typology of archives, distinguish­
ing between official and private archives and, among the latter, between family and book­
keeping archives. The classification proposed later in the chapter builds on his typology.
However, although official archives often appear to contain private correspondence and
vice versa, it is not advisable to add an extra category called “hybrid archives” (Martin
1994). The introduction of the term “personal archive” may solve this problem, as Bous­
sac (1993) suggests for Delian archives.

Official Archives and Personal Archives of Officials

Official archives (called “Amtsarchive” by Seidl) contain only or mainly papers belonging
to a governmental agency or a state official. Few such archives have actually been found
in a public building. In the winter of 1892–1893 Naville discovered several carbonized pa­
pyrus rolls in the ruins of the Roman bibliothêkê, or record office, in the ancient capital of
the Mendesian nome, ⋆Thmouis (P.Thmouis; compare chapter 2 for this and further exam­
ples).

The major part of the official archives, however, ended up on rubbish dumps or were used
as secondhand paper in mummy cases and mummified crocodiles, such as the archive of
⋆
Menches (mentioned earlier), a village scribe. Other official papers were taken home by
officials upon their retirement, for instance, and were found in their houses, as discussed
later.

Page 14 of 38
Archives and Dossiers

An official archive may center around a government office, such as a granary (e.g., at ⋆Oxyrhyn­
cha and ⋆Pyrrheia), notarial office (e.g., at ⋆Krokodilopolis and ⋆Pathyris), or record office
(grapheion, e.g., at ⋆Theogonis). Other official archives are instead linked to a particular
official and are generally named after this person (p. 232) and his title (e.g., the nomarch­
es ⋆Aristarchos, the toparches ⋆Tesenouphis, or the village scribe ⋆Isidoros). The archive of
⋆
Aurelius Heras, praepositus pagi, comprises nine documents addressed to Aurelius
Heras (alias Dionysios) in his capacity as head of the administration (praepositus) of the
eighth pagus (district) of the Oxyrhynchite nome.

The term official archive may in these cases be misleading, as these collections often in­
clude papers of a private nature. This should not surprise us, as our mailbox at work un­
doubtedly contains both business and private messages, and the ancients lacked our mod­
ern notions of separation between personal and professional. Such archives are thus best
classified as “personal archives of officials.” ⋆Kleon, chief engineer in the Fayyum under
the second Ptolemy, belonged to the Greek elite. His home city was Alexandria, where his
wife and sons lived and where they had friends in high places. But for his job, Kleon was
required to spend most of the year in the Fayyum, where he kept his official papers mixed
up with his private correspondence, which composes about 15 percent of the archive (Van
Beek 2005). His wife suggested coming to visit him, and his children begged him to come
home for an annual festival, apparently without success. In a supporting letter, his son
Philonides hints at Kleon's fall from grace with the king:

Nothing truly will be dearer to me than to protect you for the rest of your life in a
manner worthy of you and of myself, and if the fate of mankind befalls you, to see
that you enjoy all due honours; this will be my chief desire, honourably to protect
you both while you live and when you have departed to the gods. If possible, then,
make every effort to obtain your release for good, or if you see no chance of that,
for at least the time of low Nile, at which season there is no danger and Theodoros
can be left to take your place, in order that you may spend this season at least
with us. (translation Sel.Pap. I 94)

Theodoros indeed succeeded Kleon as chief engineer and took over the latter's archive, includ­
ing, surprisingly, the private letters. The archive is therefore known as the archive of ⋆Kleon and
Theodoros.
After their time in office, some officials took home documents that were important to
them personally and intermingled them with their private correspondence and even their
private library. ⋆Apollonios, strategos of the Apollonopolite Heptakomias nome in Upper
Egypt, maintained a regular correspondence with people from his hometown, Hermopo­
lis, where his wife, Aline, his daughter Heraidous, and his mother, Eudaimonis, lived.
When his archive was discovered in the Hermopolite, it became apparent that Apollonios
had carried home several official and private papers after his term in office. A substantial
number (almost 100 out of the 225 papyri) are deemed to belong to the private part of the
archive.

Page 15 of 38
Archives and Dossiers

Apollonios was in office at the time Hadrian succeeded Trajan as emperor (117 CE), when
a revolt of the Jews was raging in several countries. Surprisingly, he even took part in a
battle against the Jews at Memphis. No wonder that his dedicated (p. 233) wife was wor­
ried about him, expressing her feelings in one of her letters: “I take no pleasure in food
and drink, but stay awake continually night and day with one worry, your
safety” (C.Pap.Jud. II 436). Apollonios's mother, Eudaimonis, possessed a sharp tongue
and advised her son to “do the same as the strategos here, who puts the burden on the
notables” (C.Pap.Jud. II 437). In another letter she alludes to the risks Apollonios was tak­
ing: “Be sure that I shall pay no attention to the god [Hermes] until I get my son back
safe.” Although she also complains about receiving too little financial support from her
family, she undoubtedly exaggerates: “I already have the vision of being naked when win­
ter starts” (C.Pap.Jud. II 442). Apollonios may have been flattered by the letter from his
female servant, Taus, who appears to have been very devoted to the strategos: “I beg you,
my lord, if it please you, to send for me. Else, I die because I do not behold you daily.
Would that I were able to fly and come to you and make obeisance to you…” (Sel.Pap. I
115). Unsurprisingly, Apollonios cherished these letters.

Personal and Family Archives of a Private Nature

Personal versus Family Archives

A personal archive is kept by a single person, whereas a family archive (Familien-archive
in Seidl's terminology) must fulfill one of the following two conditions:

(a) The archive was put together by a family member who also kept papers for his
wife or other close relatives, mostly parents, sisters, sisters-in-law, and daughters.
(b) The archive was inherited by the oldest son or daughter and was thus continued
by the following generations.

In several cases both conditions are fulfilled. The family archive of ⋆Philosarapis, for instance,
had four successive archive holders, following the line of the first-born sons of the ancestor
Maron to the last central figure, Philosarapis II. They kept documents for other relatives as well.
Our knowledge of this family, one of the wealthiest in the Roman Fayyum, spans no less than 135
years.
When a family archive is passed from an older to a younger generation, it is best named
after the last owner, as Seidl (1962, 16) strongly recommended, and this practice has
sometimes been rigorously followed even when the last owner is unknown (compare the
“archive of the Anonym, ” Pestman in Pap.Lugd.Bat. XXVII, p. 93). Nonetheless, for a
number of reasons, especially when the last owner is unknown, the name of the central
figure or of an ancestor may be used: Of the more than eighty texts of the Tebtynis
archive of ⋆Patron's descendants, Patron himself did not significantly contribute to it,
whereas his three sons and some of his grandchildren did, but it is not clear who among
the Patronids was the actual archive keeper. In view of the fact that several members of
this wealthy family from (p. 234) the Fayyum exercised the highest civic magistracies of
Ptolemais Euergetis, they may have lived in the city and maintained a second residence in
the country, at Tebtynis, where they were rather absentee landlords assisted by a phron­

Page 16 of 38
Archives and Dossiers

tistes, or manager (Bagnall 2000). One of these managers was named Laches (Clarysse
and Gallazzi 1993); another, Turbo. This example brings us to the problem of the archives
of wealthy families, who often appointed a manager or steward to maintain important
records: The archive may thus have been kept by either the family or the managers.

Private versus Professional Papers

Archives of an individual or a family may comprise private papers, documents from the
professional sphere, or both. When the latter are dominant, the name of the archive usu­
ally refers to the profession of the archive keeper, such as the goldsmith ⋆Menches (about
195 BCE, Oxyrhyncha). Typical professional archives are the so-called business or book­
keeping archives (“Buchführungsarchiv”; Seidl 1962), which contain primarily correspon­
dence and accounts.

The ⋆Zenon archive is a standard example of a business archive kept by an individual.

Apart from numerous accounts and letters written to him, this famous manager also col­
lected letters addressed to his employees, agents, or partners, as well as letters that had
been forwarded to him for his information or for further investigation. Amid this serious
and sometimes boring correspondence, the following scribbled note catches the eye:
“Horos to Zenon, greetings. By February 3 [252 BCE] there will be 130 arouras [ca. 36
hectares] sown with poppy. If you please, do come and see me so you can feast your eyes
on the sight. Farewell” (P. Cair.Zen. II 59243; translationby Crawford 1973). Horos wrote
this letter with an Egyptian brush and in poor Greek, but his enthusiasm for a large Weld
full of white or light purple poppy flowers prevails. Poppy was a new oil crop that was be­
ing experimented with on the estate.

Zenon's papers constitute the largest papyrus archive known to date. Such an extensive
assemblage may be subdivided into several subarchives or files, some of which led a life
of their own for a while before ending up in Zenon's hands:

October/November 261–April 258 BCE. Palestinian file. Zenon started out as the commer­
cial representative of the minister in Palestine and learned a great deal about the pros­
pering slave trade. Meanwhile, the king granted Apollonios a large estate of 2, 750
hectares located at the desert border of the Fayyum area.

April 258–April/May 256 BCE. Alexandrian file. At the threat of the second Syrian war,
Zenon returned to Egypt and was promoted to the minister's private secretary. In this ca­
pacity he took over Apollonios's correspondence at Alexandria and accompanied his em­
ployer wherever necessary. Because Zenon endorsed the minister's letters, adding the
name of the town where he opened the letters, scholars have been able to reconstruct the
exact route that Apollonios's staff followed through the delta, where they inspected the
collection of the taxes.

April/May 256–248/247 BCE. After a serious illness, Zenon had to lead a more
(p. 235)

sedentary life: He became the private manager of the large estate belonging to the minis­
ter in the Fayyum and supervised the urbanization of the neighboring town of Philadel­
phia. Zenon took over the archive of his predecessor, Panakestor, who had failed to devel­
Page 17 of 38
Archives and Dossiers

op the new estate in a timely fashion (Panakestor's file). During the first three years,
Zenon was very much occupied with his managerial duties, and the minister kept close
track of the developments, as shown by his abundant correspondence (Zenon's file of the
estate). But the minister's interest gradually diminished, his letters became scarce, and
Zenon was concentrating more and more on his own affairs (Zenon's personal file).

248/247–end 229 BCE. Between November 248 and early December 247 BCE Zenon was
dismissed for an unknown reason. He was succeeded first by Eukles and later by Bion,
both of whom were assisted by a secretary named Apollonios. The latter kept the estate's
papers at that time (file of Apollonios, the secretary) but handed them over to Zenon
when the estate was reclaimed by the new king, Ptolemy III. Meanwhile, Zenon lived in
Philadelphia as a respected citizen and conducted his own business in a successful man­
ner (Zenon's personal file, continued). After March 240 BCE he probably handed over his
entire archive (which comprised all of the aforementioned files) to an anonymous person
—possibly Apollonios, the estate's former secretary, and possibly one of Zenon's brothers.
“Anonymous” continued the archive until the close of 229 BCE. By that time Zenon was no
longer alive.

The largest business archive of Roman Egypt is that of ⋆Heroninus, which is scattered
among several collections and made up chiefly of letters (sometimes written on the backs
of old literary texts) and accounts (253–306 CE). About 450 texts have thus far been pub­
lished, but even more texts still await publication. The available evidence has been stud­
ied by Rathbone, who focuses on the structure of the estate, its personnel, and its man­
agement. Rathbone shows how the so-called archive of Heroninus is a “useful shorthand
designation” in referring to a large, poorly defined group of texts. The core is formed by
the papers of Heroninus himself, who was manager of a unit of the well-known estate of
Appianus and of five further estates in the Fayyum; these estate papers form subarchives
or files. A number of related texts, clearly of different origin but closely linked to
Heroninus's papers, should not be considered as belonging to the archive stricto sensu
(Rathbone 1991, 6–8, 410).

These two archives belonged to managers of large estates, but small businessmen, too,
like the oil seller ⋆Phanesis (233–223 BCE, Tebtynis), produced numerous papers. When
⋆
Dionysios, son of Kephalas, was a young man waiting to be recruited as an infantryman
of the Ptolemaic garrison at Tenis-Akoris, he went into business in order to make a living,
and the forty texts of his archive give us much information about these activities. After
Dionysios joined the Ptolemaic army, the archive abruptly stops.

Some priestly functions similarly left their mark on private archives. Apart from their pri­
vate papers, the choachytai, or libation pourers, collected professional documents such as
records of the names of the mummies they looked after, payments of funerary taxes,
transfer taxes on tombs, and disputes over professional (p. 236) affairs (e.g., the archive of
⋆
Teos and Thabis). Some also kept administrative documents of the association of
choachytai, including a list of regulations (these informed members of the days of drink­

Page 18 of 38
Archives and Dossiers

ing and the expectation that they would be present at each other's embalming and funer­
al [P.Choach.Survey 61]).

Some professions, however, did not involve a mass of paperwork. Soldiers, at least those
who were not in charge of a military camp, usually had a humble archive of private pa­
pers only. Conspicuously, they produced more documents before they were recruited (like
the aforementioned ⋆Dionysios, son of Kephalas) or after they retired. The Graeco-Latin
archive of ⋆Pompeius Niger contains only two texts from the twenty-four years in which he
served as a soldier of the Legio XXII Deiotariana, but his next twenty years as a veteran
are far better represented with thirteen documents (31–64 CE, Oxyrhyncha). The same is
true of the archive of ⋆Iulius Serenus, who retired with the rank of decurio (179–216 CE,
Karanis).

Archives of Communities: Temples, Monasteries, Associations

Although the Ptolemaic administration took over several of their tasks, the Egyptian tem­
ples maintained an intensive administrative apparatus in Graeco-Roman times. Their
archives comprise accounts, lists, letters, yearly reports to the government, oracle ques­
tions, or self-dedications to the god. All temple documents were as a rule written in de­
motic until well into the first century CE, when demotic disappeared for official and legal
texts (see later discussion). The temple archive of ⋆Soknobraisis at Bakchias in the
Fayyum (116–212 CE), for instance, is exclusively Greek and includes yearly reports (ac­
companied by cover letters) to the government listing the sacred objects and declaring
the number of priests. Sebbakhin put this archive up for sale on the antiquities market in
the 1930s. Only at the end of the twentieth century was the actual temple of Soknobraisis
identified; it faces the well-known sanctuary of his twin brother, Soknokonneus (Capasso
1996, 119; Davoli 2005, 29).

In addition to their archives, priests often collected literary texts, which were rarely writ­
ten in Greek, as these temple libraries were “the last bastions of the native languages and
scripts” (Depauw 1997, 160). The most illustrious library in Roman times was that in Teb­
tynis, which gathered for the last time the entire Egyptian literature of the past millenni­
um: novellas such as the cycle of Pharaoh Inaros or of the sage and magician Setne;
mythological, astronomical, and astrological treatises; and a collection of law texts and a
herbal. One of the fragmentary papyri, which must have been more than ten meters long,
was an encyclopedia of stars, rivers, villages, and gods written in hieratic with glosses in
demotic and a few annotations in Greek characters (Ryholt 2005).

Several monasteries had similarly furnished (Graeco-Coptic) archives or libraries, such as

that of Epiphanius (P.Mon.Epiph.) and the monastery of Phoibammon at (p. 237) Deir el-
Bahari. The latter produced a large number of ostraca and papyri and a few texts on
parchment. Receipts of book loans and fragmentary inventories testify to the existence of
a library; the monastery's archives consisted of documents related to its economic and
administrative activities, short religious texts, the monks' “school” texts, and letters writ­
ten by or to its bishop, Abraham (Godlewski 1986).

Page 19 of 38
Archives and Dossiers

Archives of professional or cult associations are harder to detect. Their papers may have
been kept by a member of the association among his personal papers or, in the case of
cult guilds, by the temple.

Archives Comprising a Single Type of Document

Some archives contain a wide variety of documents, whereas others are rather monoto­
nous, sometimes consisting of a single type of document and therefore named after it.
The military correspondence of ⋆Pates and Pachrates is a collection of Greek and demotic
letters addressed to these officers and written by soldiers on campaign in the north of the
country during the Judean-Syrian-Egyptian conflict of 103–101 BCE. The private corre­
spondence of ⋆Apollonios from Bakchias consists of fourteen letters written by and ad­
dressed to this estate manager active under the Flavian emperors. The tax lists kept by
the ⋆praktores argyrikon of Soknopaiou Nesos are labeled “an early-third-century tax
archive” by Lewis (1954).

Composition of Archives
The composition of archives is here approached from two different angles.

Incoming versus Outgoing Documents

In this first approach I consider the archive holder as the central figure and evaluate his
relationship to the documents (cf. Van Beek 2007). Some documents (mostly letters) are
addressed to him; others (such as contracts or receipts) are drawn up for his benefit: All
of these texts are part of the group of incoming documents, and one expects to find them
in the archive of the key figure. Other texts, however, appear to be written by or in the
name of the archive holder and are destined for other people: this group forms the outgo­
ing documents, and their presence in the archive needs to be explained. Usually drafts or
copies of these outgoing documents are involved. A special case is that of completed peti­
tions: The original petition maybe returned to the archive holder after a decision has
been made and may contain a note by the official who dealt with the case. A standard ex­
ample of an archive with both incoming and outgoing documents is that (p. 238) of ⋆Apolli­
narios, strategos of the Panopolite nome, which consists of two administrative rolls: The
first roll is a register containing copies of letters dispatched by the strategos in Septem­
ber 298 (outgoing); the second roll has some fifty letters written to the strategos by a su­
perior in February 300 (incoming) (see figure 10.4).

The categories of incoming and outgoing documents do not cover all texts, however. Ac­
counts, lists, and the like drafted by the archive holder for internal use and never meant
to leave his dwelling, may fall under the heading of “internal documents.” The terminolo­
gy proposed here has been created by Clarysse for the Leuven project on archives.

Page 20 of 38
Archives and Dossiers

Juridical versus Sentimental Value

This second approach tries to explain why the archive holder kept or cherished certain
documents and not others. In case of official, business, and temple archives, the explana­
tion is simply that the texts were of importance to the archive holder's office, business, or
temple activities.

With regard to family and some personal archives, the answer is more complex and has
been discussed by Pestman in a fine Dutch booklet on such archives. In general, family
archives contain a rich variety of documents; hence, Pestman's enthusiasm: “Family
archives are goldmines.” He distinguishes between three groups of texts (Pestman 1989,
7–8), and to them I add a fourth group here:

(a) The first and most substantial group consists of legal documents that could be
submitted in case of a trial, such as title deeds, loans, marriage contracts, wills, min­
utes of lawsuits, and tax receipts. Contracts concerning the purchase of real estate
and even lease contracts might be presented in court to prove ownership. Loan con­
tracts and acknowledgements of debt were put in the hands of the creditor, but when
the loan was paid off, the creditor either returned them to the debtor or issued a re­
ceipt. Marriage and divorce contracts were safest in the archive of the wife's family.
Documents that were actually collected in antiquity in order to present them in court
proceedings constitute a lawsuit archive or file. The so-called ⋆Erbstreit archive deals
exclusively with an inheritance dispute among three different parties within a single
family, resulting in no fewer than five trials in a two-year period. Apart from the orig­
inal demotic title deeds and their Greek translations, it includes copies of former law­
suits and even the original documents through which the final verdict of the judges
(the chrematistai) is executed (186–133 BCE, Pathyris).

Page 21 of 38
Archives and Dossiers

Figure 10.7. When returning this loan agreement to

the debtor, the creditor canceled it by means of
crosshatched lines (London, British Library inv.
1893; P.Fam.Tebt. 22; archive of Philosarapis).

A number of family and personal archives contain a substantial lawsuit file. Ten out
of the thirteen documents of the lector priest ⋆Tefhapi from Siut are related to a trial
concerning the inheritance of his father, who was married twice (186–169 BCE). Six­
teen out of the twenty-two texts in the family archive of Gaius ⋆Iulius Agrippinus con­
cern the so-called Drusilla lawsuit. The widow Drusilla took legal action against
(p. 239) the Roman veteran Agrippianus and, on his death, against his son
(p. 240)

Agrippinus. When the archive ends (103/117–148 CE, Karanis), the case remained
unsettled.
(b) Of the second group of odd notes without any juridical value and kept for senti­
mental reasons, letters are a good example. Although correspondence was essential
and often predominant in official, business, and temple archives, letters found in
family and personal archives were meant to give a sign of life and were kept mainly
for sentimental reasons. This second category includes texts other than letters, such
as the demotic list of the birthdays of his five children, which ⋆Amenothes, son of
Horos, carefully drafted, starting with the oldest son born, on October 31, 150 BCE,
and ending with the youngest scion, born on August 28, 129 (P.Tor.Amen. 3), or the
school exercises written by Apollonios and kept by his older brother, Ptolemaios
(UPZ I 147).
(c) Literary works were rarely kept by private persons, as the archive keeper had to
have some interest in literature. Books might constitute the personal library of pri­
vate persons. The recluses ⋆Ptolemaios and his brother Apollonios, who devoted their
lives to the god in the Memphite Serapeum (164–152 BCE), not only copied a fair
number of extracts from literary texts and model letters but also borrowed terms and
phrases found in them for their petitions and dream descriptions (Clarysse 1983, 57–
60). When, however, only few literary works have been found, one can assume they
Page 22 of 38
Archives and Dossiers

were kept together with the remaining documentary papyri of the family archive
(Clarysse 1983). The cavalry officer ⋆Dryton, for instance, copied a play on the back
of one of his old loans, known as the “Alexandrian erotic fragment, ” which tells the
story of a woman who was abandoned by her lover (Bing in P.Dryton 50).
(d) However, people may have kept documents for various other reasons (perhaps
practical in nature) that we do not yet understand. They may have kept outdated con­
tracts that were no longer necessary for court cases but were useful as secondhand
paper (e.g., P.Dryton, pp. 279–280; Clarysse 1983, 58–59 and n. 86).

Greek, Egyptian, and Bilingual Archives

Spanning almost a thousand years, figure 10.8 shows the predominance of Greek archives
and the lesser place occupied by Egyptian (that is, demotic and Coptic) archives. This
should not surprise us: Greek was the language of the administration and of the upper
class. In particular, the areas where many Greeks had settled (i.e., Lower Egypt, includ­
ing the Fayyum area) furnished numerous Greek and Graeco-demotic archives even in the
early Ptolemaic period.

(p. 241) Purely demotic archives are limited to the early Ptolemaic period and to the bare­
ly Hellenized southern part of the country, mainly Thebes. One of the oldest such archives
is that of the Theban funerary priest ⋆Teos and his wife, Thabis, which dates to the late
fourth century BCE. Further demotic archives (dating through the end of the third centu­
ry BCE) were found in Upper Egypt. These include the family archive of the herdsman
and soldier ⋆Pabachtis, son of Paleuis, who was living in Edfu and serving on the southern
border until the great Upper Egyptian revolt in 206 BCE. Some title deeds in this archive
have a Greek registration docket, which is not enough to classify them as bilingual
archives. As several Upper Egyptians became Hellenized to a certain degree after the
suppression of the revolt in 186 BCE, the purely demotic archives of the south gradually
give way to bilingual archives. The government was supportive of this development and
introduced among other things Greek notaries. In particular, Pathyris, some thirty kilome­
ters south of Thebes, yielded numerous bilingual family archives of Egyptian soldiers who
were serving in the Ptolemaic army and became partly Hellenized (Pestman 1965, Van­
dorpe 2008).

In the early Roman period, new registration requirements were introduced, and the first
party of a contract was obliged to add an elaborate subscription in Greek. Consequently,
demotic notary contracts disappeared by the late first century CE, though demotic was
still used for other types of texts until the third century. The Upper Egyptian sites, for­
merly important suppliers of demotic and bilingual archives, no longer produced such
records. Fayyum towns like Tebtynis and Soknopaiou Nesos, by contrast, still yield bilin­
gual archives that date to the late first century (Depauw 2003).

Page 23 of 38
Archives and Dossiers

Graeco-Latin archives constitute a minority, and even in these, Latin texts make up only a
small percentage (see chapter 6; for an exception, see the Graeco-Latin archive of ⋆Claudius
Tiberianus). After his retirement, ⋆Flavius Abinnaeus, a cavalry officer who commanded
the military camp of Dionysias in the Fayyum, took part of his official archive home, prob­
ably to Philadelphia, and added it to his private documents and those of his wife, Nonna.
Of the more than eighty texts, only two were written in Latin; these were kept for obvious
reasons: a petition to the emperors Constantius and Constans in which Abinnaeus claims
the command of the cavalry ala at Dionysias, and a sharp letter from the dux of Egypt dis­
missing the commander two years later (341–351 CE; P.Abinn.).

In the Byzantine period, the Greek language was still dominant in official and economic
matters, but Coptic became popular for private documents. Several hundred texts consti­
tute the well-known bilingual archive of a family from sixth-century Aphrodito, named af­
ter the landlord and notary Flavius ⋆Dioscoros. Having been trained in both languages,
Dioscoros developed two types of handwriting: uncial writing, which he used for his
Greek literary and Coptic texts, and a cursive writing, found in his Greek documents.
Most of the Coptic texts, which are in the minority in this archive, still await publication.
Dioscoros's papers provide a (p. 242) wealth of information on the Christianized environ­
ment in which the family lived and more generally on the society and economy of sixth-
century Egypt. Dioscoros's native language was Egyptian, and his belief was Christian,
but he was also versed in the pagan Greek culture and had studied Roman law.

Figure 10.8. Language use in archives up to 600 CE.

In all, about one third of the Graeco-Roman and
Byzantine archives are bilingual (Greek, demotic,
Latin, or Coptic): They may be “real” bilingual
archives, which show a good balance between the
two languages, or “limited” ones, in which one of the
two languages is predominant. Graph: W. Clarysse.

The archive can be divided into several, well-delineated periods. Some of the oldest docu­
ments are related to Dioscoros's father, Apollos, who moved the family into the upper
classes of Aphrodito and was eventually accorded the honorific nomen “Flavius.” In the
last decade of his life, Apollos retired to a monastery that he had himself founded.
Dioscoros received a higher education in Antinoopolis or Alexandria. Following in his
father's footsteps, he became village headman and received numerous petitions. After
Page 24 of 38
Archives and Dossiers

acting as notary in the nome metropolis Antinoopolis for some years, he returned to
Aphrodito sometime between 570 and 573. Having a great interest in Christian and pa­
gan literatures, Dioscoros maintained a private library that included works by Homer and
the comedy writer Menander. In his spare time he appears to have been an enthusiastic
poet of wedding songs and the like, written in classical Greek meters.

(p. 243) Classifying Archival Papers and Files

Archives are useful if they are easily accessible. Graeco-Roman Egypt organized its public
records well, and especially for the Roman period we can describe in detail how these
records were kept (Wolff 1978; Cockle 1984; Burkhalter 1990). Here I focus on visible
traces of classification found among the preserved archives.

Avery simple way to classify papers is to roll them up together and write a label on the
strip of papyrus or linen used to tie up the package, such as “letters” (archive of ⋆Pete­
harsemtheus, son of Panobchounis).

Several official and business archives of the third century BCE have a registration system
for letters, which demonstrates that these papers were systematically arranged (e.g., the
archive of the official ⋆Tesenouphis and the business archive of ⋆Zenon).

Figure 10.9. The estate manager Zenon usually en­

dorsed letters he received with the place and date of
receipt, as well as the subject. On the back of this
letter from Apollonios, written in the minister's chan­
cellery, Zenon made the unusual addition of the hour
of receipt, as urgent matters were involved. The
docket, written to the left of the address, reads:
“Year 32, month of Mesore 2, at the 10th hour [that
is, about 4 PM]. [Letter by] Apollonios about the ani­
mals for the envoys from Pairisades and Argos ” (P.Lo
nd. VII 1973 verso).

Page 25 of 38
Archives and Dossiers

Figure 10.10. The Roman administration often col­

lected individual documents by pasting them togeth­
er into a tomos synkollesimos (“pasted roll”). The
sheets were numbered at the top: The photograph
shows pages 102, 103, and 104 (ρβ, ργ, ρδ) (second
century CE; Musées Royaux dʼArt et dʼHistoire, Brus­
sels inv. E 7616). Reproduced courtesy of the Musées
Royaux dʼArt et dʼHistoire, Brussels.

Beginning in the second century BCE, a more evolved system of classification was devel­
oped for official archives and eventually became the standard in the Roman administra­
tion. Documents were pasted together into a tomos synkollesimos. Usually documents of a
single type were collected in chronological, topographical, or alphabetical order. This
method of classification was applied to several types of documents, not least to declara­
tions. The individual sheets were referred to by the number of the roll and that of the
pasted sheet within the roll. (p. 244) (p. 245) Pasted rolls of more than a hundred sheets
were common (Clarysse 2003b). This classification system was rarely applied to private
documents. ⋆Asklepiades (alias Asklas), for instance, pasted his private and business let­
ters for a seven-year period into one roll in chronological order by date of receipt, added
at the top of the letter (29–23 BCE).

Ostraca Groups: Archives Sui Generis

The bulk of the archives from Graeco-Roman Egypt are papyrus archives, but groups of
interrelated documents also appear among the ostraca (and wooden tablets); these are
usually small archives of tax receipts. It may be useful to distinguish between the “older”
and the “younger” ostraca of buildings or towns. The older ones do not date from the last
period of habitation and are usually found in rubbish dumps or the lower or older layers
of buildings. The younger ostraca belong to the last period of habitation and are often
kept in a niche or a corner of the taxpayers’ dwellings. “Older” and “younger” are relative
terms: “A group of ostraca can be sent to the dump very quickly, but this discarding may
Page 26 of 38
Archives and Dossiers

be the effect of the rehabilitation of a space cleaned after months or years” (O.Claud. I, p.
21). Ostraca have sometimes been found in the same building as papyrus archives but
never in exactly the same spot. They are, as Montevecchi (1988, 248) has put it, rather
archives sui generis.

Archives of younger ostraca may be reconstructed if they have been found during legal
excavations. Few Ptolemaic ostraca were recovered under such conditions, as they usual­
ly do not belong to the last period of habitation. A large such archive is that of ⋆Hor, who
was attached to the sanctuary of the ibis in the Memphite Serapeum in the second centu­
ry BCE. His drafts of dreams, oracles, and so on, mainly written in demotic, were discov­
ered during excavations in 1965–1966 “while clearing the sand from a rough stone-built
chapel adjoining the ibis-galleries at Saqqâra, ” where Hor was working (O.Hor). Roman-
Byzantine houses have yielded more ostraca archives. In modern Aln Waqfa in the oasis of
Kharga, 82 Greek and 3 Coptic ostraca were discovered in and around two houses (called
A and D). A group of 43 ostraca comes from one room of house D and is related to the
farmer Poules (O.Waqfa).

Recent publications dealing with ostraca found during scientific excavations fortunately
pay particular attention to the archaeological context (see chapter 2). The first volume of
the Roman ostraca from Karanis in 1935 indicates the exact findspot of each ostracon:
“Most of them were found in the ruins of ancient houses, among the rubbish of floors or
cellar bins, or in wall niches where they had been carefully put aside. They were discov­
ered singly or in groups” (O.Mich. I, p. IX; (p. 246) the later editions follow the same sys­
tem). The volume by Bingen and Clarysse (O.Elkab) of the Greek and demotic ostraca
found in Elkab is a first-rate example of an edition of ostraca. The archaeological context
and the findspots are given in detail, and the work is illustrated with surveyable maps.

If ostraca were acquired on the antiquities market or discovered in rubbish dumps, the
original archives may be irretrievably lost. Most Greek and Latin ostraca from Mons Clau­
dianus in the Eastern Desert, for instance, were found on a trash heap called the “south­
ern sebakh, ” which lay south of the Roman fortress and east of the grain depot (O.Claud.
I–III; see chapter 2). As the original archives cannot be reconstructed, the ostraca are
published by type (e.g., “lettres privées”) or by topic (e.g., “la mort et la maladie”). Some
dossiers are put together, especially the correspondence of people like Petenophotes
(O.Claud. II 243–254), Dioscoros and others (O.Claud. II 224–242); the editors stress that
these letters are dossiers grouped together for prosopographical reasons (e.g., O.Claud.
II, p. 43; Martin 1994, 572). Sometimes, however, ostraca archives may be reconstructed
by using the various methods for the reconstruction of papyrus archives: the study of
prosopographical data, museum archaeology, or reliance on the archaeological context;
for instance, ostraca may be assembled into groups that were dumped together, as in the
case of the Berenike ostraca (O.Berenike).

Page 27 of 38
Archives and Dossiers

Moments of Joy
Archives may bring papyrologists moments of joy. The letters quoted here introduce us to
a happy and an unhappy family, as their private correspondence illustrates.

On several occasions Paniskos, a soldier or a merchant engaged in the armor trade, asks
his wife, ⋆Ploutogeneia, to come to Koptos. In the first letter everything seems to be going
well:

First of all, I pray daily for your good health in the presence of all the gods. I
would have you know then, that we have been staying in Koptos near your sister
and her children, so that you may not be grieved about coming to Koptos…. So
when you have received this letter of mine, make your preparations in order that
you may come at once if I send for you. And when you come, bring … six jars of
olives, four jars of liquid honey, and my shield, the new one only, and my helmet….
If you find an opportunity, come here with good men…. When you come, bring
your gold ornaments, but do not wear them on the boat. (translation P.Mich. III
214)

Ploutogeneia, however, has no intention of joining her husband at Koptos, thus infuriating
Paniskos: “I have sent you three letters and you have not written me (p. 247) even one. If you do
not wish to come up to me, no one compels you…. The letter carrier said to me when he came to
me: ‘When I was on the point of departing I said to your wife and her mother “Give me a letter to
take to Paniskos” and they did not give it’” (translation P.Mich. III 217).
A totally different picture emerges from another archive of letters addressed mostly to
⋆
Saturnila, mother of five sons. Her papers are referred to as the “happy family archive”
because of the “civilized and affectionate relationship between mother and adult
sons” (Bell 1950):

Sempronius to his mother and lady Saturnila, very many greetings. Before every­
thing I pray for your good health and that also of my sweetest brothers, and at the
same time I make daily supplication for you before the lord Serapis. Because I
found someone who is coming to you upstream, I hurried to greet you by a letter.
Do please, my lady, without delay write to me about your welfare, so that I may
feel less anxiety. (translation P.Mich. XV 752)

But even this happy family had its moments of sadness, as, for instance, when Saturnila's son
Maximus mourned after his wife's death (Sijpesteijn 1976).
Historians who rely on guides or editions of papyrus archives need to be aware that these
studies are often the result of a long process of puzzling work. The major part of the
archives have not been found in their original depository. Even in antiquity the individual
items became mixed up in the wastepaper basket, on rubbish dumps, in mummy cases,
and in mummified crocodiles. In more recent times native diggers carrying out illegal ex­
cavations split up archives and even individual texts before putting them on the antiqui­
ties market. Papyrologists have several methods at hand enabling them to rebuild the an­
cient archives in an iterative process; they can avail themselves of published, as well as

Page 28 of 38
Archives and Dossiers

unedited, texts or fragments spread all over the world. Although reconstructions repre­
sent a fascinating aspect of papyrological research, they are never considered definitive
unless the papers have been found intact in their original setting.

The richness of archival research is reflected by the numerous types of archives, by the
variety of documents constituting them, and by the differences in duration: Some
archives are short lived, while others continue for several generations. Although as a rule
only papers with a juridical value were kept, several archive holders were only acting like
human beings when they filed letters and the like for sentimental reasons, thereby allow­
ing us to witness marital rifts or a mother's love for her children.

The historian who gathers scattered clues from isolated texts may come to similar conclu­
sions as an archival researcher. But, whereas with isolated texts interpreters are forced
to make more speculative reconstructions since no immediate context is available,
archives simply provide this context. They allow the historian to reconstruct with a high
degree of certainty an overall picture of the life (p. 248) or career of an official, a business­
man, or a family: Their papers may depict their cultural background and mother tongue,
the class to which they belonged or were promoted to, their pagan or Christian belief, the
land they possessed or leased, the house they lived in or the country house they owned,
their job and ambitions, as well as their little worries and big problems and how they
dealt with them. Archival research may create a microcosm of Graeco-Roman society and
provide testing cases for models of social and economic behavior.

Archives Mentioned in This Chapter, Listed Al­

phabetically
For each archive the entry gives the provenance, the range of attested dates (ce unless
otherwise indicated), the principal edition, and information about the Leuven Homepage
of Papyrus Collections (LHPC) record.

⋆
Adamas, sitologos (199–197 BCE, Fayyum): P.Tebt. III 750–756, 941, 944–945; LHPC
Clarysse 2003–2006; ⋆Alopex, descendants of Alopex (308–355; Panopolis): P.Panop. and
P.Dubl.; LHPC Geens 2003–2006; ⋆Amenothes, son of Horos (171–116 BCE; Thebes):
P.Tor.Amen.; Depauw 1997, 157; LHPC Clarysse 2003–2006; ⋆Apollinarios, strategos of the
Panopolite nome (298–300): Skeat 1961; P.Panop.Beatty, pp. XXI–XXV; LHPC Geens 2003–
2006; ⋆Apollonios from Bakchias (about 75–84): LHPC Smolders 2003–2006; Smolders
2004, 233–237; ⋆Apollonios, strategos of the Apollonopolites Heptakomias (113–120; Her­
mopolis): P.Giss.Apoll.; Clarysse 1989; Rowlandson 1998, 118–124; Litinas 2001; Messeri
2001; ⋆Apynchis, son of Tesenouphis (83; Soknopaiou Nesos): BGU XI 2095–2100; LHPC
Geens 2003–2006; ⋆Aristarchos, nomarches (250–238 BCE, Fayyum): Héral 1991, 1992;
LHPC Van Beek 2003–2006; ⋆Asklepiades alias Asklas (29–23 BCE; Herakleopolites): Ols­
son 1925, nos. 1–7; White 1986, nos. 63–65; LHPC Geens 2003–2006; ⋆Aurelius Heras,
praepositus pagi (316–318; Oxyrhynchus): Pruneti 1994; LHPC Geens 2003–2006;
⋆
Choachytai: Theban Choachytai Osoroeris and Panas: P.Choach.Survey; Depauw 1997,

Page 29 of 38
Archives and Dossiers

157; ⋆Claudius Tiberianus (100–120; Karanis): Pighi 1964; Adams 2003, 593–596; Strassi
2004; ⋆Dionysios son of Kephalas (141–103 BCE; Tenis-Akoris): Pap.Lugd.Bat. XXII; Lewis
1986 (chapter 8.1); Pestman 1989, 71–74; Depauw 1997, 157–158; LHPC Van Beek 2003–
2006; ⋆Dioscoros (506–585; Aphrodito): Maltz 1957; MacCoull 1981 (on the Coptic pa­
pyri); Clarysse 1983, 55–57; Verbeeck 1989; Rowlandson 1998, 151–154; Fournet 1999,
2001; ⋆Dryton, his wife, Apollonia, and their offspring (172–94 BCE; Pathyris): P.Dryton;
Rowlandson 1998, 105–112; LHPC Vandorpe 2003–2006; ⋆Erbstreit (186–133 BCE; Pa­
thyris): P.Erbstreit; Ritner 1984; (p. 249) LHPC Vandorpe 2003–2006; ⋆Eutychides, son of
Sarapion (Hermopolis): P.Sarap.; Kehoe 1992, 67–72; ⋆Flavius Abinnaeus, praefectus alae
(342–351; Dionysias—Phila-delphia): P.Abinn.; Barnes 1985; Pestman 1989, 134–137; LH­
PC Geens 2003–2006; ⋆Gaius Iulius Agrippinus (103/117–148; Karanis): Meyer 1906a,
1906b; Maehler 1970, 1982; Rupprecht 2001; LHPC Geens 2003–2007; ⋆Heroninus (247–
270; Theadelphia): Pintaudi 1976; Clarysse 1983, 47; Rathbone 1991; Kehoe 1992, 92–
117; Ferro 1994; ⋆Hor(168–151 BCE; Memphis): O.Hor; Depauw 1997, 158; ⋆Horos, son of
Nechouthes: P.Adl.; Pestman 1965, 47–48; Herrmann 1975; Messeri-Savorelli 1990; De­
pauw 1997, 155; LHPC Waebens 2008; ⋆Isidoros, village scribe (161–164, Trikomia and
Lagis): LHPC Smolders 2003–2006; ⋆Iulius Serenus (179–216, Karanis): P.Hamb. 158–196;
LHPC Clarysse 2003–2006; ⋆Kleon and Theodoros: Lewis 1986 (chapter 2); Van Beek 2005;
LHPC Van Beek 2003–2006; ⋆Krokodilopolis, notarial office (?) (Fayyum): P.Petr.2 I; ⋆Kro­
nion (second century; Tebtynis): P.Kron.; P.Mil. Vogl. II, III, VI; Rowlandson 1998, 125–
133; Bagnall 2000; LHPC Smolders 2003–2006; ⋆Melitian monks of Labla (511–513):
McGing 1990; LHPC Clarysse 2003–2006; ⋆Menches, goldsmith (ca. 195 BCE, Oxyrhyn­
cha): P.Mich. XVIII 771–774; ⋆Menches, village scribe: Crawford 1971; P. W. Pestman in
P.Rain.Cent., pp. 127–134; Lewis 1986 (chapter 7); Verhoogt 1998a, 1998b, 2005; LHPC
Vandorpe 2003–2006; ⋆Milon, praktor (225–222 BCE; Edfu): Depauw 1997, 159; Clarysse
2003a; LHPC Clarysse 2003–2006; ⋆Oxyrhyncha, granary of (152–149 BCE): P.Erasm. I
and II; ⋆Pabachtis, son of Paleuis (265–208 BCE; Edfu): P.Hausw.Spiegelberg;
P.Hausw.Manning; Depauw 1997, 155; LHPC Müller 2003; ⋆Pakebkis, descendants of
(126–162; Tebtynis): Melaerts 1991; Bagnall 2000; LHPC Smolders 2003–2006; ⋆Pates and
Pachrates, correspondence of Pates and Pachrates (103–101 BCE; Pathyris): P.War of
Scepters, pp. 37–81; LHPC Clarysse 2003–2006; ⋆Pathyris, archeion of (ca. 111–110 BCE):
Vandorpe 2004, 166–167; ⋆Patronids (formerly called the archive of Laches) (108–176;
Tebtynis): Bagnall 1973; Clarysse and Gallazzi 1993; Bagnall 2000; LHPC Smolders 2003–
2006; ⋆Petaus, village scribe (182–187; Ptolemais Hormou): P.Petaus; ⋆Peteharsemtheus,
son of Panobchounis: Pestman 1965; Lewis 1986 (chapter 8.2); LHPC Waebens 2008; ⋆Phane­
sis, oil seller (233–223 BCE, Tebtynis): Muhs, Grunewaldt, and Van den Berg-Onstwedder
2002–2003; LHPC Clarysse 2003–2006; ⋆Philosarapis (89–224; Tebtynis): P.Fam.Tebt.;
LHPC Smolders 2003–2006; ⋆Ploutogeneia (296–298; Philadelphia): P.Mich. III 275–298;
Schwartz 1968; Rowlandson 1998, 147–151; LHPC Smolders 2003–2006; ⋆Pompeius Niger
(31–64; Oxyrhynychus?): Whitehorne 1988; Rathbone 2001; LHPC Smolders 2003–2006;
⋆Praktores argurikon of Soknopaiou Nesos (an early third-century tax archive): Lewis
1954, 297–298; Nachtergael 2005, 232–235; ⋆Ptolemaios and his brother, Apollonios,
recluses in the Memphite Serapeum (164–152 BCE): UPZ I; Clarysse 1983, 57–60; Lewis
1986 (chapter 5); Clarysse 1986; Hoogendijk 1989; ⋆Pyrrheia, granary of Pyrrheia:

Page 30 of 38
Archives and Dossiers

Clarysse and Hauben 1991; LHPC Clarysse 2003–2006; ⋆Sarapias (165–264/270; Tebty­
nis): Verhoogt 1998a; Clarysse 2006; LHPC Smolders 2003–2006; ⋆Saturnila (about 175–
199; Karanis?): Bell 1950; Sijpesteijn 1976; Rowlandson 1998, (p. 250) 143–147; Bar­
renechea 2001; LHPC Van Beek 2003–2006; ⋆Socrates, tax collector (106–171; Karanis):
Van Minnen 1994, 237–244; Strassi 2001; ⋆Soknobraisis, temple of Soknobraisis at
Bakchias (116–212): Gilliam 1947; LHPC Clarysse 2003–2006; ⋆Taienteus (315–274 BCE;
Thebes): P.Brit.Mus. I; P.Ryl.Dem. 10–14; Pestman 1989, 14–24; Depauw 1997, 156; ⋆Tefhapi
from Siut (185–169 BCE; Lycopolites): P.Siut; Shore-Smith 1959; Vleeming 1989; Depauw
1997, 157; ⋆Teos and his wife, Thabis: P.Teos; Depauw 1997, 156; LHPC Depauw 2003–
2006; ⋆Tesenouphis, toparches (224–217 BCE, Fayyum): P.Sorb. I 38–55; LHPC Clarysse
2003–2006; ⋆Theogonis, grapheion of Theogonis: CPR XVIII; ⋆Thmouis, record office in
Thmouis (Mendesian nome): P.Thmouis 1–4; ⋆Totoes: P.Tor.Botti, esp. 204–205; PSI IX
1014–1025; Pestman 1989, 24–29; Depauw 1997, 156; ⋆Zenon (mid-third century BCE;
Philadelphia, Fayyum): for bibliography see Pap.Lugd.Bat. XXI; Clarysse and Vandorpe
1995; Rowlandson 1998, 95–98.

Bibliography
Adams, J. N. (2003). Bilingualism and the Latin Language. Cambridge: Cambridge Univer­
sity Press.

Bagnall, R. S. (1995). Reading Papyri, Writing Ancient History. London: Routledge.

———. 2000. “Village and Urban Elites in Roman Tebtunis.” In Bancroft Library Gallery,
Berkeley. http://tebtunis.berkeley.edu/ancientlives/bagnall.html.

Bagnall, W. S. 1973. “Some Prosopographical Observations on the Laches Archive.” BASP

10: 65–70.

Barnes, T. D. 1985. “The Career of Abinnaeus.” Phoenix 39: 368–374.

Barrenechea, F. 2001. “A New Letter from the Sempronius Dossier: A Letter from Max­
imus.” BASP 38: 21–34.

Bell, H. I. 1950. “A Happy Family.” In Aus Antike und Orient: Festschrift Wilhelm Schubart
zum 75. Geburtstag, ed. S. Morenz and W. Schubart, 38–47. Leipzig: Harrassowitz.

Boussac, M.-F. 1993. “Archives personnelles à Délos”. In Comptes rendus des séances de
lʼAcadémie des Inscriptions et Belles-lettres (CRAI). 677–693. Paris: Klincksieck.

Brosius, M., ed. (2003). Archives and Archival Traditions: Concepts of Record-keeping in
the Ancient World. Oxford: Oxford University Press.

Burkhalter, F. 1990. “Archives locales et archives centrales en Egypte romaine.” Chiron

20: 191–215.

Page 31 of 38
Archives and Dossiers

Capasso, M. 1996. “Catalogo dei papiri di Bakchias, II: i PBakchias 101–133.” In Bakchias
III: Rapporto preliminare della campagna di scavo del 1995, ed. S. Pernigotti and M. Ca­
passo, 119–147. Studi di egittologia e di antichità puniche. Monografie, series maior 3,
Pisa: Giardini.

Clackson, S. J. 2004. “Museum Archaeology and Coptic Papyrology: The Bawit Papyri.” In
Coptic Studies on the Threshold of a New Millennium: Proceedings of the Seventh Inter­
national Congress of Coptic Studies. Leiden 27 August-2 Sept. 2000, ed. (p. 251) M. Im­
merzeel and J. van der Vliet, vol. 1, 477–490. Orientalia Lovaniensia Analecta 133. Leu­
ven: Peeters.

Clarysse, W. 1983. “Literary Papyri in Documentary ‘Archives.’” In Egypt and the Hellenis­
tic World, ed. E. Van't Dack et al., 43–61. Studia Hellenistica 27. Leuven: Peeters.

———. 1986. “UPZ I 60, a Reconstruction by Revillout.” Enchoria 14: 43–49.

———. 1989. “Apollonios: Ambtenaar en familievader.” In Familiearchieven uit het land

van Pharao, ed. P. W. Pestman, 85–106. Zutphen, the Netherlands: Terra.

———. 2003a. “The Archive of the Praktor Milon.” In Edfu, an Egyptian Provincial Capital
in the Ptolemaic Period, Handelingen Contactforum Brussels, 3 September 2001, ed. K.
Vandorpe and W. Clarysse, 17–27. Brussels: Koninklijke Vlaamse academie van België
voor wetenschappen en kunsten.

———. 2003b. “Tomoi Synkollesimoi.” In Archives and Archival Traditions: Concepts of

Record-keeping in the Ancient World, ed. M. Brosius, 344–359. Oxford: Oxford University
Press.

Clarysse, W., and C. Gallazzi. 1993. “Archivio dei discendenti di Laches o dei discendenti
di Patron?” AncSoc 24: 63–68.

Clarysse, W., and H. Hauben. 1991. “Ten Ptolemaic Granary Receipts from Pyrrheia.” ZPE
89: 47–68.

Clarysse, W., and K. Vandorpe. 1995. Zénon, un homme dʼaffaires grec à lʼombre des pyra­
mides (Ancorae: Steunpunten voor Studie en Onderwijs 14). Leuven: Presses universi­
taires de Louvain.

Cockle, W. E. H. 1984. “State Archives in Graeco-Roman Egypt from 30 BC to the Reign of

Septimius Severus.” JEA 70: 106–122.

Crawford, D. J. (1971). Kerkeosiris, an Egyptian village in the Ptolemaic period. Cam­

bridge: Cambridge University Press.

———. 1973. “The Opium Poppy. A Study in Ptolemaic Agriculture.” In Problemès de la

terre en Gréce ancienne, ed. M. I. Finley, 223–251. Civilisations et Sociétés 33. Paris:
Mouton.

Page 32 of 38
Archives and Dossiers

Davoli, P. (1998). Lʼarcheologia urbana nel Fayyum di età ellenistica e romana. Missione
congiunta delle Università di Bologna e di Lecce in Egitto. Monografie 1 Athenaeum.
Napoli: Procaccini.

———. 2005. Oggetti in argilla dallʼarea templare di Bakchias (El-Fayyum, Egitto): Catalo­
go dei rinvenimenti delle Campagne di Scavo 1996–2002. Biblioteca degli Studi di egit­
tologia e di papirologia 3. Pisa: Giardini.

Depauw, M. (1997). A Companion to Demotic Studies (Pap.Brux. 28). Brussels: Fondation

égyptologique Reine Elisabeth.

———. (2000). The Archive of Teos and Thabis from Early Ptolemaic Thebes: P. Brux.
Dem. Inv. E. 8252–8256. Monographies Reine Elisabeth 8. Turnhout: Brepols.

———. 2003. “Autograph Confirmation in Demotic Private Contracts.” CdÉ 78: 66–111.

Ferro, C. 1994. “I rendiconti dellʼarchivio di Heroneinos.” Analecta Papirologica 6: 37–51.

Fournet, J.-L. (1999). Hellénisme dans lʼÉgypte du VIe siècle: La bibliotheque et lʼœuvre
de Dioscore dʼAphrodité (MIFAO 115). Cairo: Institut francais dʼarchéologie orientale.

———. 2001. “Du nouveau dans les archives de Dioscore dʼAphrodité.” In Atti del XXII
Congresso Internazionale di Papirologia, Firenze, 23–29 agosto 1998, ed. I. Andorlini, G.
Bastianini and M. Manfredi, 475–485. Florence: Università di Firenze. Istituto papirologi­
co G. Vitelli.

Gallazzi, C. 1990. “La ‘Cantina dei Papiri’ di Tebtynis e ciò che essa conteneva.” ZPE 80:
283–288.

Gilliam, E. H. 1947. “The Archives of the Temple of Soknobraisis at Bacchias.”

(p. 252)

Yale Classical Studies 10: 179–281.

Godlewski, W. 1986. Deir el-Bahari V: Le monastère de St. Phoibammon. Warsaw: PWN.

Hagedorn, D. 1970. “Papyri aus Panopolis in der Kölner Sammlung.” In Proceedings of

the XIIth International Congress of Papyrology, Ann Arbor, August 13–17, 1968, ed. D.H.
Samuel, 207–211. Toronto: Hakkert.

Héral, S. 1991. Les archives bilingues du nomarque Aristarchos dʼaprès les papyrus de
Ghôran. Mémoire de D.E.A., Institut de Papyrologie, Université de Paris IV–Sorbonne.

———. 1992. “Les archives bilingues de nomarques dans les papyrus de Ghôran.” In Life
in a Multi-cultural Society: Egypt from Cambyses to Constantine and Beyond, ed. J. H.
Johnson, 149–157. Studies in Ancient Oriental Civilization 51. Chicago: Oriental Institute.

Herrmann, J. 1975. “Sachteilung und Wertteilung bei Grundstiicken: Zu den griechischen

Kaufurkunden des Horus-Archivs.” In Festschrift für E. Seidl. Zum 70. Geburtstag, ed. H.
Hübner, 53–60. Cologne: Hanstein.

Page 33 of 38
Archives and Dossiers

Hoogendijk, F. A. J. 1989. “Ptolemaios: Een Griek die leeft en droomt in een Egyptische
tempel.” In Familiearchieven uit het land van Pharao, ed. P. W. Pestman, 46–69. Zutphen:
Terra.

Jördens, A. 2001. “Papyri und private Archive: Ein Diskussionsbeitrag zur papyrologis­
chen Terminologie.” In Symposion 1997: Vorträge zur griechischen und hellenistischen
Rechtsgeschichte (Altafiumara, 8–14 Sept. 1997), ed. E. Cantarella and G. Thur, 253–267.
Cologne: Böhlau.

Kehoe, D. P. (1992). Management and Investment on Estates in Roman Egypt during the
Early Empire. Papyrologische Texte und Abhandlungen 40. Bonn: R. Habelt.

Leuven Homepage of Papyrus Collections (LHPC). Database-driven website containing a

recent list and detailed descriptions ofarchives from Graeco-Roman and Byzantine Egypt.
Contributions by W. Clarysse, M. Depauw, K. Geens, K. Muller, R. Smolders, B. Van Beek,
K. Vandorpe and S. Waebens. Leuven 2003–2008. http://www.trismegistos.org/arch/
index.php.

Lewis, N. 1954. “Miscellanea Papyrologica.” CdÉ 29: 288–298.

———. (1986). Greeks in Ptolemaic Egypt: Case Studies in the Social History of the Hel­
lenistic World. Oxford: Oxford University Press.

Litinas, N. 2001. “A Letter from the Strategos Apollonios Archive? P.Lond.inv. 1228.” In
Atti del XXII Congresso Internazionale di Papirologia, Firenze, 23–29 agosto 1998, ed. I.
Andorlini, G. Bastianini and M. Manfredi, 805–812. Florence: Università di Firenze. Istitu­
to papirologico G. Vitelli.

MacCoull, L. S. B. 1981. “The Coptic Archive of Dioscoros of Aphrodito.” CdÉ 56: 185–
193.

Maehler, H. 1970. “Neue Dokumente zum Drusilla-Prozess.” In Proceedings of the XIIth

International Congress of Papyrology, Ann Arbor, August 13–17, 1968, ed. D.H. Samuel,
263–271. American Studies in Papyrology 7. Toronto: Hakkert.

———. 1982. “Neues vom Prozess der Drusilla gegen Agrippinus.” In Symposion 1997:
Vorträge zur griechischen und hellenistischen Rechtsgeschichte (Altafiumara, 8–14 Sept.
1997), ed. E. Cantarella and G. Thur, 325–333. Cologne: Böhlau.

Maltz, G. 1957. “The Papyri of Dioscoros.” In Studi in onore di Aristide Calderini e Rober­
to Paribeni. 2, 345–356. Milan: Ceschina.

(p. 253) Martin, A. 1994. “Archives privées et cachettes documentaires.” In Proceedings of

the 20th International Congress of Papyrologists. Copenhagen, 23–29 August 1992, ed. A.
Bülow-Jacobsen, 569–577. Copenhagen: Museum Tusculanum.

McGing, B. C. 1990. “Melitian Monks at Labia.” Tyche 5: 67–91.

Page 34 of 38
Archives and Dossiers

Melaerts, H. 1991. “Autour du P.Mil.Vogl. II, 73.” CdÉ 66: 266–278.

Messeri, G. 2001. “Suggestioni da PSIIV 308.” ZPE 135: 165–168.

Messeri-Savorelli, G. 1990. “Un nuovo documento dellʼarchivio di Horos Wglio di

Nechutes.” Analecta Papyrologica 2: 53–61.

Meyer, P. M. 1906a. “Papyrus Cattaoui. II: Kommentar.” APF 3: 67–105.

———. 1906b. “Zum Drusilla-Prozess.” APF 3: 247–248.

Montevecchi, O. (1988). La Papirologia, 2d ed. Milan: Vita e pensiero.

Muhs, B., A. Grünewaldt, and G. Van den Berg-Onstwedder. 2002–2003. “The Papyri of
Phanesis, Son of Nechturis, Oil Merchant of Tebtunis, and the Ptolemaic Cloth Monopoly.”
Enchoria 28: 62–81.

Nachtergael, G. 2005. “Papyrologica II.” CdÉ 80: 229–245.

Olsson, B. (1925). Papyrusbriefe aus der frühesten Römerzeit. Uppsala: Almqvist och Wik­
sell.

Orrieux, Cl. (1983). Les papyrus de Zénon: lʼhorizon dʼun grec en Egypte au IIIe siècle
avant J. C. Deucalion. Paris: Macula.

———. (1985). Zénon de Caunos, parépidèmos, et le destin grec. Annales littéraires de

lʼUniversité de Besançon 320. Paris: Belles Lettres.

Pantalacci, L. (2008). La lettre dʼarchive: communication administrative et personelle

dans lʼantiquité proche-orientale et égyptienne. Topoi: Orient-Occident. Supplément 9.
Lyon: Maison de lʼOrient et de la Méditerranée.

Pestman, P. W. 1965. “Les archives privées de Pathyris à lʼépoque ptolémaïque: La famille

de Pétéharsemtheus, Wls de Panebkhounis.” In Pap.Lugd.Bat. XIV: 47–105.

———, ed. (1989). Familiearchieven uit het land van Pharao. Zutphen: Terra.

———. 1995. “A Family Archive Which Changes History.” In Hundred-gated Thebes: Acts
of a Colloquium on Thebes and the Theban Area during the Graeco-Roman Period, ed. S.
P. Vleeming, 91–100. Pap.Lugd.Bat 27. Leiden: Brill.

Pighi, G. B. (1964). Lettere latine di un soldato di Traiano (P.Mich. 467–472). Bologna:

Zanichelli.

Pintaudi, R. 1976. “Papiri Worentini dellʼarchivio di Heronas.” ZPE 20: 233–248.

Pruneti, P. 1994. “Lʼarchivio di Aurelius Heras praepositus pagi.” Aegyptus 74: 33–36.

Page 35 of 38
Archives and Dossiers

Rathbone, D. W. 1991. Economic Rationalism and Rural Society in Third-century A.D.

Egypt: The Heroninos Archive and the Appianus Estate. Cambridge: Cambridge Universi­
ty Press.

———. 2001. “PSI XI 1183: Record of a Roman Census Declaration of A.D. 47/8.” In Es­
says and Texts in Honor of J. David Thomas, ed. T. Gagos and R. S. Bagnall, 99–113. Amer­
ican Studies in Papyrology 42. Oakville, Conn.: American Society of Papyrologists.

Ritner, R. K. 1984. “Property Transfer.” In Grammata Demotika: Festschrift für E. Lüd­

deckens zum 15. Juni 1983, ed. H.-J. Thissen and K.-Th. Zauzich, 171–187. Würzburg: Za­
uzich.

Rowlandson, J., ed. (1998). Women and Society in Greek and Roman Egypt: A Sourcebook.
Cambridge: Cambridge University Press.

Rupprecht, H.-A. 2001. “Ein Verfahren ohne Ende: Der Prozess der Drusilla.” In Atti del
XXII Congresso Internazionale di Papirologia, Firenze, 23–29 agosto 1998, ed. I. Andorli­
ni, G. Bastianini and M. Manfredi, 1135–1144. Florence: Università di Firenze. Istituto pa­
pirologico G. Vitelli.

Ryholt, K. 2005. “On the Contents and Nature of the Tebtunis Temple Library: A
(p. 254)

Status Report.” In Tebtynis und Soknopaiu Nesos: Leben im römerzeitlichen Fajum. Akten
des Internationalen Symposions vom 11. bis 13. Dezember 2003 in Sommerhausen bei
Würzburg, ed. S. L. Lippert and M. Schentuleit, 141–170. Wiesbaden: Harrassowitz.

Schwartz, J. 1968. “Autour du dossier de Paniskos (P.Mich. 214–221).” Aegyptus 48: 110–
115.

Seidl, E. (1962). Ptolemäische Rechtsgeschichte, 2d ed. Ägyptologische Forschungen 22.

Glückstadt: J. J. Augustin.

Seidl, E. 1973. Rechtsgeschichte Ägyptens als römischer Provinz: (Die Behauptung des
ägyptischen Rechts neben dem römischen). Sankt Augustin: Richarz.

Shore, A. F., and H. S. Smith. 1959. “Two Unpublished Demotic Documents from the
Asyut Archive.” JEA 45: 52–60.

Sijpesteijn, P. J. 1976. “A Happy Family?” ZPE 21: 169–181.

Skeat, T. C. 1961. “Papyri from Panopolis in the Collection of Sir Chester Beatty.” In Pro­
ceedings of the IX International Congress of Papyrology, Oslo, 19–22 August 1958, ed. L.
Amundsen and V. Skånland, 194–199. Hertford: Norwegian Research Council for Science
and the Humanities.

Smolders, R. 2004. “Two Archives from the Roman Arsinoites.” CdÉ 79: 233–240.

Strassi, S. 2001. “Le carte di Sokrates Sarapionos, praktor argyrikon a Karanis nel II sec.
d.c.” In Atti delXXII Congresso Internazionale di Papirologia, Firenze, 23–29 agosto 1998,

Page 36 of 38
Archives and Dossiers

ed. I. Andorlini, G. Bastianini and M. Manfredi, 1215–1228. Florence: Università di Firen­

ze, Istituto papirologico G. Vitelli.

———. 2004. “In margine allʼarchivio di Tiberianus e Terentianus: P.Mich. VIII 510.” ZPE
148: 225–234.

Van Beek, B. 2005. “A Letter from Zenon to Kleon: A New Date for P.Zen.Pestm., Suppl.
B.” AncSoc 35: 119–128.

———. 2007. “Ancient Archives and Modern Collections: The Leuven Homepage of Pa­
pyrus Archives and Collections.” In Proceedings of the XXIVth International Congress of
Papyrology (Helsinki, 1–7 August 2004), ed. J. Frösén, T. Purola and E. Salmenkivi, 1033–
1044. Helsinki: Societas Scientiarum Fennica.

Van Minnen, P. 1994. “House-to-House Enquiries: An Interdisciplinary Approach to Ro­

man Karanis.” ZPE 100: 227–251.

Vandorpe, K. 1994. “Museum Archaeology or How to Reconstruct Pathyris Archives.” EVO

17: 289–300.

———. 2004. “A Greek Register from Pathyris' Notarial Office: Loans and Sales from the
Pathyrite and Latopolite Nomes.” ZPE 150: 161–186.

———. 2008. “Persians soldiers and Persians of the Epigone. Social Mobility of Soldiers-
Herdsmen in Upper Egypt.” APF 54: 87–108.

Verbeeck, B. 1989. “Dioskoros: Dorpshoofd, dichter en notaris.” In Familiearchieven uit

het land van Pharao, ed. P. W. Pestman, 138–162. Zutphen: Terra.

Verhoogt, A. M. F. W. 1998a. “Family Papers from Tebtunis: Unfolding a Bundle of Papyri.”

In The Two Faces of Graeco-Roman Egypt: Greek and Demotic and Greek-Demotic Texts
and Studies Presented to P.W. Pestman, ed. S. P. Vleeming and A. M. F. W. Verhoogt, 141–
154. Pap.Lugd.Bat 30. Leiden: Brill.

———. 1998b. Menches, Komogrammateus of Kerkeosiris: The Doings and Dealings of a

Village Scribe in the Late Ptolemaic Period (120–110 BC). (Pap.Lugd.Bat. 29). Leiden:
Brill.

———. 2005. Regaling Officials in Ptolemaic Egypt. A Dramatic Reading of Official Ac­
counts from the Menches Papers. (Pap.Lugd.Bat. 32). Leiden: Brill.

(p. 255) Vleeming, S. P. 1989. “Strijd om het erfdeel van Tefhapi.” In Familiearchieven uit
het land van Pharao, ed. P. W. Pestman, 30–45. Zutphen: Terra.

White, J. L. (1986). Light from Ancient Letters. Philadelphia: Fortress.

Whitehorne, J. E. G. 1988. “More about L. Pompeius Niger, Legionary Veteran.” In Pro­

ceedings of the XVIII International Congress of Papyrology, ed. B. G. Mandilaras, vol. 2,
445–450. Athens: Greek Papyrological Society.
Page 37 of 38
Archives and Dossiers

Wolff, H. J. (1978). Das Recht der griechischen Papyri Ägyptens in der Zeit der Ptolemäer
und des Prinzipats. Vol. 2, Organisation und Kontrolle des privaten Rechtsverkehrs. Hand­
buch der Altertumswissenschaft 10.5.2/Rechtsgeschichte des Altertum 5.2. Munich: Beck.

Notes:

(1.) Archives discussed in this chapter are listed at the end with further information.
Their names are preceded by an asterisk (⋆) in the text.

Katelijn Vandorpe

Katelijn Vandorpe is Professor of Ancient History, Katholieke Universiteit Leuven.

Page 38 of 38
The Ancient Book

The Ancient Book

William A. Johnson
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0011

Abstract and Keywords

From the beginnings of Greek written literature until deep into the Roman era, a “book”
was fashioned by taking a premanufactured papyrus roll, writing out the text, attaching
additional fresh rolls as the length of text required, and, when finished, cutting off the
blank remainder. This article notes that literary texts were produced, in general, with
strict attention. It describes what constituted the ancient book. Books on papyrus in the
form of rolls (bookrolls) were the norm from the beginnings through the early Roman era.
Over the course of the second to the fourth centuries CE, the codex came to replace the
bookroll. The article also considers the content of the books, and the Oxyrhynchite texts.

Keywords: bookroll, papyrus roll, codex, Oxyrhynchite texts, Greek written literature

From the beginnings of Greek written literature until deep into the Roman era, a
“book” was fashioned by taking a premanufactured papyrus roll, writing out the
text, attaching additional fresh rolls as the length of text required, and, when fin­
ished, cutting off the blank remainder. Needed were the papyrus rolls, ink, pen,
sponge, glue, and knife. This could have been a casual process. But, in fact, as the
papyri show us, it was not. A lot follows from this fact—that literary texts were
produced, in general, with strict attention rather than casually—and our first or­
der of business is to understand clearly what constituted the ancient book.

The Bookroll
Books on papyrus in the form of rolls (“bookrolls”) were the norm from the beginnings
through the early Roman era. The first Greek vase representation dates to the early fifth
century, but the Egyptians had used papyrus bookrolls for at least two thousand years be­
fore that. Over the course of the second to the fourth centuries CE, the codex came to re­
place the bookroll (discussed later), but the bookroll was the dominant format for Greek
and, later, Roman literary texts for about one thousand years. During that lengthy period,

Page 1 of 26
The Ancient Book

the look and feel of the bookroll varied in only relatively minor points of style. The ap­
pearance of the book-as-object is remarkably stable over time and place, and this fact al­
lows us to advance some generalities.

The papyrus writing material was, as we have already seen (chapter 1), manufac­
(p. 257)

tured in two-layer sheets made from papyrus pith. The resulting sheets were not unlike
paper today: somewhat thicker and certainly tougher but more or less in keeping with
modern page size; for quality productions, typically 20–25 cm (8–10 in.) in width and 19–
33 cm (7.5–13 in.) in height (Johnson 2004, 88–92). The sheets, however, were not sold
separately, as paper is today but were joined together into rolls of a fixed number of
sheets (typically twenty; Lewis 1974, 54–55; Skeat 1982, 169–72), pasted left over right
so that the writing would flow easily—downhill, as it were—over the join. Manufactured
papyrus rolls were of quality that varied widely, from wrapping paper to writing material.
The elder Pliny (HN 13.74–78) gives us Latin terms for the grades, along with information
on the criteria used for grading. Important were the color, the texture, and the width of
the sheets: Better was white, thin, smooth, and wide sheeted (Johnson 1993). Use of the
top grades appears to have been typical for the production of books. Despite heavy dam­
age to most surviving papyri, one finds evidence of painstaking attention to detail in the
manufacture of the rolls. Manufactured joins, for example, were often constructed in the
manner of a rabbet in cabinet joinery, with a layer of one of the two-layered sheets metic­
ulously stripped away so that the join itself comprised only three layers—thus both
smoother and less prone to damage (J. Rea in Turner 1978, 20; Coles et al. 1985, 115). Lo­
cating the join can be nearly impossible on the back, where the vertical fibers run, but al­
so sometimes surprisingly challenging on the front of the roll, where one would think the
mismatch of horizontal fibers would make the join readily apparent. (The Arden papyrus,
MP 1233, is a good example: See figure 11.1 and comments at Johnson 2004, 91n14.)
Such was the care given to smoothing and joining in the manufacture of better-made
rolls.

Once the scribe took the manufactured roll in the hand, attention to detail did not stop.
The scribe's payment was based on the quality of the writing (P. Lond. inv. 2110 [Bell
1921]; Diocletian, Edictum de pretiis rerum venalium, col. 7.41–43). Books on papyrus are
generally written in a “book hand,” a style of writing in which the letters are kept wholly
or mostly separated to improve legibility. A fair number, perhaps a third of the total, are
written in a formal or semiformal script with at least pretension to elegance. Importantly,
relatively few are written in cursive or otherwise substandard scripts.1 Those written in
book hands, whether calligraphic or nondescript, generally show signs of having been
written by a trained scribe. Most telling is the layout, which tends to be exceedingly exact
in several respects (see later discussion).

Page 2 of 26
The Ancient Book

Figure 11.1. MP 1233 (P.Lond.Lit. 132). The Arden

papyrus, cols. 29–34.

Bookrolls from our earliest direct witnesses (fourth century BCE) are laid out along the
length of the roll in columns running left to right. (See figure 11.2.) Prose texts were writ­
ten in columns that were narrow relative to their height, roughly analogous to a modern
newspaper. Over time, the preferred style may change—wider (it seems) in Ptolemaic
texts, narrower in the early Roman era, wider in the later empire—but the column width
almost always falls within 4.5–9 cm (2–3.5 in.) (p. 258) (p. 259) and usually within 4.5–7.5
cm (2–3 in.), a narrow range. The column height, by comparison, ranges broadly but
tends to be at least twice that of the width—usually within 12–26 cm (5–10 in.), thus en­
suring the general look of a tall, slender column.

Verse texts are different since the column width is simply a function of the script size and
the verse length; for hexameters, the column was quite wide, and for trimeter texts some­
what less so (see table 11.1). Again, over time minor changes in style assert themselves:
For example, Ptolemaic verse texts often evince a preference for as narrow a space as
possible between the columns, to the extent that a long line from one column is allowed
to run right up to the next, while Roman-era verse texts tend toward somewhat wider,
more distinct intercolumnar spacing. Still, like prose texts, bookrolls that contain verse
are generally stable in look and feel over time, stylistic particulars aside. Worth remark is
that prose, with its narrow and (mostly) justified columns, is immediately distinguishable
from poetry.2

Page 3 of 26
The Ancient Book

Figure 11.2. Anatomy of the bookroll.

As described, the layout is remarkably exact for a hand-produced item. The measurement
from the left edge of one column to the next column stays typically within ± 1.5 mm as
one ranges along the roll—the width of a broad pen stroke. Also typical are evenness in
the straightness of the written line, rough evenness in spacing between lines, fairly exact
alignment in the run of the left edge of the (p. 260) (p. 261) column (and, by the Roman
era, rough justification of the right), and the alignment of the top and bottom of the
columns relative to the top and bottom of the roll. A telling detail is scribal attention to
“Maas's law,” named after the scholar (Paul Maas), who first drew attention to the phe­
nomenon. In accordance with this “law,” bookrolls often exhibit a forward tilt in the col­
umn, such that the left and right edges of the column move steadily to the left as the
scribe works his way down to the bottom. (See figure 11.1.) This tilt becomes a regular
feature in the early Roman era and dominates in the second and third centuries. That
Maas's law is a deliberate feature (and not a product of scribal inattention) is confirmed
both by its regularity—in a sample of 192 Oxyrhynchus papyri, 70 percent showed a dis­
tinct, measurable tilt forward, and only two a definite tilt in the other direction—and by
the presence on several papyri of ruling dots set out as guides to keep the angle of the tilt
consistent (Johnson 2004, 91–99).

Table 11.1. Typical Bookroll Dimensions (Johnson 2004)

Dimension Normative Range

roll height Ptolemaic: 19–25 cm

Roman era: 25–33 cm

column height class I: 12–16 cm

class II: 16–21 cm

class III: 21–27 cm

Page 4 of 26
The Ancient Book

column width, prose class I: 4.5–6 cm

class II: 6–7.5 cm

class III (rare): 8–9 cm

column width, verse trimeter: ca. 8–11 cm

hexameter: ca. 11–14 cm

intercolumn, prose class I: ca. 1.5 cm (1.2–1.8)

class II: ca. 2.0 cm (1.9–2.5)

upper margin everyday ms: 3–4 cm

deluxe ms: 3–6 cm

lower margin everyday ms: 3–5 cm

deluxe ms: 3–7 cm

letters per line 13–24 (at extremes 10–30)

lines per column 25–50 (at extremes 18–64)

The bookroll, here and elsewhere, shows distinct signs of deliberate design and attention
to what is stylish, as well as exactness in execution involving both measurement and ex­
pert estimation. All of this is consistent—as a general picture—with the conclusion that
bookrolls were generally the product of scribes trained for the task, that is, to an artisan
apprentice trade. The trade clearly also involves a strong sense of the cultural demands
on the product. The bookroll signaled culture and learning, but for a bookroll to qualify as
such required a particular look and feel with well-defined traditions of detail. Counterex­
amples exist, of course, but—school texts of Homer to one side—it is important to stress
how rare these are.

The writing itself, formed with a view to clarity, is laid out in continuous letters (scriptio
continua), that is, without spaces between words. It is sometimes said that ancient books
lacked punctuation, but that is not strictly true; what is true is that in ancient books the
punctuation was far less elaborated. Minor points of articulation or breath pauses, where
we now place a comma or a colon, are left unpunctuated for the most part; when marked,
raised dots are normally used, and these are often additions by a reader.3 Punctuation is,
however, routine for marking periods (i.e., at the end of a sentence), changes between
speakers in drama and dialogue, and other major points of division, such as the poems
Page 5 of 26
The Ancient Book

within an epigram collection. Points of major division are most often signaled by the para­
graphos (a horizontal line at the left edge of the column); where the division occurs in
midline, the paragraphos is conjoined with a space (more usual in Ptolemaic texts) or a
raised dot (·) at the appropriate point in the line of text. Changes of speaker, whether in
drama or philosophical dialogue, are usually marked with a dicolon (:), in form like today's
colon, to distinguish from the dots that mark the separation of sentences. Other sigla
such as diplê (>) and diplê obelismenê (>—) appear but only sporadically (McNamee 1992).
Little interferes, then, with the run of the text aside from the paragraphoi. For the reader,
the paragraphoi naturally act as landing points for breath and mental pauses and as visu­
al cues for returning to a (p. 262) passage when a reader looks up from the text (Johnson
1994). But the overall effect is (to us) of a radically unencumbered stream of letters.

A few other details of the written text are worth cataloguing as well. In all periods, rules
for the division of words between lines were strictly observed.4 Here again we find evi­
dence of attention to detail and an apparently deliberate distinction from documentary
texts, where such rules are routinely violated. Iota adscript was written in every period,
though erratically applied by the Roman era; the literary papyri show here a clear desire
to maintain traditional orthography even in the face of a significant shift in pronunciation
(most documentary texts drop iota mutum wholesale in the Hellenistic era; Clarysse
1976). Also in every period, scriptio plena (e.g., unelided δέ before a vowel) was a normal
if irregular feature for both poetic and prose texts, written even when not spoken so as to
help with the parsing of the words. From the second century BCE on, an apostrophe, as in
our texts today, was sometimes used to mark elision (e.g., δʼ). In the early Roman era di­
aeresis came to be added over iota and upsilon at the beginning of a word (e.g., ϊνα ϋιος)
and in “organic” uses (e.g., Αϊδι), also apparently to help with word division. In the third
century CE, the habit arose of placing an apostrophe between doubled mutes or liquids
(e.g., τετʼτιγων, P.Oxy. VII 1016.232; cf. Turner and Parsons 1987, 11). Very occasionally
there were other aids to reading, such as a sling underneath (the ancient hyphen), which
indicated that the reader should take the letters together as part of the same word; or a
diastole (in form like our comma), indicating the separation of a letter group. Also occa­
sional was the addition of a breathing or an accent to help disambiguate a letter group.
Except in school exercises, accents are common, however, only in the difficult dialectical
terrain of Greek lyric texts; the same is true for marks of quantity. Common in documents
but rare in bookrolls is the use of indentation (eisthesis) and reverse indentation (ekthesis);
in literary papyri these are normally used only to indicate verse groups or change in me­
ter. In general, then, lectional aids were few, and little by way of help or intervention in­
terrupted the flow of the letters. Thorough training was necessary for one to be able to
read this scriptio continua readily and comfortably.

Note that the net effect is designed for clarity and beauty but not ease of use, much less
mass readership. Importantly, this design is not one of primitivism or ignorance. The an­
cients knew perfectly well, for instance, the utility of word division—the Greek school
texts on papyri (see chapter 14) bear eloquent testimony to the need for emerging read­
ers to practice syllable and word division. Similarly, philhellenism in the early empire led
to the adoption of scriptio continua in Latin literary texts, which earlier had used inter­
Page 6 of 26
The Ancient Book

puncts (raised dots) to divide the words—that is, word division was discarded by the Ro­
mans in deference to Greek aesthetic and cultural traditions. As already mentioned, read­
ers would sometimes add detailed punctuation to texts as a guide to syntax and breath
pauses, yet the punctuation does not become more complex over time: In general the
(p. 263) deliberate scribal practice was to copy only the bare-bones punctuation of major

points of division even when detailed punctuation was available. Strict functionality,
clearly, is not a priority in bookroll design. The bookroll seems, rather, an egregiously
elite product intended in its stark beauty and difficulty of access to instantiate what it is
to be educated. This comes as no surprise. Whatever the cost of papyrus (a topic of de­
bate; see Skeat 1995), the Edictum Diocletiani de pretiis rerum venalium, col. 7.41–43, al­
ready mentioned, makes clear that the major cost of a book—the scribe's work—is fully 2–
2½ times what it costs to have a scribe scribble out a contract: “To a scribe for best writ­
ing, 25 denarii per 100 lines; for second-quality writing, 20 denarii per 100 lines; to a no­
tary for writing a petition or legal document, 10 denarii per 100 lines.”

The documentary papyrologist will want to interpose the observation that there are many
well-made documents, and that is certainly true. At the level of detail, documents rarely
look just like a bookroll—the vertical spacing will be different, for instance, or the
columns will be too wide. But a great many documents look bookish in one respect or an­
other. A book hand may be employed, for example. Fine-grade papyrus stock may be
used. The text may mimic contemporary literary habits of textual presentation like punc­
tuation or the use of iota adscript. Conversely, literary texts written on rolls very rarely
have any specifically documentary features. Still, one must be careful not to overstate the
case. If a third of bookrolls were written with some elegance, more than half were written
in fairly rapid, nondescript hands (mostly “second quality” in the terms of the Edict of
Diocletian, one supposes). Nonetheless, these do not look like tax receipts. Even in the
case of bookrolls written in nondescript hands, there is generally enough attention to de­
tail of layout and text to support the iconography of the bookroll here delineated. A
bookroll was not always a fully elegant product, then, yet also not cheap: Even a second-
quality book hand was twice the expense of the hand typical of a legal document under
Diocletian's Edict.

The bookroll itself also had traditional features that instantiate the need to preserve and
endure, features that are, again, not regular or even usual for documentary texts. The
margins above and below the columns tended to be quite wide: 3–5 cm was a typical mea­
sure for run-of-the-mill manuscripts, 3–7 cm for deluxe, with extreme examples ranging to
8 cm and more (Johnson 2004, 130–141).5 Practicality is in play since the papyrus roll
tended to fray at the edges, but a margin of 2 cm would have served to prevent this.
There is surely something here of aesthetic display and even of conspicuous consumption.
Typical bookrolls used only 40–70 percent of the front papyrus surface available for writ­
ing. The back was typically left blank, again to protect the text written on the inside. At
the front edges of the finished roll was attached a blank sheet as protection against fray­
ing and damage to the front of the written text, and we see such sheets (the protokollon)
occasionally among the papyri. Among the carbonized rolls from Herculaneum are exam­
ples of the rods (umbilici) used to facilitate rolling (p. 264) and unrolling the papyrus with­
Page 7 of 26
The Ancient Book

out crushing the front and end of the roll (Capasso 1995, 73–98)· Literary and artistic
sources describe wrappers for fancy rolls (διϕθέρα, membrana: Lucian, Ind. 16; Tibullus
3.1.9; Martial 1.66.11) and of cases for carrying them (τєῦχος, capsa, scrinium: Horace
Ep. 2.1.268; Martial 1.2.4), but the papyri naturally do not provide direct evidence. Re­
mains of bookrolls do, however, include a number of title tags, often on parchment, which
stuck out from the rolled edged of the papyrus so as to provide a handy label when the
papyrus was stored on a bookshelf or in a carrying case.

In general terms, we can assume that one work or one book of a work was equivalent to
one roll, but complications and exceptions occur. In particular, we are uncertain of the in­
cidence of rolls containing more than one work.6 An Oxyrhynchus papyrus (P.Oxy. 1810)
from the early second century probably contained the five early Philippic speeches of
Demosthenes, for example; a book list in a third-century papyrus (P.Laur. inv. 19662v =
Corp.Pap.Fil. 3) specifies one roll containing Plato's Meno and Menexenus and another
containing the two Hippias dialogues and Euthydemus; a late Ptolemaic papyrus
(P.Berol.inv. 16985 = MP 980) almost certainly contained books 19–22 of the Iliad, and
several other papyri bear witness to the habit (more common early on) of putting two
“books” of Homer in one bookroll. Still, the evidence seems limited to Homer and to short
speeches and philosophical dialogues; there is no hint, for instance, that multiple plays
were placed in a single bookroll.

From the length of dramas and other poetry books, we can with reasonable certainty in­
fer that rolls as short as 3 meters (4 cm in diameter) were common. But how large could a
roll be? Here, too, we encounter complications: A conceptual “book” was not always
housed in a single bookroll. Pliny tells us of a work by his uncle divided into three
“books” (libri) but written in six bookrolls (volumina) “on account of their size”; among
the Herculaneum papyri, the fifth book of Philodemus's On Poems survives in two recen­
sions, one in a single roll and the other (in a slightly more generous format) split into two
rolls. “Clearly, at some point a large book” might, as a matter of convenience, be divided
into two rolls. Judging from the example of On Poems 5, as well as statistical accumula­
tions, that point was somewhere around 15 m, which translates into a nine-centimeter di­
ameter (roughly equivalent to a two-liter bottle of Coca-Cola). Occasional rolls were even
longer: The Iliad roll already mentioned (f!Berol.inv. 16985) seems to have been about 19
m long; and new evidence from Herculaneum suggests that Philodemus's On Piety was a
single roll of about 23 m. In general terms, however, based on both specific examples and
broad statistical data, a normative range of 3–15 meters seems reasonably certain for the
ancient bookroll (Johnson 2004, 143–152).

The diameter and heft of the rolled-up bookroll could itself be an iconographic feature of
design. At least by the Hellenistic era, the poetic bookroll was typically slim in diameter, a
fact made programmatic and famous by Callimachus. Book (p. 265) divisions in the an­
cient novel, too, seem to have accommodated to this thin and lightweight model. Many
historical texts, on the other hand, were different, probably in every period and certainly
from the Hellenistic era on: a much longer roll, thus thicker and heavier in the hand,
“monumental” in aspect, as well as content. (A book of poetry hardly ever contained more

Page 8 of 26
The Ancient Book

than sixteen hundred lines and rarely much more than a thousand; a book of history could
contain the equivalent of well more than four thousand.)

Note that there is no “standard” or “average” length to which bookrolls tended, as some
earlier researchers supposed. Since the roll end was determined by the stroke of a knife,
there was no incentive to fill out the contents; to extend the book, the scribe simply glued
on another blank roll. Note also how extreme the upper range is. It is hard for us to imag­
ine routinely managing a “book” the size of a wine bottle (for a bookroll of ten meters) or
a supersized Coke (fifteen meters), yet that, clearly, was what happened. This detail of vo­
luminology corresponds with social attitudes toward the use of the bookroll, including, in
particular, the use of lectors—yet another aspect of the elitism often associated with the
ancient bookroll.

The Codex
Codex books—that is, the style of “book” we use today—come firmly into the historical
record in the first century CE (Martial 1.2; 14.184–192; Roberts and Skeat 1983, 24–29;
Harris 1991) and over the next three centuries came to supersede the bookroll for liter­
ary texts. Bookrolls were almost exclusively made of papyrus in every period (there is on­
ly scattered evidence for leather rolls), but parchment quickly came to be the usual mate­
rial for the codex.7 Early codices were nonetheless often made with papyrus and deserve
brief remark here.

The first thing to notice is that for codices, too, one began with the papyrus roll, not (as
we moderns might expect) with sheets. To produce a codex required sheets of the same
height but twice the width of the desired page since the sheet was folded in the center to
make up the leaves of the quire.8 (For details of quire construction see chapter 25.) The
roll, as indicated, was the standard way that papyrus was sold, and thus for an ancient
Greek or Roman the natural procedure was to fashion the leaves by cutting of sheets of
appropriate size from a manufactured roll. There is some contrary evidence: For a very
few codices, it appears that sheets may have been manufactured to a special, large for­
mat so as to accommodate the codex manufacture. From what we can tell, however, that
appears to be a rarity. In papyrus codices we commonly see the telltale glue join (kollêsis)
that signals a roll as the ultimate source for the sheets that make up the quires. This is
(p. 266) not to say that scribes were unaware of the point of kollêsis: The desire to avoid

glue joins (or at least to avoid more than one join per sheet) contributed to the narrow­
ness of the leaf in papyrus codices as compared to the relative squareness characteristic
of codices made from parchment (Turner 1977, 49–51).

In the writing, scribes of Greek and Latin codices generally follow the same rules of word
division, punctuation, and the like as for the bookroll. One might also expect the earliest
papyrus codices to mimic the look of the roll, with (for prose) tall, narrow columns. This
presumption has been so strong that an earlier generation of scholars thought to see a
developmental pattern whereby codices with multiple columns were assumed to be early
in date. Accumulation of evidence (and cogent analysis by Eric Turner) has shown the
Page 9 of 26
The Ancient Book

contrary, however: Papyrus codices in every period are mostly written in single columns,
with only rare multiple-column examples—and those examples distributed over time
(Turner 1977, 35–37).9 From the earliest witnesses, that is, the codex—even while keep­
ing in line with the conservative traditions of writing—seems to exhibit a somewhat differ­
ent aesthetic in its pages. The basic look and feel are noticeably distinct from the roll
since the individual (double) pages confront the reader rather than a continuous stream
of written columns.

A different history accompanies this dissimilarity in aesthetic. The facts are these: (1) In
the second century, when codices appear in any numbers, bookrolls still account for more
than 90 percent of surviving books; by the fourth century, codices account for 80 percent
of the total; by the sixth century, the changeover is complete (Leuven Database of Ancient
Books; hereafter LDAB).10 (2) Early codices (from the second or third century) in the main
are more likely than bookrolls to be written in workaday hands (labeled “reformed docu­
mentary” in Roberts [1979, 12–15]; cf. Turner [1977, 36–37, 86]). Calligraphic and pre­
tentious scripts are a rarity rather than the substantial percentage (ca. 30 percent) al­
ready noted for bookrolls. (3) Christian texts are almost always written in codex form. On­
ly five of one hundred New Testament papyrus fragments listed in the LDAB are written
on bookrolls, and Christian writings in the broader sense tend strongly to favor the codex
form (in excess of 80 percent of all examples). Conversely, only a small percentage of
classical texts are written in codex form in the early period (second or third century); pa­
gan texts written in codex form come into their own only from the fourth century on. (4)
Coincident to the changeover from roll to codex is a shift in book content from classical
literature to Christian texts. Only a tiny percentage of surviving books are Christian in the
second century; perhaps 10 percent in the third; about 40 percent in the fourth; and more
than 50 percent by the fifth (LDAB; including codices on parchment). (5) Also coincident
to the transition to the codex is a change of material from papyrus to parchment. As early
as the fourth century, a quarter of the surviving witnesses are parchment; by the seventh,
parchment predominates.

From these facts flow historical supposition. Christians appear to be instrumental

(p. 267)

in the adoption of the codex. In 1954 C. H. Roberts speculated that the codex format may
reflect a desire to distinguish the Gospels iconographically from the rolls associated with
Jewish scripture; T. C. Skeat (with support from an elderly Roberts [1983]) drew parallels
to the use of tablets for writing the Jewish Mishnah and supposed that a “proto-Gospel”
with the sayings and passion of Jesus may have been the source of Christian fondness for
the codex form; H. Gamble (1995) has hypothesized as reason for the preference an early
collection of Pauline epistles in codex form. Others have highlighted the ability of the
codex to embrace in one package the whole of the developing Christian canon (probably
wrongly: Early codices were not normally so bulky, and the Gospels continued to circulate
individually). Scholars have also been quick to associate the workaday character of early
codices with the working-class Christian community. However all this may be, it seems
clear enough that up to the end of the second century the bookroll constituted the endur­
ing image of what a classical literary text was meant to be; that early Christians deliber­
ately adopted the different codex format for their scriptures; and that from the second
Page 10 of 26
The Ancient Book

through the fourth centuries, as Christianity grew more accepted and influential and as
the codex format became better known and appreciated, the codex gradually supplanted
the roll as the essential idea of the “book.”

The shift in technology from roll to codex had important practical results. The codex had
the following essential and consequential advantages: (1) Compass and compactness:
Several bookrolls could be fashioned into a single codex (Turner [1977, 82] presents
some statistics). The ability to collect corpora into one cover would have great conse­
quence for the use and survival of classical literature. (2) Durability: The covers helped
protect the content, and the codex was less prone to squashing, tearing, and other dam­
age. (The constant rolling and unrolling of papyrus rolls led to frequent tears and other
deterioration, as repairs on papyrus bookrolls attest.) The concomitant change to parch­
ment results in a book of unprecedented durability, again a critical point for the survival
of classical texts. Frequently cited as advantages of the codex (but probably red herrings)
are also (3) ease of access and reference (pace Harris 1991, the assumption that a reader
could mark and locate a passage more readily is based on exaggerated modern notions of
the difficulty of using a bookroll) and (4) economy of material (less material was needed
since both sides were used, but the use of ample margins, for instance, makes it unlikely
that this was an important ancient consideration). In any case, the codex also had its dis­
advantages, principally that book production was more complicated since it involved the
planning of the quires and the sewing of a binding and covers. Without implying direct
cause and effect, we can see that medieval characteristics such as the rise of scriptoria,
renewed encyclopedism, and the habit of extensive marginal annotation can be located
within the series of changes we associate with the transition to codex form.

(p. 268) Books and Their Content

Fragments of more than 3,000 bookrolls, 1,000 papyrus codices, and another 1,000
parchment codices survive from antiquity (LDAB). As we expect, these books reflect a
great diversity of content. Technical treatises of all types survive, including medical, as­
tronomical, mathematical, musical, philosophical, and grammatical discourses; commen­
taries, lexica, mythographica, and other paraliterary texts also survive in “book” form (cf.,
e.g., C.Pap.Med., C.Pap.Fil., C.Gramm., C.Comm.Lex., and CPP; Rossum-Steenbeek 1998).
But the great bulk of what a bookroll or literary codex comprises is not only literary—
that, of course, is a consequence of the modern categorization—but high literary and, as
we have seen, in a distinct, recognizable format.

I tabulate in table 11.2 the prevailing authors on papyrus bookrolls, about a third of what
survives. I do not differentiate by date or provenience, but the huge majority are of the
Roman era, and a disproportionate number are—as always for the literary papyri—from
Oxyrhynchus. The most common authors are in aggregate the direct or indirect result of
scholastic influence. We do not need the papyri to tell us of the great importance of
Homer in Greek education and culture, but they certainly do so: To the hundreds pub­
lished, we should add several hundred unpublished fragments of the Iliad and the

Page 11 of 26
The Ancient Book

Odyssey that languish in collections for the simple reason that there are many more sig­
nificant papyri awaiting editorial attention (see chapter 27). Dozens of the Homeric pa­
pyri show the telltale marks of students as they practiced accentuation and scansion;
Homer in any case is the text we most expect to find in a household interested in main­
taining a sense of Greek identity. Hesiod is a similar case, though not nearly so common
as Homer—perhaps something like finding a copy of Shakespeare next to a Bible in an
English household. The remainder of the authors at the top of the list are determined by
preference in the schools. Most of these do not surprise us: Demosthenes, Isocrates, and
indeed Thucydides and (differently) Menander are what one expects as a result of the
needs or tastes of rhetorical schools (especially of the Roman era); similarly for Plato and
the philosophical schools. Euripides' overwhelming popularity is also at least in part of
the schools' making: Lines by Euripides are common as writing practice on materials like
ostraca, tablets, and cut sheets of papyrus. But there are unanticipated results as well.
We would not, I think, have predicted the relative scarcity of Aristotle, Sophocles, or even
Aeschylus.11

Significant is that what does survive is, in the main, pretty demanding. The canonical
texts that survive in quantity were themselves not easy fare for Hellenistic- or Roman-era
readers. When we think of authors that survive mostly or only through papyri, the list is
decidedly highbrow: Sappho, Alcaeus, Alcman, Bacchylides, Hyperides, Herondas,
Philodemus, Posidippus, and Timotheus. In contrast, the lighter literature of antiquity—
the so-called ancient novel, for instance—is evident but relatively rare (Stephens and Win­
kler 1995). Uncontroversial but worth (p. 269) repeating is the conclusion that the papyri
give no evidence for mass readership of books. Rather, the ancient book is—not always,
but in general—a product to be associated with the intellectual and social elite. These
books are best situated within the general context of Greeks in a non-Greek land working
to maintain their sense of Hellenic identity. In the high empire, where our papyrological
evidence is most dense, the Graeco-Roman elite favored highbrow texts for personal at­
tainment and recreation, as we see, for example, in the pages of Lucian, Gellius, and
Athenaeus. The papyrological record is—again in broad stroke—a clear reflection of this
kind of elite tendency.

Table 11.2. Common Authors in the Papyri (LADB)*

Author Bookrolls Papyrus codices

Homer-Iliad 363 (+100?) 95

Demosthenes 115 (+8?) 24

Homer-Odyssey 93 (+26?) 24

Euripides 84 (+27?) 22 (+1?)

Page 12 of 26
The Ancient Book

Hesiod 76 (+26?) 10 (+3?)

Plato 64 (+12?) 4

Isocrates 61 (+9?) 14 (+1?)

Thucydides 56 (+15?) 2

Callimachus 45 (+9?) 16

Menander 43 (+34?) 14 (+4?)

Herodotus 33 (+5?) 0

Xenophon 30 (+3?) 2

Aeschines 28 (+13?) 3

Aeschylus 25 (+4?) (note: 22 of these are a matched 0

set)

Apollonius Rhodius 23 (+6?) 5

Pindar 20 (+11?) 4 (+1?)

Sophocles 18 (+7?) 6

Aristophanes 16 (+7?) 26

Sappho 14 (+6?) 1?

Theocritus 11 (+4?) 5

Hippocrates 11 (+3?) 5

Archilochus 10 (+7?) 0

Aristotle 9 (+1?) 1

(*) Question mark indicates uncertainty about either attribution of author or book
form.

Page 13 of 26
The Ancient Book

Books and Society: the “Scholars” of

(p. 270)

Oxyrhynchus
Much of the importance and wonder of papyrology is the ability to reach through the win­
dow of time to grasp details of the ancient world otherwise lost to us and to write through
individual witnesses the sort of “thick” history that is more fully descriptive of ancient so­
ciety. As we turn to questions of ownership and use of the papyri, we must steer away
from generalities to address a specific time and place. I choose as a case study the sec­
ond century and the evidence for a group of “scholars” in Oxyrhynchus. This “scholarly”
group is more broadly representative of Graeco-Roman elite society than one might sup­
pose and serves as an illustration of how the papyri can lead to a fresh understanding of
the use of books and indeed of the very nature of reading in antiquity.

I begin with a well-known example. P.Oxy. XVIII 2192, a letter excavated in Oxyrhynchus
and thus presumably (though not certainly) a letter sent to that city, is assigned to the
second century of our era on the basis of the scripts. The letter is written in at least three
(more probably four) hands. The body of the letter, mostly lost, is written in the flowing
semiliterary hand of a practiced scribe. Following the body of the letter, as often with an­
cient letters, the sender adds a subscription in his own hand (“I hope you're well, my dear
brother”). The same hand, that of the sender, adds a substantial postscript underneath
the subscription. A second postscript, also substantial, follows the first in what is clearly a
third hand (which we will call that of the sender's colleague). Following that the papyrus
breaks off, but there are remains of what appears to be yet a third postscript, in what
seems to be a fourth hand (apparently, a second colleague).

The postscripts deserve our close attention, for they combine to convey a vivid impres­
sion of one side of “scholarly” activity in second-century Oxyrhynchus. The first post­
script, in the sender's hand, writes: “Have a copy made of books six and seven of Hypsi­
crates' Men Who Appear in Comedies and send it to me. Harpocration says that Pollio has
them among his books, and probably others have it too. And he also has epitomes in prose
of Thersagoras's Myths of Tragedy.”

A different hand—the sender's colleague—then adds another paragraph in postscript:

“Demetrius the bookseller has them (that is, the two books of the Men Who Appear in
Comedies), according to Harpocration. I have ordered Apollonides to send to me some of
my own books—which ones you'll find out from him. And if you find any volumes of
Seleucus's work on Tenses/Metrics/Rhythms that I don't own, have copies made and send
them to me. Diodorus's circle also has some that I don't own.” (Text as in Hatazilambrou
2007.)

Not every point is clear, but a number of inferences can be reasonably drawn. First,
though the books are personally owned (“some of my own books,” writes the (p. 271) col­
league), we garner a strong impression of a group that shares books—and not just any
books but those that are learned compilations of background information for classical
drama, rather like a scholar's accumulation of reference works today. We sense the vigor
Page 14 of 26
The Ancient Book

with which this group of people pursues the collecting of such books and the serious in­
terest in having some rather arcane knowledge available. Second, the sender's colleague
seems to assume that the recipient will both recognize him by his hand (though it is possi­
ble that the lost beginning or final subscription to the letter identifies him) and know all
about Seleucus as well as the contents of his own library (“collect the volumes that I don't
already own”). The focus in the letter on reference books for tragedy and comedy seems
to imply something akin to a scholarly project, or at least a circle of readers with focused
attention for the moment on, it appears, classical drama. The colleague in any case seems
eager to acquire all available books within a defined area of knowledge. The ease with
which he orders the recipient to collect “whatever volumes [of Seleucus] I don't own” is
also telling: We gain the impression of a man of some means. Third, the tightness of the
network merits comment. Harpocration (whom Eric Turner [1952, 92] identified with the
Alexandrian lexicographer)12 appears to be the central node of the personal network, but
both the sender and the sender's colleague give independent suggestions for either alter­
nate books to be obtained or alternate sources for the books. One has the vivid sense of
both a small, friendly community of those interested in books of this sort and of close con­
nections between Oxyrhynchus and whatever town the request derives from. The alter­
nate source suggested by the sender's colleague—oἱ πεpὶ Διόδωρ[ον]—leads to our final
point, which is that this circle of book collectors and readers knows of another circle that
seems in some sense an analogue to themselves—“Diodorus's circle”13 is evidently also a
group interested in (and perhaps defined by) its interest in books of a learned sort. One
wonders, though it cannot be proven, whether those in “Diodorus's circle” might well
have referred to these two writers and their compeers as “Harpocration's circle.”

The details in this letter are echoed by other documents of similar period from Egypt.
From second-century Alexandria, Theon (again identified by Turner as the grammarian of
same name; Turner [1952, 93]) writes a letter to Heraclides in Oxyrhynchus, accompa­
nied by six Stoic texts on self-improvement (P.Mil.Vogl. I 11). In elevated (indeed
pompous) Greek, Theon writes, “Inasmuch as I take great pains to furnish you with useful
books, especially those that contribute toward a better life, I think it behooves you not to
be careless in the reading of them since the serviceableness that comes from these books
is not trifling for those who take pains to acquire the advantage.” Theon continues: “I am
sending the books you requested via Achilles,” and after the subscription he adds a list of
the books. Another, very lacunose, second-century letter, SB XIV 11996 (= Corp.Pap.Fil.
5), contains similar verbiage about sending books, apparently among friends, via a couri­
er. Written in the second century and (probably) from the Fayyum, P.Lit. (p. 272) Lond. 97,
a prose farce, has a note on the back that “Heraclides made the copy from the library of
Praxias.”

Several book lists (usually of somewhat obscure import) come down from antiquity, but at
least a few from the second or third century CE are clearly either lists of desiderata or
catalogues of existing bookrolls, and—of interest here—these book lists are usually rather
abstruse materials like rare works, philosophical works, and commentaries.14 Yet another
second-century letter (P.Petaus 30), this time from Ptolemais Hormou, is written by a son
to his father and reads: “When Deios was with us, he showed us the six parchments
Page 15 of 26
The Ancient Book

(μεμβράνας). We did not take any of these, but we made collations against eight others,
for which I paid 100 drachmas on account.” Apparently a traveling scribe or book dealer
of some sort offers “parchments” (meaning parchment rolls? codices?) for sale but as an
ancillary service provides an opportunity to collate texts in his collection. These various
papyrus documents, though a thin stream, help to make vivid a second-century scene in
which a small community of literate men in the regional capitals and perhaps also the
towns of provincial Egypt (cf. van Minnen 1998) were actively searching for and sharing
literary texts, commentaries, and works of reference, seeking to improve both the text
and their knowledge of it.

With this understanding as background, let us return to P.Oxy. 2192, the first text we ex­
amined. What is particularly striking about this letter is the fact that the scholars pursue
their interests as a group and that what the letter seems to imply is not so much the indi­
vidual scholar busy with his studies but a circle of readers with scholarly interests and
one with contacts in Oxyrhynchus, which, as the metropolis, has similar readers' circles,
along with other resources of interest to “scholarly” readers.

This suggestion of scholarly activity as a group undertaking dovetails interestingly with

another set of documentary evidence from the sands of Oxyrhynchus, namely, the actual
artifacts of scholars' texts themselves—for, to anticipate, the marginal annotations in pa­
pyrus bookrolls that belong to scholars are also often the product of multiple people's
work. I focus on a group of texts already collected as “scholar's texts” by a couple of earli­
er researchers (Turner 1956;McNamee 1981, 2007). These are annotated texts that con­
tain textual remarks and variants and in which the textual variants are attributed to a
person by name; that is, the marginal annotations read, for instance, “Thus was [the read­
ing] in Theon's”; “So in Apion's”; and so forth. The interest in attribution—in applying a
name to the variant reading—seems to mark an enterprise or an attitude toward the text
that is fully consistent with our notion of a “scholars' text.” The challenge is to define bet­
ter the social context for these texts.

The number of such texts is not large, but it is also not small: nineteen in all. In the re­
marks that follow, I will restrict my remarks to the sixteen Oxyrhynchite texts that form
the bulk of the evidencei15 since I wish to explore the possibility that these texts consti­
tute a group of some sort. The Oxyrhynchite texts share a surprising number of uniformi­
ties:

(1) Almost all the bookrolls comprise noncanonical texts, and most are diffi­
(p. 273)

cult works to read. Aside from P.Oxy. 445 (the Iliad), the texts fall, moreover, into two
distinct types. First are a number of dramatic works, most of which are unusual
plays: four plays by Sophocles (Ichneutai, Eurypylos, Trachiniai, Theseus) and sever­
al plays (it appears) by Epicharmus; all of these (except, of course, the bookroll of
the Trachiniai) are unique surviving witnesses to their texts. Second are a number
oflyric and elegiac texts that make up several bookrolls of Pindar, a couple of Stesi­
chorus, a couple of Alcaeus, and one each of Alcman, Bacchylides, and Simonides.

Page 16 of 26
The Ancient Book

(2) Rarely is the scribe involved in the annotation of textual variants, and when he is,
there are always other annotators as well.
(3) It is unusual, however, to find only the scribe of the text and a single annotator,
though that does happen three times. In thirteen of the sixteen examples, multiple
annotators are in play, often as many as four, five, or six. Note that for many of these
texts, the very fragmentary state of preservation tends to understate the number of
annotators since only a few margins, at most, are likely to survive.
(4) Most of the texts (ten of the sixteen) show signs of substantial additions of read­
ers' marks (such as added punctuation, word division, or elision markers), and al­
most all have at least some readers' marks (as is generally common in bookrolls, es­
pecially for difficult texts). Often multiple readers are involved.
(5) Most of the texts are also marked up with chi or other sigla in the left margin
(again, ten of the sixteen, even though in some cases very little of the left margin
survives). Such sigla have sometimes been taken as keys to commentaries, but, to
follow McNamee (1992, 19–22), these are best understood as a variety of ways of sig­
naling nota bene; that is, they are marks by readers that signal passages of interest
or those that need further attention.
(6) A common remark by the editors of these texts is that the annotator's hand is
“contemporary with the text,” and my review of these manuscripts in photograph
confirms that very few (no more than a couple) of the annotators here write with
hands that are disconsonant with the time period of the scribal hand. Certainly
palaeography is a crude measure of date, but it is interesting that the fact of multiple
annotators, whatever the exact social context, does not seem to be the product of a
book's passing from hand to hand over numerous generations.
(7) Two of the Sophocles bookrolls are written by the same scribe, and four others
are also by scribes who are identified as having written other bookrolls among the
Oxyrhynchus publications.16 The fact that six of the sixteen texts here are written by
identified scribes must be considered against the fact that, among many hundreds of
published literary texts from Oxyrhynchus, only forty-ive scribes in total have been
identified as having written multiple bookrolls (Johnson 2004, 61–65). It is hard to
know what exactly to make of the correspondence, but a reasonable inference is that
the coincidence of scribal hands is the consequence of the disposal in the Oxyrhyn­
chite town dump of a set of bookrolls by a single owner (p. 274) or group (in the man­
ner of the Hypsipyle archive; Cockle [1987, 21–22]; and see also Houston [2009]). If
that is the case, then of interest is the fact that the other bookrolls by these identi­
fied scribes are also predominantly bookrolls of tragedy, lyric, and elegiac.

I do not mean to overstate the uniformities since there is certainly a great deal of varia­
tion in detail, such as the fact that a number of bookrolls are freely annotated by a num­
ber of hands, whereas others seem principally annotated by one hand with occasional ad­
ditions by others. Yet for all that the uniformities remain remarkable. If I am seeing this
at all clearly, we seem to be dealing not just with a random collection of examples but al­
so with a group that in use and function represents a type. That is, the use and function of
the manuscripts seem to reflect some rather specific sociocultural context that prevailed

Page 17 of 26
The Ancient Book

in second-century Oxyrhynchus. By that I do not mean to suggest that the use of the man­
uscripts can be localized to a particular set of contemporaries—the fact that the manu­
scripts are assigned all along the spectrum of more than a century defeats that hypothe­
sis. Nor do I wish to draw too tight a line between the letter we examined earlier (P.Oxy.
2192), in which a group of scholars are collecting background data for the study of
tragedy, and the prominence of bookrolls of tragedy in the group here. Nonetheless, I do
find of great interest the fact that the group centers on a couple of genres in a manner
similar to the “scholarly project” implied in P. Oxy. 2192. In general terms, what seems to
me so fascinating about this group of manuscripts is what they may suggest about schol­
arly habitudes, that is, what sociocultural models of behavior we can find that account for
the ways in which these papyri seem to have been used.

Noteworthy at once is that two standard models do not seem adequately to account for
the consistencies here noted:

(1) The diorthôtês model. In this model one expects that once the scribe is finished
copying, the bookroll is handed over to a “corrector” (the diorthôtês), who collates
the text for errors—whether this happens at the instigation and under the control of
the scribal shop or the owner is immaterial. This model does not seem adequately to
account either for the style of annotation, in which the attributions are carefully
named, or for the number of hands involved in annotating the text with variant read­
ings and other “learned” data.
(2) The inheritance model. We know of bookrolls in antiquity that were passed along
with an estate (Bagnall 1992; Clarysse 1983), which is hardly a surprise given the
value of the artifact and, for exotic texts, the rareness of the product. In that case,
the reuse of the bookroll from generation to generation might account for the pres­
ence of multiple annotators. But that model, which is frankly what I expected to find,
does not obtain here. The hands for any given manuscript are too contemporary (and
often too numerous) to allow it, whether by inheritance we mean from relative to rel­
ative (as, for example, parent to child) or from master to pupil.

Moreover, the list presents some surprises of detail. I draw attention in particular to
Oxyrhynchus papyrus 2387 (Alcman's Partheneia). In the intercolumn to P.Oxy. (p. 275)
2387 we find the usual syntax of attribution: μό(νον) Π(τολεμαίον) [fr. 3.ii.19: “only in
Ptolemy's”], οὕ(τως) Π(τολεμαίον) [3.ii.22, “thus in Ptolemy's”]. Here, as always, the ref­
erent of the possessive is not entirely clear: Ptolemy's what? His commentary? His copy?
In this case, however, we have more information, for the upper margin (2387, fr. 1) con­

tains an extremely interesting note: π]αρενγρά (ϕεται)

(“This [passage] is wrongly inserted in copies [also in the] fifth (book), and in that book it
was bracketed [in] Aristonicus's copy but was left unbracketed in Ptolemy's”). That is, the
writer of the note claims not only that the passage is generally interpolated into copies of
book five but also that, in two specific antigraphs (those belonging to Aristonicus and
Ptolemy respectively), the interpolation is variously treated. Throughout the list of papyri
with attributed annotations, as mentioned, one is confronted with the question of the ref­

Page 18 of 26
The Ancient Book

erent for the possessive: “Thus it was in Theon's”; “Thus only in Ar( )'s” (whether the ab­
breviation “Ar” means Aristophanes or Aristarchus or someone else). Where the name
may be taken to match that of a known grammarian, the impulse has been to take the
possessive with the notion of an accompanying hypomnema (commentary): “Thus the
reading was in Aristophanes' (or Aristarchus's or Theon's) commentary on Sophocles.”
And I suppose that this resolution is correct for at least some of these manuscripts.
Nonetheless, in the case of P.Oxy. 2387, it seems, it is not Aristonicus's and Ptolemy's
commentaries that are being collated but rather the copies that belong to these two peo­
ple. Moreover, Aristonicus is very likely the Augustan-age grammarian of that name.17

In any case, the fact that the annotation in P.Oxy. 2387 witnesses a group of readers with
access, whether directly or at a remove, to the unique Alcman copies used by Aristonicus
and Ptolemy accords with other evidence in our list. For instance, in P.Oxy. 2452, the vari­
ant reading is attributed not simply to a scholar whose name begins Ari (again, the reso­
lution of the abbreviation is uncertain), but the variant is specifically said to be ἐν ἑτέρῳ
ʼAρι( ) (“in the second copy, that belonging to Ari(?)”). In several other cases, too, attrib­
uted variant readings are accompanied by remarks like “in the other (copy),” “n the first
(copy),” and “in the antigraph,” which makes clear that at least some of the general activ­
ity we seek to tease out of this body of evidence is certainly collation against actual
copies as opposed to collation and annotation on the basis of commentaries. The heavy
use of sigla like chi and ζή(τει) likewise helps to flesh out the picture since these are
notes (always placed at the left of the column of text) that mark something that needs to
be examined further or specially noted. That is, they are the skeletal markings of an ac­
tive investigation into points of detail in the text.

What image then can we construct of the use to which these texts are put? In attempting
to answer that question, I want to consider, by way of illustration, one last piece of evi­
dence, again from the second century (ca. 170 CE) but this time from the literary tradi­
tion and—though written by a man who spent his later life in (p. 276) Egypt—a text most
likely written in Athens. The text is the adversus Indoctum of Lucian, a diatribe that in at­
tacking an “uneducated” Syrian and his circle makes clear by contrast how an intellectual
circle should work. The Syrian does everything wrong: He collects expensive antique
books but does not know how to judge their textual value (Ind. 1); he does not know how
to read the texts aloud so as to bring out the meaning (Ind. 2); he is not well educated in
things literary (Ind. 3, cf. 26–27, where he is ridiculed for his ignorance of Plato, Antis­
thenes, Archilochus, Hipponax, Aeschines, and Eupolis); he collects a circle of friends
who are not men of culture but Xatterers or worse (Ind. 7; cf. 17, 20); his sham attempts
at elite intellectual display and discussion thus devolve into drunken parties (Ind. 22–23).
Several points of intense interest emerge from the scene Lucian paints: (1) The context
for the use of high literary books is an intellectual circle; (2) for a true intellectual circle,
elite education is assumed not just for the central figure but for the group as well; (3) ed­
ucation assumes (a) an ability to read performatively, with detailed, deep knowledge of
the meaning, style, structure, and conventions, (b) a broad knowledge of rather arcane
persons and works in the literary tradition, and (c) strikingly, the ability to judge the val­
ue and correctness of the text of a bookroll. Note, however, that Lucian does not present
Page 19 of 26
The Ancient Book

the Syrian as someone aspiring to be a leading scholastic or scholar. Rather, the Syrian
seems simply to aspire to be among the elite but in a time period when the elite (under
the “scholar emperor” Aurelius; cf. Ind. 22) particularly valued learning and when leading
elites sought to achieve a strong sense of intellectual refinement among their ranks.

The portrait in Lucian interlocks with the papyrological evidence in some intriguing ways.
For instance, readers' marks in the papyri (such as added punctuation) may of course be
successive attempts to make sense of the text, but the prevalence of multiple readers'
marks may also indicate that members of these intellectual groups repeatedly made per­
formative readings of the text whether by way of entertainment or as a springboard for
discussion. The annotations in the papyri of variant readings and other details from
named antigraphs of previous scholars are in keeping with Lucian's hints of a strong in­
terest in antique, autograph texts. A general interest in the correctness and the antiquity
of the text and, importantly, the vesting of the judgment of the value of a particular copy
of a text in the hands of “the educated” also seems quintessential to attitudes of the time.
Most important, though, is the link between what we find in Lucian and what the multiple
hands of the annotators and the evidence in papyrus letters seem to suggest about a so­
cial context for the creation of a “scholarly” product.

The combination of evidence does not, that is, seem to sit well with the dominant me­
dieval and modern image of a scholar quietly sitting in his study, reading a difficult text
together with a commentary and transferring notes from that commentary into the mar­
gins of his bookroll or collating a text side by side with its exemplar. Nor do the mechan­
ics of reading in a precodex culture encourage that type of behavior, given the (p. 277)
awkwardness of opening more than one bookroll at a time. Rather, the presence of textu­
al variants in multiple hands seems likely to be, in some sense, the result of repeated
group discussion and analysis of the text. As we have seen, that group work would itself
include a variety of activities, including, of course, not only the reading of hypomnemata
but also, it appears, collation against other copies of the text, especially any available
copies of particular value; the reading of works that provided background or ancillary in­
formation, such as the Men Who Appear in Comedy and Myths in Tragedy, volumes
sought from among Pollio's books in papyrus letter P.Oxy. 2192; and no doubt (to infer
from Lucian) vigorous discussion of points of style, structure, convention, as well as (to
infer from the papyrus annotations) discussion of the constitution of the text. This “schol­
arly” work may well have typically revolved around a central figure—think of Diodorus's
circle and the references to Harpocration in that same Oxyrhynchus letter—but the scene
is by no means necessarily professional or scholastic: The other men in the cubiculum and
at dinner need hardly be “scholars” in any conventional sense.18 Lucian and his Syrian re­
mind us that the “scholarly” enterprise was, in the second century, at least as closely
linked to a particular type of cultural elitism and indeed exclusionism, focused on Greek
letters, as to professional “scholarship,” and that those engaged in the criticism, interpre­
tation, and constitution of canonical Greek texts probably went well beyond the group of
grammarians and scholars usually cited.

Page 20 of 26
The Ancient Book

Bibliography

Abbreviations

CPP
= Catalogue of Paraliterary Papyri, ed. Marc Huys. http://cpp.arts.kuleuven.be/.
LDAB
= Leuven Database of Ancient Books, ed. W. Clarysse. http://ldab.kuleuven.be/.
Der neue Pauly
= = Der neue Pauly: Enzyklopädie der Antike, ed. C. von Hubert and H. Schneider.
Stuttgart: J. B. Metzler, 1996–2003.
.

Bagnall, R. S. 1992. “An Owner of Literary Papyri.” Classical Philology 87: 137–140.

Bell, H. I. 1921. “The ‘Thyestes’ of Sophocles and an Egyptian Scriptorium.” Aegyptus 2:

281–288.

Capasso, M. (1995). Volumen: Aspetti della tipologia del rotolo librario antico. Naples:
Procaccini.

Clarysse, W. 1976. “Notes on the Use of the Iota Adscript in the Third Century BC.” CdÉ
51: 150–166.

———. 1983. “Literary Papyri in Documentary ‘Archives.’” In Egypt and the Hellenistic
World: Proceedings of the International Colloquium, Leuven, 24–26 May 1982, ed. E.
Vanʼt Dack, P. van Dessel, and W. van Gucht, 43–61. Studia Hellenistica 27. Leuven: Orien­
taliste.

Cockle, W. E. H. (1987). Hypsipyle, Euripides: Text and Annotation Based on a Re-exami­

nation of the Papyri. Rome: Ed. dellʼAteneo.

Coles, R. A., M. Manfredi, P. J. Sijpesteijn, and A. S. Brown. (1985). The Harris Papyri, vol.
2. Zutphen, the Netherlands: Terra.

Fraser, P. M. (1972). Ptolemaic Alexandria. Oxford: Clarendon.

Gamble, H. (1995). Books and Readers in the Early Church: A History of Early Christian
Texts. New Haven, Conn.: Yale University Press.

Harris, W. V. 1991. “Why Did the Codex Supplant the Book-Roll?” In Renaissance
(p. 280)

Society and Culture: Essays in Honor of Eugene F. Rice Jr., ed. J. Monfasani and R. G.
Musto, 71–85. New York: Italica.

Hatzilambrou, R. 2007. “P. Oxy. XVIII 2192 revisited.” In Oxyrhynchus: A City and its
Texts, ed. A. K. Bowman et al. 282–286. Graeco-Roman Memoirs 93. London: Egypt Explo­
ration Society.

Page 21 of 26
The Ancient Book

Houston, G. 2009. “Papyrological Evidence for Book Collections and Libraries in the Ro­
man Empire.” In Ancient Literacies: The Culture of Reading in Greece and Rome, ed. W.
A. Johnson and H. Parker. New York: Oxford University Press.

Johnson, W. A. 1993. “Pliny the Elder and Standardized Roll Heights in the Manufacture
of Papyrus.” Classical Philology 88: 46–50.

———. 1994. “The Function of the Paragraphus in Greek Literary Prose Texts.” ZPE 100:
65–68.

———. (2004). Bookrolls and Scribes in Oxyrhynchus. Toronto: University of Toronto

Press.

Lewis, N. 1963. “The Non-scholar Members of the Alexandrian Museum.” Mnemosyne

(ser. 4)16: 257–261.

———. (1974). Papyrus in Classical Antiquity. Oxford: Clarendon.

———. 1981. “Literati in the Service of the Roman Emperors: Politics before Culture.” In
Coins, Culture, and History in the Ancient World: Numismatic and Other Studies in Honor
of Bluma L. Trell, ed. L. Casson and M. Price, 149–166. Detroit: Wayne State University
Press.

———. (1995). On Government and Law in Roman Egypt: Collected Papers of Naphtali
Lewis. Atlanta: Scholars Press.

Marshall, P. 1976. “Library Resources and Creative Writing at Rome.” Phoenix 30: 252–
264.

McNamee, K. 1981. “Greek Literary Papyri Revised by Two or More Hands.” In Proc. XVI
Int. Congr. Pap., ed. R. S. Bagnall et al., 79–91. Chico, Calif.: Scholars Press.

———. (1992). Sigla and Select Marginalia in Greek Literary Papyri (Pap. Brux. 26). Brus­
sels: Fondation égyptologique Reine Élisabeth.

———. (2007). Annotations in Greek and Latin Texts from Egypt. American Studies in Pa­
pyrology. Oakville, Conn.: American Society of Papyrologists.

Otranto, R. (2000). Antiche liste di libri su papiro (Sussidi eruditi 49). Rome: Edizioni di
Storia e Letteratura.

Radt, S. 1980. “Noch einmal Aischylos, Niobe Fr. 162 N. 2 (278 M.).” ZPE 38: 47–58.

———. 1997. “Zu P. Merton 19, 2 F. und P. Oxy. 2192, 43 F.” ZPE 119: 6.

Roberts, C. H. 1954. “The Codex.” Proceedings of the British Academy 40: 169–204.

———. (1979). Manuscript, Society and Belief in Early Christian Egypt. London: Oxford
University Press.

Page 22 of 26
The Ancient Book

———, and T. C. Skeat. (1983). The Birth of the Codex. London: Published for the British
Academy by Oxford University Press.

Rossum-Steenbeek, M. van. (1998). Greek Readers' Digests? Studies on a Selection of

Subliterary Papyri. Leiden: Brill.

Skeat, T. C. 1982. “The Length of the Standard Papyrus Roll and the Cost-Advantage of
the Codex.” ZPE 45: 169–175.

———. 1995. “Was Papyrus Regarded as ‘Cheap’ or ‘Expensive’ in the Ancient World?” Ae­
gyptus 75: 75–93.

Stephens, S. A., and J. J. Winkler. (1995). Ancient Greek Novels: The Fragments: Introduc­
tion, Text, Translation, and Commentary. Princeton: Princeton University Press.

Turner, E. G. 1952. “Roman Oxyrhynchus.” JEA 38: 78–93.

———. 1956. “Scribes and Scholars in Oxyrhynchus.” In Akten des VIII. Interna­
(p. 281)

tionalen Kongresses für Papyrologie, Wien 1955, ed. H. Gerstinger, 141–146. MPER, n.s.,
5. Vienna: Rudolf M. Rohrer.

———. (1977). The Typology of the Early Codex. Philadelphia: University of Pennsylvania
Press.

———. 1978. “The Terms Recto and Verso: The Anatomy of the Papyrus Roll (Pap.Brux.
16).” Actes du XVe Congrès international de papyrologie, Bruxelles-Louvain, 29 août–3
septembre 1977, pt. 1, ed. J. Bingen and G. Nachtergael. Brussels: Fondation égyp­
tologique Reine Élisabeth.

———. (1980). Greek Papyri: An Introduction, 2d ed. Oxford: Clarendon Press.

———, and P. J. Parsons. (1987). Greek Manuscripts of the Ancient World, 2d ed. London:
University of London, Institute of Classical Studies.

Van Minnen, P. 1998. “Boorish or Bookish? Literature in Egyptian Villages in the Fayum in
the Graeco-Roman Period.” Journal of Juristic Papyrology 28: 99–184.

Willis, W. H. 1968. “A Census of the Literary Papyri from Egypt.” GRBS 9: 205–241.

Notes:

(1.) Most of the substandard scripts (about 15 percent of the total) are Homeric papyri.
Some problems of editorial selection come into play here since many Homeric papyri re­
main unpublished. Percentages are derived from tables 3.1 and 3.2 in Johnson 2004, 162–
184.

(2.) Lyric texts created before the colometric researches of early Alexandrian scholars
were presumably written as prose, but these, too, were written in verse units from the
third century BCE on. See Turner and Parsons (1987, 12).
Page 23 of 26
The Ancient Book

(3.) Ancient grammarians speak of other punctuation dots (ἡ μέση στιγμή, ύποστιγμή)
that mark lesser pauses in a sentence, but of this the papyri show evidence that is spo­
radic and inconsistent—certainly not part of a generalized system used and understood
by readers (Turner and Parsons 1987, 9; Johnson 1994).

(4.) “A syllable divides after its vowel: but division is permitted between doubled conso­
nants or two consonants, the first of which was a liquid or a nasal or a sibilant; and after
a single consonant, if that letter is part of a preposition forming a compound
word” (Turner and Parsons 1987, 16–17).

(5.) Earlier scholars have seriously underestimated the typical size of margins.

(6.) Sometimes cited in this regard is a statement by Tzetzes (Proleg. de comoedia

Aristoph., Prooemium 2.10–11 Koster), ascribed to Callimachus, in which it is said that at
the time of Ptolemy III the Library of Alexandria contained 400,000 “mixed” books and
90,000 “single and unmixed,” but it is unclear whether “mixed” rolls refer to bookrolls
containing multiple works.

(7.) There is considerable debate over whether the parchment codex or the papyrus codex
came first; see Turner (1977, 35–42); Roberts and Skeat (1983, 29).

(8.) Large vellum skins and, of course, the large-format paper used for modern books can
be folded both horizontally and vertically to make up quires of four, eight, or even sixteen
sheets. The maximum height of papyrus (about thirty-five cm), as normally manufactured,
precluded horizontal folding for all but very small books.

(9.) By contrast, the double column was a favored format for parchment codices of liter­
ary prose texts. Turner (1977, 35) has speculated that these later, parchment codices
were deliberate imitations of high-class bookrolls.

(10.) Statistics from the LDAB in this chapter are based on its state as of January 2006.

(11.) The sample is, of course, largely restricted to towns and villages in Egypt, but that
in effect gives a snapshot of the scene at the far reach of Hellenism—probably represen­
tative, that is, of the broad swath of the Hellenic population around the Mediterranean.
With the exception of the Homeric texts, the problem of editorial selection in the publica­
tion of literary fragments (an important consideration for earlier analyses like Willis
[1968]), is no longer a serious difficulty.

(12.) This remains only a probability, however, as does the identification of Pollio with lex­
icographer Valerius Pollio of Alexandria and of Diodorus (a common name) with Pollio's
son, a second-century Alexandrian who was a member of the Alexandrian Museum (note
that Pollio and Diodorus are not linked in the letter itself). Turner's suggestion that the
Harpocration mentioned here is the Alexandrian has now been elevated to a fact in Der
neue Pauly.

Page 24 of 26
The Ancient Book

(13.) The P.Oxy. editor (Roberts) translates this as “Diodorus and his friends”; Radt (1997,
6) thinks it “more probable” that the phrase here, as commonly enough in classical and
Hellenistic Greek, is simply a paraphrase for “Diodorus” (see LSJ s.v. πєρί, as well as the
extensive treatment of the phrase in Radt [1980, 47–56]). Nonetheless, the use of this
phrase for “Diodorus” seems characteristic of high literary rather than documentary
Greek; the examples in Preisigke's lexicon of 01 πєρὶ τὸν δєῖνa (2.288) in papyrus letters
and other documents all seem consistent with the translation “X and intimate associates,”
whether family or friends. In any case, it is the focus on the man within the context of his
(somewhat invisible) friends and followers that I seek to capture in the translation
“Diodorus's circle.”

(14.) Examples include P.Oxy. XLVII 3360, P.Ross.Georg. I 22 = Corp.Pap.Fil. 2, P.Laur. inv.
19662 = Corp.Pap.Fil. 3, P.Vars. 5 = Corp.Pap.Fil. 4, P.Turner 9, and P. Lond. inv. 2110
(Bell 1921); for a comprehensive collection of book lists on papyrus see Otranto (2000).

(15.) P.Oxy. 445 (Iliad 6), P.Oxy. 841 AB (Pindar, Paeans), P.Oxy. 1174 + 2081a + Pap.Flor.
X (Sophocles, Ichneutae), P.Oxy. 1175 + 2081b (Sophocles, Eurypylos), P.Oxy. 1234 +1360
+ 2166 (Alcaeus), P.Oxy. 1361 (Bacchylides, Scolia), P.Oxy. 1805 + 3687 (Sophocles, Tra­
chiniae), P.Oxy. 2295 (Alcaeus), P.Oxy. 2327 (Simonides), P.Oxy. 2387 (Alcman), P.Oxy.
2394 (Doric lyric, Alcman?), P.Oxy. 2427 (Epicharmus, various plays), P.Oxy. 2442 (Pin­
dar), P.Oxy. 2452 (Sophocles, Theseus), P.Oxy. 2803 (Stesichorus), P.Oxy. 3876 (Stesicho­
rus). Attributed textual annotations in papyri not from Oxyrhynchus: P.Par. 71 (Alcman,
from Memphis), P.Hawara 24–28 (Iliad 1–2, from Hawara), P.Lit.Lond. 30 (Od. 3, said to
come from Soknopaiou Nesos). List and analysis here are based in part on evidence as­
sembled in McNamee (2007), kindly shared in manuscript; cf. also McNamee (1981).

(16.) Following the numeration in table 2.1 in Johnson (2004), Scribe #B1 (Sophocles:
P.Oxy. 1174 + 2081a + Pap.Flor. X, P.Oxy. 1175), #A19 (P.Oxy. 2327), #A22 (P.Oxy. 2427),
#A24 (P.Oxy. 2452), #A30 (P.Oxy. 2442).

(17.) Or at least it seems a telling coincidence that the rare ancient notices of Aristonicus
cite both a son (Athen. 11.481d, Iliad scholia 4.423a1 Erbse) and a father (Suda P 3036)
named Ptolemy, both grammarians and teachers in Rome. McNamee (2007) points out
that, even though a given abbreviation may be uncertain, all of the attributed names list­
ed in the appendix seem plausibly to be names of known grammarians, or at least all of
the abbreviated names can be so resolved.

(18.) For analogous circles of readers in Rome attached to private libraries (hence to
prominent political and social figures), see Marshall (1976). The scene suggested here,
where the social elite in Hellenic Egypt play at being intellectuals, may help explain why
it was unnecessary for those appointed to the Alexandrian Museum in the Roman era to
be scholars per se, though all seem to be members of the social elite. (For details on mu­
seum members who were not scholars see Lewis [1963] = Lewis [1995, 94–98]; Fraser
[1972, 333–334]; Lewis [1981] = Lewis [1995, 257–274].)

Page 25 of 26
The Ancient Book

William A. Johnson

William A. Johnson, Associate Professor of Classics and Head of the Department of

Classics, University of Cincinnati.

Page 26 of 26
Papyrology and Ancient Literature

Papyrology and Ancient Literature

Timothy Renner
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0012

Abstract and Keywords

This article discusses literary and subliterary papyri; papyri and Egyptian literature; the
study of Greek literature; and papyri and Latin literature. The texts inscribed on these
materials are the source for the longest and most important Egyptian literary composi­
tions known from the Pharaonic and Hellenistic periods. “Subliterary” papyri include pa­
pyri containing texts such as commentaries, lexica, and grammatical treatises, which are
in some sense ancillary to the study of the major genres and have traditionally been so re­
garded. Hieratic and demotic papyri, including wooden writing boards and ostraca, are
responsible for our knowledge of most of the Egyptian texts that contain narrative tales
and fables, instructions or precepts, and love poetry. Meanwhile, the body of ancient
Greek literature continued to expand on the basis of papyrological evidence.

Keywords: literary papyri, Greek literature, Latin literature, Egyptian Literature, papyrological evidence

Literary and Subliterary Papyri

WHILE the vast majority of literary texts on papyrus, parchment, ostraca, or wood tablets
are written in Greek, the texts inscribed on these materials are also our source for the
longest and most important Egyptian literary compositions known from the Pharaonic and
Hellenistic periods. Further, in spite of the overriding use of Greek as both an administra­
tive language and a mark of elite culture in Egypt during the Roman period, a modest
number of papyri shed light on the use or study of Latin literary texts in that province and
on occasion make specific additions to the study of Latin literature more generally. Taken
together, the varied contributions of papyri continue to bring new energy to the study of
ancient literature and ensure that the boundaries of the texts that are the objects of its
study shall not remain finite. Because as material objects they transmit works to us di­
rectly from their world, literary papyri (although often fragmentary and in need of much
attention by modern editors to make them accessible to nonpapyrologists) bring the study
of ancient literatures and their cultures alive with special immediacy.

Page 1 of 21
Papyrology and Ancient Literature

The term literary papyri at first calls to mind texts of major genres such as epic, drama,
lyric, history, and other types of prose narrative. However, papyrologists today, particular­
ly those who study papyri written in Greek, usually include under (p. 283) “literary” sever­
al categories of “subliterary” or “paraliterary” papyri. These include papyri containing
texts such as commentaries, lexica, and grammatical treatises, which are in some sense
ancillary to the study of the major genres and have traditionally been so regarded. Fur­
ther, the term subliterary is often used to refer to other types of technical works such as
those on mathematics, science, and medicine, which in this book are treated in chapter
15.1

Papyri and Egyptian Literature

If it were not for papyri and their decipherment and publication over the past two cen­
turies, our familiarity with Egyptian literature today would be much reduced. Although
literary texts of some length have been preserved by inscriptions on tombs and stelae, hi­
eratic and demotic papyri, including wooden writing boards and ostraca, are responsible
for our knowledge of most of the Egyptian texts that contain narrative tales and fables, in­
structions or precepts, and love poetry.2 The most recent count of demotic literary
papyri,3 including some items not yet published, totals 539. The number of hieratic liter­
ary papyri, excluding funerary texts, is estimated to be substantially less than this. Al­
though their numbers are small, the role of papyri in transmitting this ancient literature
is very great by comparison with Greek or Latin literatures, which in the absence of the
important additions that the papyri bring to them would still be available, to a great ex­
tent, through the medieval tradition had papyri never been known to moderns. Another
basic difference between Graeco-Roman literary works and literary works in Egyptian is
that the latter are usually anonymous, with the exception that some are pseudepigraphi­
cally attributed to pharaohs, sages, or other figures of past eras.

Here it is possible only to mention general categories of texts and to highlight a few de­
tails. The earliest hieratic literary compositions preserved on papyrus, ostraca, or wooden
tablets are “instructions” that dispense advice on how best to live life. The earliest copies
of the texts themselves date to the Middle Kingdom, although in the case of some of the
most famous instructions the dramatic setting is the Fifth or the Sixth Dynasty. Texts com­
posed during the Middle Kingdom and attested on papyrus include not only further in­
structions but also a wider variety of poetic and prose works. There are hymns to rulers
and to the Nile. Especially, there is a wide range of prose tales such as the rather lengthy
Eloquent Peasant, which mixes seriousness and irony in a demonstration of the value of
rhetoric, or the Story of Sinuhe, detailing the life of a courtier sojourning in foreign lands.
The Story of the Shipwrecked Sailor continues the fascination with biography or pseudo­
biography in the context of traveling and public service, while Three Tales of Wonder
(with its dramatic setting in the Fourth Dynasty) touts the cleverness of (p. 284) lector-
priests and glorifies the birth of a new pharaoh. Worthy of special mention in New King­
dom literature are a series of prayers used as school texts, love poems, further instruc­

Page 2 of 21
Papyrology and Ancient Literature

tions (including some used in school), and a fascinating composition on the immortality of
writers (Bresciani 1990; Lichtheim 1973–1974, vols. 1–2).

With the Prophecies of Neferti, a Middle Kingdom hieratic text preserved on ostraca and
an Eighteenth Dynasty St. Petersburg papyrus (Lichtheim, vol. 1, 1973, 139–145), begins
our record of a series of compositions that foretell the coming of evils upon Egypt
through unrest or invasion, followed by the rise of a great king who brings peace. In a
guise of this kind, such compositions glorify a regime in power or, in the case of works
from the Persian period and later, serve to console Egyptians and express nationalistic
fervor. With the theme of the conquest of Egypt by evil foreigners accentuated, these lat­
er works include the Demotic Chronicle, preserved in an early second-century BCE
papyrus in Paris, and the Oracle of the Lamb, which survives in a demotic Vienna papyrus
of the Augustan period and purports to have been spoken under Pharaoh Bocchoris of the
Twenty-fourth Dynasty. The continuation of this tradition of prophecy into the Roman pe­
riod as seen in the Oracle of the Potter, attested in Greek papyri of the second and third
centuries CE, serves as a reminder that, even where literary texts are concerned, the di­
vide between Egyptian and Greek is not always absolute (Koenen 1968). Demotic papyri
of the Ptolemaic and Roman periods containing the adventures of Setne Khamwas, a New
Kingdom prince, and tales entitled Egyptians and Amazons exemplify the interpenetration
of Greek and Egyptian story elements, while the Myth of the Eye of the Sun is based more
strictly on Egyptian elements (Lichtheim 1980, vol. 3).

Papyri and the Study of Greek Literature

The most spectacular and extensive finds of completely new Greek literature occurred for
the most part at the end of the nineteenth century, when papyrology brought to light sev­
eral entire ancient literary works that were previously known only as names (Turner
1968, 17–24). The body of ancient Greek literature continued to expand on the basis of
papyrological evidence. By the early part of the twentieth century, more Greek authors or
individual compositions had been “rediscovered,” and many additions made to the exist­
ing corpora of several Greek authors, often forcing scholars to rethink their approaches
to these writers' works. Indeed, the papyri had begun to substantially change the land­
scape of ancient Greek literature as it was known to the modern world (Powell and Bar­
ber 1921). Today, with (p. 285) approximately six thousand items published to date as list­
ed in Mertens-Pack3 (see the list of selected electronic resources), texts containing Greek
literature or subliterary texts make up some 87 percent of published literary papyri from
Egypt up to the Arab conquest. These papyri include carefully produced, expensive
bookrolls or codices, informal copies of literary works made by educated readers or stu­
dents, and subliterary texts intended to accompany the reading of authors such as Homer.
School copies and subliterary papyri ancillary to major works especially provide us with
insights into the ways in which works of literature were used for schooling or individual
study (see chapters 11 and 14). Indeed, most aspects of the study of ancient Greek litera­

Page 3 of 21
Papyrology and Ancient Literature

ture and culture have been affected in some way by the recovery, study, and publication
of papyri. Here there is space only to touch upon highlights.

Among Greek poetic texts, our acquaintance with archaic and classical lyric and elegy is
heavily dependent on papyri, before the availability of which authors could be sampled
only through quotations by other ancient writers. For instance, although a number of pa­
pyrological contributions to his varied compositions had been made earlier in the twenti­
eth century, a more recently published Cologne fragment of an epode by Archilochus,
with its passion and sexual explicitness, showed the poet in a new, if not wholly unexpect­
ed, light (Merkelbach and West 1974). About 60 percent of the poetry of Alcaeus known
today derives from papyri, especially those of the Oxyrhynchus collection housed in Ox­
ford. The passages of Alcaeus that are attested on papyrus, although many of their verses
are fragmentary, are nearly always individually longer than those that were previously
known through citation by authors such as Athenaeus and Strabo. Substantial portions of
Sappho's surviving work, too, are known only on papyrus, and discoveries of her work
continue: Thus, a striking new fragment of a poem alluding to the old age of Tithonus as a
parallel to the writer's own graying hair came to light recently from mummy cartonnage
in the Cologne collection, its early third-century BCE date in fact making it one of the ear­
liest known of Sappho's texts (Gronewald and Daniel 2004). The several dozen existing
quoted passages or tags from Alcman and with them our knowledge of archaic Spartan
choral lyric were greatly enhanced by the publication in the later nineteenth century of a
Louvre papyrus containing one hundred verses plus marginal notes, as well as by com­
mentaries on the poet's works that were uncovered in the Oxyrhynchus collection during
the twentieth. Bacchylides would be unknown except as a name were it not for the 1896
discovery in Upper Egypt of a papyrus roll containing several of his dithyrambs and odes.
Pindar himself has become much more comprehensively understood by Hellenists today
now that papyrology has brought to light parts of his paeans and dithyrambs to supple­
ment the Epinician Odes preserved in medieval manuscript. The fragments of the elegiac
poems of Simonides have increased Wvefold (see the case study later on) as a result of
papyrus discoveries, and the papyri have shed considerable new light on the “Homeric”
qualities of Stesichorus while broadening (p. 286) our knowledge of his production overall.
Finally, in the early days of papyrology, a good portion of the popular fourth-century BCE
poet Timotheus's dithyramb Persae was discovered on a roll—itself one of the earliest
Greek papyri, attributable to the late 300s—that had been disinterred from cartonnage.

In the area of hexameter verse, most of the surviving lines of the Hesiodic Catalogue of
Women are still fragmentary. However, thanks to papyri, which are largely responsible for
what we know of the work in detail, we are now in a much better position to assess the
content, connectedness, and overall shape of the poem, as well as its relationship to the
mythographical tradition (Merkelbach and West 1967, 1–120). A Strasbourg papyrus rec­
ognized during the 1990s as preserving some 130 verses of Empedocles' On Nature is
noteworthy for demonstrating that a text of a Presocratic work was in circulation in Egypt
in the Wrst century CE and that the author combined a rationalizing approach to the
physical world with a doctrine of reincarnation (Martin and Primavesi 1998; Janko 2004).
A papyrological window into the biographical hexameters of Erinna, who may have writ­
Page 4 of 21
Papyrology and Ancient Literature

ten during the fourth century BCE, was opened by a papyrus in Florence containing more
than fifty lines attributed to her poem Distaff, which may now be added to several epi­
grams by and about her that are known from the Anthologia Palatina. The skepticism of
one prominent philologist over the likelihood of a self-identified wool worker from a small
Aegean island having written sophisticated poetry in a unique combination of Aeolic and
Doric should not detract from the excitement of recovering more of what may well be an­
other all-too-rare female Greek literary voice (West 1977).

New discoveries on papyrus have enhanced the few works of Attic drama preserved in
manuscript by giving us substantial fragments of additional plays, especially by Euripides
and Aeschylus (Diggle 1998). In the case of the latter these include a large portion of a
satyr play, the Diktyoulkoi, only the second such known to the modern world after Euripi­
des' Cyclops. With respect to Euripides we are especially indebted to the papyri for sub­
stantial knowledge about such tragedies as Hypsipyle and Cresphontes (Bastianini and
Casanova 2005, 1–9). Although the papyri have brought to light commentaries, sum­
maries, and other subliterary accompaniments to classical authors too copious to detail
here (see Bastianini et al. 2004–), a series of Roman-era prose hypotheseis (short sum­
maries) has especially enhanced our knowledge of Euripides' lost or partially preserved
tragedies (Luppe 2002). A general point worth noting with respect to ancient hypotheseis
and summaries of classical works is that the origins, aims (whether for the purpose of ac­
companying the reading of a major work or for use in schools, for instance), and relation­
ships of such texts to their main works, as well as to scholia, ancient commentaries, and
mythographical texts, have in recent decades increasingly become the objects of special­
ized study (Rossum-Steenbeek 1998; cf. Cameron 1995, 119–127).

With respect to New Comedy, previously known largely through Latin adaptations by
Plautus and Terence, papyrology has revolutionized our state of knowledge, (p. 287) espe­
cially by providing us first, nearly a century ago, with the Cairo codex containing multi­
scene portions of two of Menander's plays (plus parts of two others) and second, since the
1950s, with additional fragments of his plays, together with a complete play, the Dyskolos,
via a series of third-century CE codices in the Bodmer collection in Geneva. Beginning in
1891 with the publication of a British Museum roll containing some seven hundred verses
of the literary mimes of Herondas, an author known previously but only through testimo­
nia and brief quotations (Kenyon 1891b, 1–39), the papyri have transformed our picture of
Hellenistic poetry. During the late nineteenth and earlier twentieth centuries, before the
publication of several dozen significant papyri, Callimachus was a poet known for his
Hymns preserved in manuscript and as the author of numerous other poetic and scholarly
works attested in quotations and references by ancient authors or in Byzantine compila­
tions such as the Homeric scholia and the Suda. Extended passages of the Aetia, Hecale,
and Iambi such as we have today were lacking. Although new discoveries continue, the
publication in midcentury of a monumental edition collecting and annotating all of the
known texts, on papyrus and from elsewhere, made possible a much fuller appreciation of
Callimachus's pivotal role with respect to literary innovation (Pfeiffer 1949–1953;
Cameron 1995). The role of the small, damaged, but informative group of papyri contain­
ing scholia and summaries, known as Diegeseis, of Callimachus's major poems should be
Page 5 of 21
Papyrology and Ancient Literature

highlighted here as an example of the contribution of subliterary texts to understanding

and supplementing the fragmentary works themselves, as was mentioned earlier in con­
nection with Euripides.4 However, it is not only Hellenistic authors from Alexandria for
whom papyri provide vital new texts and information. This is illustrated by the enormous­
ly rich assemblage of Hellenistic poetic fragments on papyrus attested by the standard,
late-twentieth-century scholarly corpus (Lloyd-Jones and Parsons 1983; cf. Zetzel 1987).

Furthermore, a blockbuster discovery riveted the attention of Greek and Latin poetry spe­
cialists alike following reports during the 1990s of an extensive, late third-century BCE
papyrus roll recovered from mummy cartonnage and containing more than one hundred
epigrams attributable to Posidippus of Pella. Published in 2001, the “new Posidippus,”
with its thematic grouping of poems that center around divination, stones, dedications,
and the like, has already generated hundreds of scholarly studies on topics ranging from
the arrangement of ancient poetry books to the physical characteristics of bookrolls (Bas­
tianini and Gallazzi 2001; Acosta-Hughes and Renner 2002). All of these papyrological
discoveries both have affected our appreciation of Hellenistic poetry in its own right and
have important consequences for students of Latin authors, many of whose Hellenistic
Greek models or predecessors would otherwise remain but dimly known. For example,
not only can a translation of a Hellenistic original such as Catullus's Coma Berenices
poem (no. 66)now be directly compared with a substantial part of the corresponding pas­
sage in the Aetia, but our new grasp of the spirit and the innovative qualities of
Callimachus's work as a (p. 288) whole also provides numerous opportunities to better ap­
preciate the ways in which Roman poets utilized or reworked Callimachean approaches to
poetry, in the process usually creating something uniquely Roman.

With regard to prose as well, the papyri have infused new texts into the body of available
Greek literature. Probably the most critical single discovery of prose on papyrus for all of
classical studies was revealed by the publication in 1891, by F. G. Kenyon of the British
Museum, of a text written on a group of four first-century CE rolls obtained at an undeter­
mined location in Upper Egypt and containing the nearly complete Athenaion Politeia, at­
tributed to Aristotle (Kenyon 1891a). This “constitution,” which traces the history of
Athenian government from the Dark Age to the fourth century, was reported in ancient
references to have formed part of a series of studies of different forms of polity by Aristo­
tle and his pupils. Only a small part of the text had been known since 1880, on the basis
of a papyrus fragment in Berlin. The Athenaion Politeia has become a fundamental source
of information on Athenian history and one in the absence of which historians of ancient
Greece today would scarcely recognize their own field of study.5

Our access to lost Attic prose was also enriched by the recovery in Egypt, between 1847
and the 1880s, of four papyrus rolls of Greek oratory ranging from the first century BCE
to the second CE. They made it possible for the first time in the modern world to sample
six speeches—one of them nearly complete, the others less so—of the fourth-century ora­
tor Hyperides alongside the remaining corpus of Attic orators transmitted to us in manu­
script. Another event provided a new historical voice for our understanding of the con­
cluding years of the Peloponnesian War and the period immediately following. This is the

Page 6 of 21
Papyrology and Ancient Literature

publication early in the twentieth century of a portion of the so-called Hellenica

Oxyrhynchia, which has occasioned lively scholarly debate concerning its author's take on
events and relationship to contemporary sources such as Xenophon and Diodorus. Prose
from the classical period, too, is a papyrus text written in an “epigraphic” script attribut­
able to as early as the mid-fourth century BCE and containing commentary on Orphic
theogonic verses. Unique for the fact of its deriving not from Egypt but from an excava­
tion in Greece (near Thessaloniki), the so-called Derveni papyrus is of great importance
for Greek religion in the age of the Sophists. It also appears to be the oldest known per­
ishable object carrying a Greek literary text (Janko 2002). To move somewhat later in
time, what we know of the history and the subject-matter range of the ancient novel has
changed considerably as a result of papyrus finds. Fragments of Chariton datable to the
second and third centuries CE overturned the early twentieth-century assertion that this
relatively early author of novels—and therefore the entire genre—was a product of the
late Roman world. Further, the discovery of fragments of ancient Greek novels, such as
the Ninus Romance or the Phoinikika of Lollianos, with its crime and violence—both of
which were previously unknown except through the testimony of other authors—has
shown that not all ancient Greek novels may be itted into the traditionally recognized cat­
egories of ideal and rhetorical romance (Stephens and (p. 289) Winkler 1995; Stephens
2003). Finally, in addition to the hypotheseis and summaries designed to accompany ma­
jor authors that were discussed earlier, we ought to mention one other type of subliterary
composition, the grammatical treatise. It would be fair to say that papyri have significant­
ly changed the possibilities for the questions we might ask about the history of such texts.
Whether focused on a special topic or aimed at covering schematically a series of stan­
dard points, as discussed in the case study later on, ancient grammatical works on pa­
pyrus provide an important supplement to treatises known from manuscript and allow us
to begin to see what kinds of grammar summaries or studies were current in one Greek-
speaking province, as well as how those texts might actually have been utilized.

Papyrology has also changed many of the ideas once firmly held about the textual history
of Greek authors whose main works are fully preserved in the manuscript tradition and
has often provided material for debate about the nature of that tradition. The discovery of
early Hellenistic papyri of Homer was especially unsettling and provocative for textual
critics because many such papyri contain so-called “plus verses,” verses not found in the
medieval manuscripts. The generally accepted solution is to attribute the elimination of
most of these after the second century BCE to the in Xuence of Aristarchus's editing and
commenting. On the whole, the textual characteristics of Iliad and Odyssey papyri of the
later Hellenistic and Roman periods do not differ greatly from those of the medieval man­
uscript traditions, which formed the basis of the first modern critical editions. Neverthe­
less, recent scholarship shows that even the Homeric papyri of the Roman period—which
are so numerous as to make Homer the undisputed leader in terms of frequency of finds
of literary papyri—need to be taken into account in assessing any Homeric verse if they
provide evidence for it (Haslam 1997; Skafte Jensen 2005). Among classical authors,
Homer is somewhat of an anomaly because of the oral aspects of his composition and be­
cause of modern scholars' intense interest in the earlier stages of transmission. However,

Page 7 of 21
Papyrology and Ancient Literature

a study of the papyri of nearly any poetic or prose author—papyri will often be distributed
over time from the Hellenistic period to the Arab conquest—leads to new insights into the
ways in which one may view the medieval manuscript tradition. Further, such an exami­
nation often impresses upon us the relative lack of adherence, by comparison with mod­
ern expectations, of ancient editions to a single received text (see for example Haslam
1978).6

Papyri and Latin Literature

Papyri of Latin authors are relatively rare in Egypt. Totaling about two hundred known
items at present, they consist mainly of several dozen fragments of known works of classi­
cal authors such as Terence, Cicero, Sallust, Vergil, Livy, Lucan, and Juvenal, as well as a
number of legal texts. However, two especially noteworthy (p. 290) contributions of papy­
rology to Latin letters deserve mention here. The first is a papyrus fragment containing
several new verses of Latin elegy that was discovered at Qasr Ibrim, in Egyptian Nubia
about 150 miles south of Aswan, dated by archaeological, palaeographical, and historical
evidence to between 50 BCE and 25 CE, and argued persuasively to be the work of the
early Augustan poet and ill-starred military leader Cornelius Gallus (Anderson, Parsons,
and Nisbet 1979). Although many of the twelve lines represented are fragmentary, the pa­
pyrus is of special interest to historians of Latin literature seeking to understand the ba­
sis for the flowering of Roman elegy in the Augustan world and hungry for even a scrap of
new text which might shed light on the development of this genre between Catullus and
Propertius. The second contribution is a Michigan papyrus (strictly speaking, it is a page
of a vellum codex) containing about forty verses from Seneca's Medea which appears to
preserve intact readings that became corrupt in the archetype of the medieval tradition.
The papyrus's fourth-century CE date shows that Seneca's tragedies were read in the
East during this relatively late period and seems to help confirm a surge in interest in
learning Latin in Egypt at this time (Markus and Schwendner 1997).

Resources for Editing and Studying Literary

Papyri
Although Greek and Latin literary papyri share many commonalities with documentary
papyri and are frequently published in volumes together with documents,7 their study has
often been seen as a separate branch of Graeco-Roman papyrology.8 Indeed, a senior Hel­
lenist and papyrologist who works on both literary and documentary Greek papyri once
remarked that editing a literary papyrus is something that any Greek and Roman classi­
cist ought to be able to do because many of the basic principles for establishing and criti­
cizing the text are similar where both medieval manuscripts and literary papyri are con­
cerned. However, editing literary texts that are recovered from papyrus requires a specif­
ic set of resources which overlap with but are not in all respects the same as those used
by scholars who are concerned primarily with Greek and Roman literature known only
through the medieval tradition. Obviously the editor or critic of a literary papyrus em­
Page 8 of 21
Papyrology and Ancient Literature

ploys whatever part of the vast resources of classical philology are necessary. These in­
clude edited texts of authors, scholarly commentaries on works of literature, collections
of fragments of writers in a particular genre, and lexica, whether in print or electronic
form. Among the electronic tools developed by classical studies in general, the University
of California—Irvine's online Thesaurus Linguae Graecae (TLG), which collects and
presents all texts attributed to Greek authors from Homer to the fall of (p. 291) Byzantium
in searchable form, has become an essential tool for papyrologists seeking to identify
newly found texts within the corpus of Greek literature or to locate parallel passages to
new texts on papyrus.9

More specialized compilations that focus wholly or mainly on the papyri and are useful
for literary papyrology consist of three groups. Those of the first are organized around
types of literary or subliterary works and may include texts other than those on papyrus.
A number of Greek literary texts, especially many of those that are revealed by papyri,
are anonymous because the names of their authors are not preserved or on account of
their subliterary nature. Thus, many are not contained in the TLG. Along with texts to
which names of authors are attached, some categories of these unattributable texts have
been included in printed corpora specific to their genres, notably collections of papyri
containing philosophy, biography, ancient novels, and grammar (Adorno et al. 1989–1999;
Gallo 1975–; Stephens and Winkler 1995; Wouters 1979). An additional tool, the Cata­
logue of Paraliterary Papyri (CPP), currently an electronic data bank in process, contains
descriptions of Greek papyri and other written materials which, because of their paralit­
erary, diverse, and anonymous character, cannot be found in the TLG and tend to be ac­
cessible only in scattered publications. Begun as a database of mythographical papyri, the
CPP is currently adding papyri with all kinds of lists and catalogues. The CPP's stated
goal is to include digital versions of full-text editions of the paraliterary papyri and to in­
corporate other materials such as grammatical papyri and commentaries, as well as med­
ical and liturgical texts.

A second type of resource is organized primarily around papyri and other ancient books
as physical objects in historical time and in space (both their places of origin, insofar as
known, and their location in modern collections). The Advanced Papyrological Informa­
tion System (APIS), whose goal of providing online data and, where possible, images of all
papyri in collections is approaching completion for North American centers of papyrologi­
cal study, and has made substantial progress with respect to Europe. The APIS essentially
treats literary papyri in the same way as it does the documentary texts that make up the
majority of its entries. Another resource, one that focuses solely on literary texts, is the
online Leuven Database of Ancient Books (LDAB). Moreover, it covers not only literary pa­
pyri but also early books of other kinds, such as the great biblical codices of the late Ro­
man period.10 At present the LDAB comprises more than twelve thousand items that date
from the fourth century BCE to 800 CE. The LDAB's data bank includes some thirty-seven
hundred anonymous texts, most of which are on papyrus. The project uses the term books
to mean “texts that were intended to reach the eyes of a reading public or at least pos­
sessed a more than ephemeral interest or usefulness.”11 The LDAB does not attempt to

Page 9 of 21
Papyrology and Ancient Literature

give a full bibliography for each item, whereas the institutional web pages that are linked
to form APIS do.

Into a third category falls the single most useful organizational tool for studying literary
papyri and one that is also essential for the scholar of Greek (p. 292) and Roman literature
concerned with authors represented on papyrus: R. A. Pack's annotated catalogue of
Greek and Latin literary papyri from Egypt. The second print edition, designated Pack2
(Pack 1965), in which 3, 026 literary and subliterary papyri are catalogued according to
author, if known, and according to genre in the case of anonymous texts, is now forty
years old. Its successor, created by the Centre de Documentation de Papyrologie Lit­
téraire (CEDOPAL) of the Université de Liège and referred to as Mertens-Pack3, is based
on the numbering system of Pack2 but is now an online resource that is updated regularly.
Like the earlier versions of Pack, it gives a full bibliography for each item.12 To ascertain
what papyri covering a particular genre of Greek or Latin literature have been published
so far, their dates, locations, and other metadata, as well as what studies have been pub­
lished concerning them, one goes first to Pack2 and Mertens-Pack3.13

It is obvious that these various resources overlap to a greater or lesser extent in terms of
coverage. Some are linked—for example, LDAB and CPP. There is not sufficient space
here to discuss in detail the somewhat differing varieties of metadata that are given by
each of these compilations and their arrangement in different database fields. For in­
stance, compared with the LDAB and Mertens-Pack3, the CPP provides more detailed de­
scriptions of the contents, as well as the state of preservation, hand, and lectional signs.
The majority of these electronic tools do not provide actual texts or images; exceptions
are the images provided on many of the institutional Web sites which make up APIS and
the intention of the CPP to include online texts. Finally, ongoing bibliographical compo­
nents of two long-established papyrological journals remain especially useful: (1) the sys­
tematic “Literarische Texte unter Ausschluss der Christlichen” sections which appear pe­
riodically in Archiv für Papyrusforschung and include summaries with brief evaluations of
new texts; (2) the relevant parts of the thorough “Testi recentemente pubblicati” and
“Bibliografia” sections in Aegyptus.

Electronic bibliographical tools and databases for Greek and Latin papyri are thus quite
well developed. These items do not, however, include hieratic and demotic literary papyri,
for which one must turn to the newly conceived Demotistische Literaturübersicht, housed
at the Katholieke Universiteit in Leuven.

A Look at the “New Simonides”: A Brief Case

Study
The expansion of what we know about the elegiac compositions of Simonides' poetry as a
result of papyrological discoveries is a good example of how new literary papyri can have
far-reaching implications for classical studies not simply by winning additional lines of
text for us to read but also by contributing to literary (p. 293) historical debates or provok­

Page 10 of 21
Papyrology and Ancient Literature

ing new ones. The publication of P.Oxy. LIX 3965 by P. Parsons in 1992, following up on
suggestions made by E. Lobel, who in 1954 had published P.Oxy. XXII 2327, containing
poetry by an unknown author, allowed the surviving text of the two rolls to be combined.
This resulted in a more extensive group of lines comprising one or more elegiac poems by
Simonides that center around events of the Persian Wars. The two rolls in question, each
datable to the second century CE on the basis of their writing and each consisting of
many fragments defined by a common hand and fabric, overlap at various points and may
be different copies of the same collection of poetry although they were not written by the
same person.14 The identification of P.Oxy. LIX 3965 as Simonides is representative of
how literary papyrologists make the most of their often scattered data. A few words of
one scrap were shown to belong to a passage of Simonidean elegiac verses quoted by Sto­
baeus. Another tiny scrap, labeled Fr. (i.e., fragment) 5, bore bits of three lines of writing
from the interior of a column and included only eight Greek letters that could be read
without dispute. Two of these scanty lines overlap, however, with Simonidean verses
quoted by Plutarch.

Let us take a closer look at the editing process for Fr. 5. As is customary for new literary
texts in P.Oxy. volumes, the scrap was first represented in “diplomatic” transcript15
accompanied by a critical apparatus describing the surviving ink of the uncertain letters
in detail and in as objective terms as possible.16

1]., foot of upright with junction to left, ν acceptable.

2]. δ‥ [, first, probably crossbar and right-hand side of η; second, perhaps upper
right-hand arc of small circle; third, short horizontal trace at upper level.
3] π̣ολυ̣ [, of π the top, part of the left-hand junction, and the right-hand side with a
rather messy join (so rather than γ [or the like] ε Lobel); of υ the left-hand branch (ν
Lobel, but the stroke extends too far to the left).

To this bare information let us add what we know from elsewhere. A previously known
pair of fragments of Simonides' elegiacs, designated Fragments 10–11 in M. West's first
edition of elegiac and iambic poets, consists of three couplets quoted successively almost
in their entirety by Plutarch (De Herodoti malignitate 42, 872d), who attributes them to
Simonides and cites them as evidence that the Corinthians stayed in their assigned posi­
tion during the battle of Plataea and did not run away as Herodotus implies. The final two
verses overlap with the first two fragmentary lines (p. 294) of the scrap (in West's second
edition the passage was renumbered as Fragments 15–16; see West 1992, 121–122).
Parsons's restored transcript of Fr. 5 of P.Oxy. 3965 uses halved square brackets (└┘) to in­
dicate the boundaries of the papyrus scrap within the text derived from Plutarch:

P.Oxy. 3965 Fr. 5 + frr. 10–11 (West)

Page 11 of 21
Papyrology and Ancient Literature

Fr. 5

In itself Fr. 5 added only four letters—]π̣ολυ̣[, two of which are in fact uncertain—to what
was already known about Simonides' work. However, the identification of 3965 (and con­
sequently 2327) as containing elegiacs by him enabled the more extensive parts of both
rolls to be used to significantly enrich our knowledge of the poet's work. Prior to these
developments, a comment in the Byzantine lexicon Suda that Simonides composed poems
in elegiac meter on battles of the Persian Wars had been viewed skeptically. The number
and nature of the separate poems that may be represented by the rolls, together with
their relationship to possible public performance, to the historical sympathies of different
Greek allies, and to the narrative of Herodotus, are still the subject of lively debate. How­
ever, the contents of the new papyri do indeed appear to include elegiac verses on several
notable military actions such as Plataea and Artemisium, as well as a symposium scene
that may be part of yet another elegiac composition.17

Perhaps most striking is the poem represented by a group of fragments which, it is plausi­
bly argued, celebrates—probably for a public occasion within a year or two of the actual
event—the battle of Plataea. This composition must have been one hundred or more vers­
es in length, perhaps considerably more. As reconstructed by Parsons and commented up­
on by others, the earliest extant lines discuss the Trojan War and the death of Achilles,
thereby setting up a presumed parallel between an ancient East-West conflict and the
contemporary one with Persia. Invoking the Muse, the poet describes the departure of the
Spartan army, accompanied by “the horse-taming sons of Zeus, the hero Tyndaridai, and
mighty Menelaos,” from home and its progress to Corinth, Megara, and Eleusis before ar­
riving at Plataea (a catalogue of forces before the battle furnishes a plausible context for
the praise of the Corinthians in Fr. 5, described earlier).

Little or nothing remains of the battle description. However, a measure of the attention
that the contents of these two rolls have attracted among scholars of (p. 295) ancient liter­
ature is the flurry of publications that have been devoted to the “New Simonides,” includ­
ing a special edition of the classical periodical Arethusa, with articles by a wide range of
Hellenists and, most recently, an edited volume exploring Simonides' elegiacs (Boedeker
and Sider 1996, 2001). The extent to which Simonides in the Plataea poem provides an
epic framework for recent events is striking. In many other respects as well, the labors of
the P. Oxy. scholars over what would seem to many but philological minutiae have pro­
duced consequences that are far reaching for an understanding of the Greek poetic tradi­
tion, as well as the history and culture of the early fifth century BCE.

Page 12 of 21
Papyrology and Ancient Literature

“Open” Grammatical Texts: A Brief Case Study

To a greater or lesser degree, texts of major authors in ancient times often vary from copy
to copy more than we may be used to seeing. After all, nineteenth- and twentieth-century
classical philology has taught us to establish “the” text. Let us now look at a passage from
a Greek grammatical fragment, P.Mich. inv. 30, which illustrates how certain types of sub­
literary texts are “open” to a much greater extent. The papyrus, of unknown provenance,
arrived in Ann Arbor in 1930 after purchase on the antiquities market.18 Its contents most
resemble a Technê Grammatikê, a relatively brief but in a certain sense comprehensive
treatise on Greek grammar that is typified by a composition preserved in manuscripts
from the tenth century on, where it is attributed to Dionysius Thrax, a second-century
BCE pupil of Aristarchus. A group of around two dozen papyrus fragments ranging in
date from the first through the sixth century CE are close in character and content to
Dionysius, albeit with numerous individual variations in detail (Wouters 1979, 33–210). 19

By examining the discussion of adverbs (ἐπιρρήματα) in Column II of the Michigan pa­

pyrus, we can appreciate the fluidity of this grammatical tradition more fully. Written in a
practiced but informal hand that shows fluctuation in the straightness of the lines as well
as variant forms for several of the letters—utilitarian qualities that are quite common in
subliterary papyri—it appears to be part of a leaf of a codex assignable to the third or
fourth century CE on the basis of palaeography. After the text was initially written, addi­
tions—perhaps by the writer of the text, although they have a slightly more formal quality
and appear to be in darker ink—were made above the line at various points. In line 8,
someone has crossed out several letters with vigorous diagonal strokes. High stops were
used from time to time, especially to separate words in a list of examples:

(p. 296) Col. II (Recto)

–––––––––––

Page 13 of 21
Papyrology and Ancient Literature

2‥: Part of horizontal in midline; alternatively, read ]ανδηλ̣ [.].μ[ 4. Read ἐπιρρημάτ [ων 5.
αυτικα˙ pap. 6‥ [: ι, ρ, φ, ψ 10. κατω˙ ανω˙ pap. 11.].: α, ε, κ? 12. Read ἐπιρρήματα; next,
possible space-filler ink 15‥ [: φ or c, then high horizontal 16. ἐπι]ρρήματα seems less
likely‥ [: τ or υ? 17.].: α or ω? 18.] μας˙: pap‥ [: ν? 19. Read λικριφίς 20.]α˙: pap.

3. προ (or ύπο-)τακ]τικὴ ῥήματ[ος, “placed before (or after) a verb.”

4. After the basic definition of adverb, a distinction is made between those that are simple
or uncompounded (άπλᾶ, άσύνθετα) and those that are compound (σύνθετα).

5. νῦν, added above the line, is perhaps intended as a second example (alongside αὐτίκα)
of adverbs that are ἁπλᾶ.

6. Perhaps τά δὲ] ἀσ[ύ]νθετα σάφ[α, “and the uncompounded (adverbs): sapha …” with
σάφα, “clearly,” being the Wrst example of this category cited.

7–9. Whether or not our suggested supplement in 7–8 is correct, in some way the writer
mentions παρονομασία or some form of παρονομάζειν, a term that is not easily paralleled
in Technê-type texts. Then something has been deleted, (p. 297) followed by discussion in­
volving at least two examples, καλῶς and τηνικάδε: Perhaps the point is that these two
adverbs are “derived,” παρονομασμένα.

8. The deleted letters look somewhat like ρον.

Page 14 of 21
Papyrology and Ancient Literature

9. Although τῃνικάδε refers to time, it may be cited here because it is said to be “derived”
like the words cited in 7–8. Adverbs of time seem to be discussed as a category in 11–12.

10. ] ιν: It does not seem possible to read the expected οἷον, “for example.” κάτω ἄνω
should belong to the category of ἐπιρρήματα τοπικά, or directional adverbs.

11–12. In 11 the writer apparently turns again to adverbs of time: The word ἐπιρήματα in
12, together with the fragmentary adjective—perhaps δηλωτι] κά or τοπικά or παραστατι]
κά (cf. 14), “expressive of”—might still be part of this discussion.

13–14. “Under these must be classified.” The reference must be to the category of ad­
verbs referred to in 12.

14. π] ᾳραστατικά, “expressive of.”

11. ἐ] κτω̃ν χρόνου?

17. ναί μω̃ [ν, as a pair of examples? As the comparanda (cf. below) show, the examples of
adverbs given in 18–21 belong to the list of adverbs of ποιότης (or μεσότης) indicating
manner.

18. Perhaps ν[ύξ = γνύξ, “with the knee.”

19. λιγρυφίς = λικριφίς, a rare Homeric adverb meaning “sideways” or “aside” appears
in Il. 14.463 and Od. 19.451. It is cited by Herodian four times in discussions of Homeric
prosody. In the Technê papyrus P.Lit.Lond. 182, it is part of a long list of adverbs that are
μεσότητος καά ποιότητος δηλωτικά, “expressive of state and quality.”

21. σφόδρ] α? αἰψ]α? ἃμ] α?

The closest parallels to the Michigan text are provided by passages in Dionysius Thrax
and in two Technê-type papyri of the first century CE and around 300, respectively. They
are of great help in understanding the papyrus and in supplementing the missing sections
—but none of them offers anything like a word-for-word correspondence:

Dionysius Thrax, Technê Grammatikê 19 (Grammatici Graeci I l, 72–86)

P.Yale 25.37–53 (I cent. CE, assigned) ἐπίρημα δʼ ἐστὶν λέξις κατὰ μίαν ἐκφορὰν

Page 15 of 21
Papyrology and Ancient Literature

(p. 298)

P.Lit.Lond. 182.80–105 (ca. 300 CE, assigned) Ἐπίρρημα. Ἐπίρρημα τί ἐστιν; Λέξις

The surviving examples of adverbs in the Michigan papyrus show quite an impressive spectrum,
from pedestrian καλω̃ς to the exotic Homeric λικριφίς (19), a word used in the Iliad to describe
Polydamas's jumping aside to avoid Ajax's spear. Such an eclectic mixture of Homeric, Attic, and
perhaps contemporary koinê items can, however, be para11e1ed in ancient grammatica1 1itera­
ture, for examp1e, in P.Lit.Lond. 182, which dates to the general period of the Michigan text. Our
papyrus is obviously eclectic with respect to all of the other known Technê compositions— just as
they themselves differ from one another.
As we struggle to understand and restore the fragmentary lines, we sometimes find our­
selves returning to the same comparanda as “trots”: for example, Dionysius or the Lon­
don papyrus. These may themselves be separated in time by as many as five centuries:
The London papyrus is assigned on palaeographical grounds to around 300 CE. The work
attributed to Dionysius is purportedly from the second century BCE, although arguments
have been advanced to place all or part of it as late as the fourth CE. However, even if
Dionysius's Technê originated so late, it is clear that the Michigan papyrus regularly
shares features of detail with the comparanda (both those cited earlier and others) rang­
ing from the first through at least the fourth centuries CE. But the features are never the
same if we look at the discussion of adverbs overall. Thus, categories of adverbs are nev­
er identical. In some texts, too, the classification system is more elaborate than in others.
The examples that are given by these texts seem to (p. 299) draw from a common pool but
never quite match, and the definitions of the adverb vary. Ancient grammatical treatises
of this type, like several other categories of subliterary texts such as scholia, did not form
a single tradition with one line of evolution in antiquity—as the papyri have been especial­

Page 16 of 21
Papyrology and Ancient Literature

ly effective in demonstrating (Swiggers and Wouters 1995). In a sense, there can be as

many different Technai as there are copies executed over several centuries. Each gram­
matikos, or other interested person who copies (or has copied) a Technê can potentially
make additions to, or subtractions from, the prior text based on personal preference or on
schoolroom needs. The Technê-type treatise is thus the ultimate “open” text. The Michi­
gan papyrus, in spite of its fragmentary state, helps to reinforce these lessons and at the
same time offers scholars who are used primarily to the philological study of standard
Greek and Roman authors a new perspective on the transmission of texts.

Bibliography
Acosta-Hughes, B., and T. Renner. 2002. “Special Review Article.” Review of Bastianini
and Gallazzi (2001). BASP 39: 165–187.

Adorno, F., A. Carlini, F. Decleva Caizzi, M. S. Funghi, D. Manetti, M. Manfredi, and F.

Montanari, eds. 1989–1999. Corpus dei papiri filosofici greci e latini: Testi e lessico nei
papiri di cultura greca e latina. Florence: Leo S. Olschki.

Anderson, R. D., P. J. Parsons, and R. G. M. Nisbet. 1979. “Elegiacs by Gallus from Qasr
Ibrim.” Journal of Roman Studies 69: 125–155.

Bastianini, G. 1996. “Un luogo di ritrovamento fantasma.” In Atti del II convegno

(p. 301)

nazionale di egittologia e papirologia di Siracusa, 1–3 dicembre 1995, 69–84. Syracuse:

Istituto Internazionale del Papiro.

———, and A. Casanova, eds. 2005. Euripide e i papiri: Atti del convegno internazionale di
studi, Firenze, 10–11 giugno 2004 (Studi e Testi di Papirologia, n.s., 7). Florence: Istituto
Papirologico “G. Vitelli.”

———, and C. Galazzi, eds. (2001). Papiri dellʼUniversità degli Studi di Milano VIII: Po­
sidippo di Pella: Epigrammi (P. Mil. Vogl. VIII309). Università degli Studi di Milano, Pub­
blicazioni della Facoltà di Lettere e filosofia, CC. Con la collaborazione di Colin Austin.
Milan: LED—Edizioni Universitarie di Lettere Economia Diritto.

Bastianini, G., M. Haslam, H. Maehler, F. Montanari, and C. Römer. 2004–. Commentaria

et lexica graeca in papyris reperta (CLGP), adiuvante Marco Stroppa. Munich: Saur.

Boedeker, D., and D. Sider. 1996. “Fragments 1–22: Text and Apparatus Criticus.”
Arethusa 29(2): 155–166.

———. (2001). The New Simonides: Contexts of Praise and Desire. New York: Oxford Uni­
versity Press.

Bresciani, E. (1990). Letteratura e poesia dellʼantico Egitto, 2d ed. Turin: Einaudi.

Cameron, A. (1995). Callimachus and His Critics. Princeton: Princeton University Press.

Page 17 of 21
Papyrology and Ancient Literature

Diggle, J., ed. (1998). Tragicorum Graecorum Fragmenta Selecta. Oxford: Oxford Univer­
sity Press.

Gallo, I. 1975–. Frammenti biografici da papiri. Rome: Edizioni dellʼAteneo.

Gronewald, M., and R. W. Daniel. 2004. “Ein neuer Sappho-Papyrus.” ZPE 147: 1–8.

Haslam, M. W. 1978. “Apollonius Rhodius and the Papyri.” Illinois Classical Studies 3: 47–
73.

———. 1994. “The Contribution of Papyrology to the Study of Greek Literature: Archaic
and Hellenistic Poetry.” In Proceedings of the 20th International Congress of Papyrolo­
gists, Copenhagen 23–29 August 1992, ed. A. Bülow-Jacobsen, 98–105. Copenhagen: Mu­
seum Tusculanum Press.

———. 1997. “Homeric Papyri and Transmission of the Text.” In A New Companion to
Homer, ed. I. Morris and B. Powell, 55–100. Leiden: Brill.

Janko, R. 2002. “The Derveni Papyrus: An Interim Text.” ZPE 141: 1–62.

———. 2004. “Empedocles, On Nature I 233–364: A New Reconstruction of

P.Strasb.gr.Inv. 1665–6.” ZPE 150: 1–26.

Kenyon, F. G., ed. 1891a. Aristotle on the Constitution of Athens. London.

———, ed. 1891b. Classical Texts from Papyri in the British Museum. London.

Koenen, L. 1968. “The Oracle of the Potter.” ZPE 2: 178–209.

Lichtheim, M. 1973–1980. Ancient Egyptian Literature: A Book of Readings. 3 vols. Berke­

ley: University of California Press.

Lloyd-Jones, H., and P. J. Parsons, eds. 1983. Supplementum Hellenisticum. Berlin: W. de

Gruyter.

Lowe, E. A. 1934–1966. Codices Latini Antiquiores: A Palaeographical Guide to Latin

Manuscripts prior to the Ninth Century. Oxford: Clarendon.

Luppe, W. 2002. “Scholia, Hypomnemata, und Hypotheseis zu griechischen Dramen auf

Papyri.” In Der Kommentar in Antike und Mittelalter, ed. W. Geerlings and C. Schulze, 55–
77. Leiden: Brill.

Manfredi, M. 1992. “LʼAthenaion Politeia di Aristotele e i papiri.” In Proceedings of the

XIX International Congress of Papyrology, Cairo 2–9 September 1989, ed. A. H. S. El-Mos­
allamy, vol. 1, 447–460. Cairo: Center for Papyrological Studies.

Markus, D., and G. Schwendner. 1997. “Seneca's Medea in Egypt (663–704).” ZPE 117:
73–84.

Page 18 of 21
Papyrology and Ancient Literature

Martin, A., and O. Primavesi, eds. (1998). LʼEmpédocle de Strasbourg. Berlin: W.

(p. 302)

de Gruyter.

Merkelbach, R., and M. West, eds. (1967). Fragmenta Hesiodea. Oxford: Clarendon.

———. 1974. “The Cologne Epode of Archilochus.” ZPE 14: 97–112.

Pack, R. A. (1965). The Greek and Latin Literary Texts from Greco-Roman Egypt, 2d ed.
( = Pack2) Ann Arbor: University of Michigan Press.

Pfeiffer, R., ed. 1949–1953. Callimachus. 2 vols. Oxford: Clarendon.

Powell, J. U., and E. A. Barber, eds. (1921). New Chapters in the History of Greek Litera­
ture: Recent Discoveries in Greek Poetry and Prose of the Fourth and Following Centuries
B.C. Oxford: Clarendon.

Quack, J. F. (2005). Einführung in die altägyptische Literaturgeschichte III: Die demotis­

che und gräko-ägyptische Literatur. Vol. 3, Einführung und Quellentexte zur Ägyptologie,
Band III. Münster: LIT Verlag.

Rossum-Steenbeek, M. van. (1998). Greek Readers' Digests? Studies on a Selection of

Subliterary Papyri (Mnemosyne, suppl. 175). Leiden: Brill.

Skafte Jensen, M. 2005. Review of G. Nagy, Homer's Text and Language (Champaign, Ill.
2004). Bryn Mawr Classical Review 2005.04.0474:47.

Stephens, S. 2003. “Fragments of Lost Novels.” In The Novel in the Ancient World, rev.
ed., ed. G. Schmeling, 656–683. Leiden: Brill.

———, and J. Winkler, eds. (1995). Ancient Greek Novels: The Fragments: Introduction,
Text, Translation, and Commentary. Princeton: Princeton University Press.

Swiggers, P., and A. Wouters. 1995. “Techne et empeiria: La dynamique de la grammaire

grecque dans lʼantiquité à la lumière des papyrus grammaticaux.” LALIES 15: 83–100.

Turner, E. G. (1968). Greek Papyri: An Introduction. Oxford: Clarendon.

West, M. L. 1977. “Erinna.” ZPE 25: 95–119.

———, ed. (1992). Iambi et Elegi Graeci ante Alexandrum cantati, 2d ed. Oxford: Oxford
University Press.

Wouters, A. (1979). The Grammatical Papyri from Graeco-Roman Egypt: Contributions to

the Study of the “Ars Grammatica” in Antiquity. Verhandelingen van de Koninklijke Acad­
emie voor Wetenschappen, Letteren, en Schone Kunsten van Belgie, Jaargang XLI, 92.
Brussels: Academie voor Wetenschappen, Letteren, en Schone Kunsten van Belgie.

Zetzel, J. 1987. “Fragmentary Pleasures.” Classical Philology 82: 347–362.

Page 19 of 21
Papyrology and Ancient Literature

Notes:

(1.) However, some of the electronic resources discussed in the present chapter include
such texts. In addition, Christian texts are treated in chapter 25.

(2.) Under the category of literature written in Egyptian I here do not include funerary
texts.

(3.) The preliminary online version of the Demotistische Literaturübersicht (Enchoria);

see the section on resources for editing and studying literary papyri and the list of select­
ed electronic resources.

(4.) An important set of scholia on Aetia, book 1, happens to have been written on one of
the British Museum rolls containing the Athenaion Politeia rolls; see note 5.

(5.) For the mystery surrounding where and how the rolls were acquired in Egypt, see
Bastianini (1996) and Manfredi (1992).

(6.) For considerable variation from the medieval text of Plato in a papyrus of the second
or third century CE, see Adorno et al. 1991, 1.1 (1999), pp. 212 ff. on P.Mich. inv. 5980.

(7.) From a cultural studies perspective it is interesting to note that, since the late nine­
teenth century, the canonical order of presentation in a “mixed” volume of edited papyri
has been biblical texts first, then Greek and Latin literary texts, and finally documents.

(8.) Although the original motive for its creation was principally to assemble and stan­
dardize references to the very large number of volumes containing documentary papyri,
the Checklist includes a number of entries for volumes that contain only literary or sublit­
erary papyri (e.g., P.Lond.Lit.), although it does not claim thoroughness in this respect.

(9.) For full references to this and other web-based tools, see the selected list of electron­
ic resources on page XV.

(10.) Included are all of the items in Lowe (1934–1966).

(11.) The LDAB, therefore, excludes papyrus documents quoting lines of Homer, ques­
tions to oracles, and magical texts other than those that are probably handbooks.

(12.) It is interesting as a measure of the increased publication of literary papyri during

the past half century that the first edition of Pack, completed in 1952, listed only 2,369
published or otherwise noticed papyri, while Mertens-Pack3 currently lists about 7,000.
Although Mertens-Pack3 as a whole is an online tool, since 1981 P. Mertens, the initiator
of the Liege project, and collaborators have published a number of separate articles that
present current information about the papyri of classical and Hellenistic authors, as well
as medical writers; see the Mertens-Pack3 webpage for full details. In spite of the refer­
ence to “Greco-Roman Egypt” in the title, Pack2 includes the literary papyri from Dura-
Europos.

Page 20 of 21
Papyrology and Ancient Literature

(13.) Note that the LDAB groups in a single entry multiple texts contained on a single roll
if they were intended as parts of one book. This is especially the case for anthologies,
which in Pack and Mertens-Pack3 are split up among entries for the different authors
quoted in them. However, the LDAB makes two entries when a literary papyrus is reused
for another literary text and also follows this practice in the case of composite codices
when the different works were originally written on individual quires.

(14.) Lobel had already suspected the Simonidean authorship of P.Oxy. XXII 2327. In 1981
in P. Turner 3, he succinctly set forth his criteria for naming Simonides as author of this
and two other P.Oxy. rolls containing lyric verse. See also Parsons's introduction to P.Oxy.
LIX 3965.

(15.) In addition to reflecting the fact that only one form of sigma was used in ancient
texts, the use of lunate sigma in P.Oxy. is a long-standing practice intended, in the case of
fragmentary texts, to maintain an initial stance of editorial neutrality with respect to
whether the letter falls in the middle of a word or at the end.

(16.) The P.Oxy. series customarily gives more detailed descriptions of damaged or uncer­
tain letters than do many publications of literary papyri. The practice of providing exten­
sive descriptions of ink in the critical apparatus of a literary papyrus text serves as an ad­
ditional kind of editor's commentary. It also stems from the fact that many literary texts
do not lend themselves to restoration by close parallels from other texts in the way that
documentary texts often do, and it reflects a long-established tendency among classical
studies to treat papyri containing literature with greater ceremony and awe than the
more numerous and, as traditionally viewed, more mundane documentary papyri.

(17.) Both of these papyri fall into a larger group of second-century CE Oxyrhynchus rolls
that contain lyric and elegiac poetry from various archaic and early classical poets or
commentaries on them.

(18.) The papyrus was first presented in a conference paper, “A New Michigan Grammati­
cal Fragment and the Study of Greek Grammar in I-IV Century Egypt,” at the XXIV Inter­
national Congress of Papyrology, Helsinki, August 2004.

(19.) Wouters (1979) discusses sixteen Technê-type texts on papyrus; additional papyri
have been published since or are being prepared by Wouters.

Timothy Renner

Timothy Renner is Professor of Classics and General Humanities, Montclair State

University.

Page 21 of 21
The Special Case of Herculaneum

The Special Case of Herculaneum

David Sider
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0013

Abstract and Keywords

Papyrologists have many archives but only one Greek library—the collection of rolls found
in a richly endowed villa that had once been on the shore of the bay of Naples in Hercula­
neum. The modern recovery of Herculaneum began in 1709, when workers digging a well
in the Italian town of Resina struck a theater many feet below the surface. The excava­
tions came under the control of the Spanish, who expelled the Austrians in 1734 and
whose king, Charles III, ordered the tunnels reopened in 1738, looking to the ancient
town as a source of decoration for his own palace, located not far from that of Prince
d'Elboeuf in Portici. A noxious or pestilential emanation, mephitis, fact led to the prema­
ture closing of the excavation tunnels in Herculaneum. In July 1750, Roque Joaquín de Al­
cubierre put the highly competent Karl Weber, a Swiss engineer, in charge of supervising
the excavations.

Keywords: Greek library, Herculaneum, papyrologists, Prince d'Elboeuf, Charles III, mephitis, Roque Joaquín de
Alcubierre, Karl Weber

PAPYROLOGISTS have many archives but only one Greek library; that is, only one find­
spot of literary texts that gives the appearance of having been assembled by one person,
or at any rate by a like-minded group of people with the same literary (in this case, more
specifically philosophical) interest. This library is the collection of rolls found in a richly
endowed villa that had once been on the shore of the bay of Naples in Herculaneum, lo­
cated between Naples itself and Pompeii.1 The quantity and literary value of this one li­
brary alone would make it special; its particular history and the difficulty of reading its
papyrus rolls make Herculaneum even more of a special case For all that the architecture
and sculpture of this villa are of extraordinary quality, it is noteworthy that this building
is today known, indeed named, for its aesthetically far less appealing papyri: the Villa dei
Papiri, this name having ousted the earlier designation, Villa dei Pisoni, even though most
people are willing to accept the conclusion that the villa was built by L. Calpurnius Piso
Caesoninus in the first century BCE and owned by his descendants until the eruption of

Page 1 of 19
The Special Case of Herculaneum

Vesuvius in 79 CE.2 As we shall see, almost without exception these books contain works
of either Epicurus and his school or of Stoics, whom the Epicureans saw as rivals.

The modern recovery of Herculaneum began in 1709, when workers digging a well in the
Italian town of Resina struck a theater many feet below the surface, containing statues
and inscriptions, several of which presented honors to individuals by “the citizens of Her­
culaneum,” Herculanenses, which immediately identified this buried and long-forgotten
town Archaeology may not have yet existed, but the robbing of ancient sites was well es­
tablished. In this case, Prince dʼElboeuf of Austria, (p. 304) the ruling power at this time,
took over from the well diggers in his eagerness to discover more statues with which to
furnish his new villa in the nearby town of Portici (between Herculaneum and Naples); he
thus became the first modern excavator of Herculaneum, but one more in line with min­
ers interested in extracting only gems or ore and with no concern for their surrounding
matrix. In dʼElboeuf's case, the gems were the statues; the matrix the buildings, which
became riddled with tunnels until digging ceased when dʼElboeuf returned to Austria in
1716.

Although in the course of later excavations earlier tunnels were discovered, some of them
perhaps dating to antiquity, it was this early eighteenth-century unearthing of treasures
that brought knowledge of Herculaneum to the awareness of modern Europe. We can
pass over the details of the fits and starts of further excavation (for which see Parslow
[1995]; the story is vividly told for a popular audience by Deiss [1985]). The important
facts are that the excavations came under the control of the Spanish, who expelled the
Austrians in 1734 and whose king, Charles III, ordered the tunnels reopened in 1738,
looking to the ancient town as a source of decoration for his own palace, located not far
from that of dʼElboeuf in Portici. Charles asked Roque Joaquín de Alcubierre, an engineer
in the Spanish army but no better trained than his immediate predecessors in archaeolog­
ical fieldwork, to restart excavations. Since Charles was not interested in the many wall
paintings, Alcubierre too felt free to break through walls in his quest for more portable
treasures. He was famously described by Winckelmann as “this man, who (to use the Ital­
ian proverb) knew as much of antiquities as the moon does of lobsters, has been, through
his want of capacity, the occasion of many antiquities being lost” (Winckelmann 1771, 22).

There was, however, a four-year suspension from 1741 to 1745 because Alcubierre and
others were feeling ill from the air they were breathing under ground. This ailment was
attributed to “mephitic” gases, a now archaic word derived from the Latin mephitis, “a
noxious or pestilential emanation, esp. from the earth” (Shorter Oxford English Dictio­
nary). This mephitis in fact led to the premature closing of the excavation tunnels in Her­
culaneum and the suspension later in the eighteenth century of further work under
ground until well into the twentieth century. Where, however, the town has been opened
to the air, there is no threat to archaeologists or tourists. Open-air excavations were car­
ried on from 1828 to 1835 over most of Herculaneum, but sites still under ground, like
the villa and the theater, remain a threat. (Only the small part of the villa that is exposed
to the air is safe.) Although it was the malodorous gases, especially hydrogen sulfide, that

Page 2 of 19
The Special Case of Herculaneum

drove people out, the air of underground Herculaneum also (still) contains odorless radon
gas, a by-product of radium disintegration.3

In 1750 Alcubierre was transferred to Naples but not before supervising the initial explo­
ration of what was to be the largest villa found in Herculaneum. Since it was set farther
along the seashore from the main body of the town, it was found not by the extension of
the tunnel system but rather by the drilling of yet another well, which indeed found water
about 100 feet below the surface (water that is still (p. 305) seen today alongside the now
partially excavated Villa dei Papiri)—and also a domestic dwelling. Two months later, in
July of 1750, Alcubierre put the highly competent Karl Weber, a Swiss engineer, in charge
of supervising the excavations. Tunnels were directed into the villa, and a second shaft
was dug to allow for quicker entry and exit for the workers, as well as to facilitate the cir­
culation of air and the removal of objects, most notably the bronze statues of family mem­
bers, philosophers, dancers, and mythological characters that now grace the National
Museum in Naples, where one may also see Weber's large architectural rendering of the
villa, although not all of it had (or indeed has yet) been excavated.

The Discovery of the Library

Weber complained that working conditions were “dangerous and abhorrent, if not despi­
cable”;4 nonetheless, statues, wall paintings, and floor mosaics could be recognized im­
mediately for what they were. It is understandable that, working in the poor light and air
in the narrow tunnels, the excavators did not immediately recognize that some of the
blackened objects found in wooden cases and even underfoot were papyrus rolls. Thus, in
the letter sent to King Charles in May just quoted, Weber tallied up the valuable objects
found in the villa with reference to mosaic floors, marble and bronze statues, marble vas­
es, bronze candelabras, and wall paintings; there was not a word about books. Yet just a
few months later, Camillo Paderni—at this point only a royal draftsman hired to make
sketches of wall paintings and other objects in preparation for the engravings that Bayar­
di was to publish in 17525—reported for the first time the presence of books in the villa
(although at the time Paderni was unaware that the many rooms under excavation in this
part of Herculaneum were part of but one large structure) in a letter written to an Eng­
lish acquaintance dated November 18. This letter was later translated and published in
the (London) Philosophical Transactions.6 It reads in part (with original italics and archa­
ic spelling):

It is not a month ago,7 that there have been found many volumes of papyrus, but
turnʼd to a sort of charcoal, so brittle, that, being touched, it falls readily into ash­
es. Nevertheless, by his Majesty's orders, I have made many trials to open them,
butt all to no purpose; excepting some words, which I have picked out intire,
where there are divers bits, by which it appears in what manner the whole was
written. The form of the characters, made with a very black tincture,8 that over­
comes the darkness of the charcoal, I shall here, to oblige you, imitate two short
lines; my fidelity to the king not permitting me to send you any more.

Page 3 of 19
The Special Case of Herculaneum

N. ALTERIUS. DVLC

DEM. CVRIS. CRVDE9

This is the size and shape of the characters. In this bit there are eight
(p. 306)

lines. There are other bits with many other words, which are all preserved in or­
der for their publication.

“There have been found.” This phrase conceals what Paderni reports in a later letter (of
April 10, 1755) that at first the rolls were not recognized for what they were: “In a cham­
ber10 … there has been found a great quantity of rolls, about half a palm long, and round;
which appeared like roots of wood, all black and seeming to be only of one piece. One of
them falling on the ground, it broke in the middle, and many letters were observed, by
which it was first known, that the rolls were of papyrus.” Similar is the later report of
John Hayter to his sponsor, the Prince of Wales (later King George IV):

The manuscripts … which the Director [Paderni] and the equally ignorant, but
clearly guiltless, labourers, mistook for pieces of charcoal, or burned timber, and
which, in consequence, were removed, and applied by them, to the usual domes-
tick purposes. In the course of their removal, however, some detached fragments
happily fell from one, or two of these devoted volumes, and displayed upon their
surfaces very distinguishable characters.11

That is, the dropped roll broke into (at least) two pieces, so that its interior columns, more pro­
tected from the heat than the surface, were visible.
The workers who broke open the roll showed their discovery to Alcubierre, but since
Alcubierre's knowledge of Greek was no better than theirs, he took the rolls in turn to a
local scholar, Canon Alessio Simmacho Mazzochi, who quickly took charge to the extent
that he, along with Paderni, called for the careful and immediate removal of all the rolls
to the royal museum at Portici. Numbers were assigned, but the total reached, about
1,800, includes all the “pieces” found, many of which were merely portions of rolls, so
that often one original roll was (and still is) represented by several P.Herc. numbers.
Thus, Philodemus's On Poems, book 1, has recently been edited from seven separate in­
ventory items: P.Herc. 444, 460, 466, 1013, 1073, 1074, and 1081.12 The villa has yielded
800–1,100 rolls, some of which are but parts (“books”) of works too long for one roll. On
Poems, for example, was in five books.

Unrolling the Books

As Paderni noted in his letter of November 1752, the rolls now stored in Portici were
dried and made brittle by the pyroclastic flow that filled the villa; their outer layers were
usually blackened as well. Attempts to unroll them were unsuccessful, although their po­
tential as a source of new classical texts was clear. Paderni, seeking royal support, as well
as personal glory, and too impatient to wait until (as was soon to happen) a safe method
of unrolling was devised, went before the king and queen with rolls and a broad-bladed

Page 4 of 19
The Special Case of Herculaneum

knife and, “having opened one of the rolls (that (p. 307) is, he cut it with a knife) in their
presence, made them conceive the value of the hidden treasure he had uncovered.”13
Paderni thus illustrated his initial plan for getting to the legible parts of the text; that is,
he sliced many rolls vertically into two parts, a process that crushed into dust all the let­
ters along and on either side of the cut. Unfortunately, many rolls were cut before inven­
tory numbers were assigned, with the result that two halves of the same roll were given
quite different numbers, and the close relationship of the two halves was forgotten or lost
in the archives.

Any sheets (called husk or bark, scorza) that had escaped blackening could be peeled
from the interior (the marrow, midollo; the process is called scorzatura), although not al­
ways easily, as moisture over the years would sometimes cause one layer to cling to the
next.14 Moreover, the act of peeling would often destroy the sheet. At this early stage,
then, it was seen that draftsmen would be needed to copy the exposed text before it was
removed from the half roll. The drawings (disegni) made by these draftsmen, all of whom
were ignorant of Greek, thus became the only record of the text whenever the original
was lost.15

Various other ways of opening the rolls were attempted. Raimondo di Sangro experiment­
ed with mercury, hoping that this heavy liquid would insinuate itself between the layers of
a roll placed vertically in a box and force them apart. Instead, the large drops simply pul­
verized the constraining material. Undeterred, di Sangro completely immersed some rolls
in mercury, with equal lack of success. Next, rosewater (for unfathomable reasons) was
applied, but this proved to be almost as disastrous as the mercury treatment. More rolls
were destroyed. The next experiment was to try to separate the layers with gas, which
may have seemed like a reasonable course, but one wonders why the chemist in charge,
Gaetano la Pira, chose a gas so smelly (hydrogen sulfide?) that it drove people from the
royal palace, which would perhaps have been forgiven had it not atomized the papyri as
efficiently as had the mercury treatment.

Yet another experiment took place: M. Mazzochi put a roll in a bell jar so that the sun
would shine on it, in the hope that, as the heat evaporated the moisture in the roll, the
layers would separate. There was, it seems, some separation, but the steamy atmosphere
also caused the ink to run, rendering the letters so altered that the text was now taken
(no doubt with some sense of excitement) to be Oscan.16 Similar attempts with gas and
steam were again attempted when twenty rolls were brought to England in 1802 and
1816 (to join the disegni that Hayter brought back). The English experimenters, among
whom was the famous chemist Humphry Davy, essentially did little more than repeat the
work done in Italy.17

It is extremely fortunate that King Charles at this point, in 1793, invited Father Antonio
Piaggio to help with the unrolling. Piaggio, then employed by the Vatican library, was well
known for his careful work in the transcribing of illegible texts.18 It was natural that the
skilled and scholarly Piaggio was repelled by the careless and destructive activities of
Paderni and that this soon developed into a dislike of the man himself.19

Page 5 of 19
The Special Case of Herculaneum

Piaggio soon devised a more cautious and effective way of unrolling. Although the outer
parts of many rolls were too friable ever to survive separation, somewhere in (p. 308) its
interior the more flexible midollo seemed more promising. Piaggio realized that a small
but constant pressure could safely separate the top layer from the one below. To apply
this pressure he devised the first of his famous machines: The exposed edge of a midollo
was attached to a thin animal membrane (bladder or gut) known as goldbeater's skin
(since this membrane, when holding a thin layer of gold leaf, was pressed or beaten over
a substance designed to receive a gold surface pattern), which in turn was attached to
long ribbons or strings to be tied to a bar set over the roll, which was allowed to hang
down. The papyrus, whose weight provided the main separative force, was attached to a
roller that could be turned with a screw. As the roll opened to a width of several columns,
pieces would be carefully cut off. This process was repeated until the midollo was opened
to the subscriptio, at which point the work's identity would finally be revealed.20 Such a
slow process produced an obvious need to construct several of Piaggio's machines so that
a number of rolls could be unrolled at the same time. The revealed columns were sent to
the disegnatori.

Once again we have a vivid contemporary description of the process:

It is incredible to imagine what this man [Piaggio] contrived and executed. He

made a machine, with which (by the means of certain threads, which being
gummed, stuck to the back part of the papyrus, where there was no writing), he
begins, by degrees, to pull, while with a sort of ingraver's instrument he loosens
one leaf from the other (which is the most difficult part of all), and then makes a
sort of lining to the back of the papyrus, with exceeding thin leaves of onion [be­
fore he changed to animal membranes] (if I mistake not), and with some spirituous
liquor, with which he wets the papyrus, by little and little he unfolds it. All this la­
bor cannot be well comprehended without seeing. With a patience superior to
what a man can imagine, this good father has unrolled a pretty large piece of pa­
pyrus, the worst preserved, by way of trial…. Father Antonio, after he has loos­
ened a piece, takes it off where there are no letters; and places it between two
crystals for the better observation; and then, having an admirable talent in imitat­
ing characters, he copies it with all the lacunae, which are very numerous in this
scorched papyrus…. The worst is, the work takes up so much time, that a small
quantity of writing requires five or six days to unroll, so that a whole year is al­
ready consumed about half this roll.21

Early Attempts at Reading, Editing, and Pub­

lishing
Success at last: The promise of an ancient library was to be revealed, a collection of man­
uscripts of an age “which outruns the date of any other Manuscripts upon earth” (Hayter
1811, 5). There was, however, a certain sense of disappointment when it soon became

Page 6 of 19
The Special Case of Herculaneum

clear that the bulk of the books seemed to be not the hoped-for lost (p. 309) works of the
famous poets and historians but the rather dry disquisitions written by Philodemus,
known previously almost exclusively from his elegant epigrams preserved in the manu­
scripts of the Greek Anthology. The first four titles to be identified by their subscriptiones
were Philodemus's De Musica IV, Rhetorica II, Rhetorica I, and On Vices and Their Corre­
sponding Virtues. Winckelmann was only the first of many to express his disappointment
in print and to put forward his own wish list of what he would have preferred to see re­
covered from the ashes. Who, he asked, needs another treatise on rhetoric and ethics
when we already have Aristotle? Would it not be better to find lost historical works of
Diodorus Siculus and Theopompus and some plays of Sophocles, Euripides, and Menan­
der (reasonable enough choices), to say nothing of the lost “rules of symmetry composed
by Pamphylius for the use of painters”?22 This last odd choice reflects the personal desire
of an art historian, as wish lists tend to do. The lyric poet William Wordsworth, for exam­
ple, wanted more Pindar and Simonides. Note that almost all the names listed here
(though not Pamphilus) have in fact been augmented by papyrological discoveries, al­
though more from the sands of Egypt than from Herculaneum, where, however,
Philodemus's quotations of classical authors occasionally provide new fragments or better
readings of previously known lines (cf. Delattre 1996).

Although I have more to say about attempts to open and read the rolls from the late eigh­
teenth century to the present, it is at this former point that publication of the papyri be­
gins. Appropriately enough, the first roll to be opened was the first to be printed—in
1793: Philodemus's De Musica IV (=P.Herc. 1497), edited by Carlo Maria Rosini, the bish­
op of Pozzuoli, who provided a learned introduction to Philodemus. This treatise appeared
in the first series of Herculanensium Voluminum quae supersunt (VH).23 Each column was
shown in a full-page engraving made from its disegno, facing a two-column page that
comprised both a printed version of the Greek text (with the lavish touch of having re­
stored letters printed in red to set them off from the more fully preserved letters) and a
Latin translation. Dotted lines indicated lost lines of text. To meet the call for more rapid
publication (a request made by, among others, Carolina Bonaparte, the wife of the then
king of Naples), a second series began in 1862 (VH2). This Collectio Altera also consisted
of eleven volumes and included the texts of sixteen rolls. Without the intensive editorial
work done on the text of the first series, it contained little more than reproductions of the
disegni. Publication ceased in 1876.24

After Piaggio died in 1796, work on the papyri slowed down and then had to cease com­
pletely during the political turmoil of these years. King Ferdinand IV of Naples (and III of
Sicily) and his wife, Maria Carolina (daughter of Empress Maria Theresa of Austria), had
done much to make Naples powerful and cultured, but, in opposing the overthrow of the
monarchy in France, they sadly underestimated Napoleon's power and ambition. Al­
though at peace with France since 1796, they hoped to take advantage of Napoleon's ex­
pedition in Egypt. Encouraged by his wife (p. 310) and with the help of Lord Nelson, Ferdi­
nand invaded Rome but could not hold it for long.25 Nor could they remain in Naples any
longer. In 1798 work on the papyri was suspended completely when Ferdinand and Maria
Carolina, as well as their entire court, fled to Palermo in Sicily in order to escape
Page 7 of 19
The Special Case of Herculaneum

Napoleon's invading forces. Ferdinand and the queen left on one of Nelson's ships, taking
with them all the papyri, which had been packed in sawdust.

Southy (1873, 213–214) describes the scene imaginatively:

Lady Hamilton, like a heroine of modern romance, explored, with no little danger,
a subterraneous passage, leading from the palace to the seaside: through this pas­
sage the royal treasures, the choicest pieces of painting and sculpture, and other
property [including the papyri, too unglamorous to name here] … were conveyed
to the shore, and stowed safely on board the English ships. On the night of the
21st, at half-past eight, Nelson landed, brought out the whole royal family … and
carried them safely, through a tremendous sea, to [his ship] the Vanguard.

The French arrived the next year, when they helped the republican class of Naples estab­
lish the short-lived Repubblica Partenopea, in memory of the land's early Greek name of
Parthenope. The republicans, however, were unable to make a success of their new gov­
ernment. Ferdinand soon regained power and, having already sent the papyri (still cush­
ioned by sawdust in their original packing crates) in 1802, was able to recall his court to
Naples. In 1806 the court again fled to Palermo ahead of Napoleon's invading army. This
time, though, the papyri stayed put, now housed in the Royal Museum in Portici.

As one can see, interest in the newly discovered papyri was such that accounts of them
were published with some frequency in the London press,26 and the famous art historian
Winckelmann incorporated an extensive account of the papyri in his open letter of 1771.
More scholarly interest is demonstrated by the works of Schütz (1795) and von Murr
(1806), only the first of a long line of German scholars, who dominated Herculaneum
studies well into the twentieth century.27 For example, for decades the Bibliotheca Teub­
neriana published the only or the most thorough edition of Rhetoric, De Ira, De Musica 4,
On Vices 10, De Oeconomia, De Rege Bono secundum Homerum, and De Libertate Dicen­
di. Similarly C. Jensen's edition of Philodemus's On Poems 5 (1923) and H.Diels's of
Philodemus's On Gods 1 and 3 (1916, 1917) offered texts that influenced much later
scholarship on these subjects.

After a period in which little was done on the papyri themselves, Marcello Gigante almost
single-handedly revived interest in these texts when he actively invited scholars from
every country to edit and reedit the papyri in the Officina dei Papiri, housed in the Na­
tional Library in Naples, and began the still ongoing series La Scuola di Epicuro (1978–).
Furthermore, the new journal he and others edited, Cronache Ercolanesi (see CErc in the
bibliography), published many texts too small for book format, most notably fragments of
Epicurus's own On Nature.

(p. 311) The Contents of the Library

As mentioned, the bulk of the discoveries proved to be Epicurean texts, and of these in
turn the majority belong to Philodemus. With the most minor of exceptions, all the texts

Page 8 of 19
The Special Case of Herculaneum

were previously lost; indeed, apart from Epicurus's On Nature, even their existence was
unknown. With regard to Philodemus, only his epigrams were extant, and Cicero told of
his acquaintance with Lucius Calpurnius Piso Caesoninus, consul in 58 BCE and the fa­
ther-in-law of Julius Caesar. The only title ascribed to him in the scanty testimony is his
Syntaxis of Philosophers, some parts of which have been found in the villa. A bare but
comprehensive list of the Epicurean material includes the following, all, however exigu­
ous, identified by subscriptiones:28 Epicurus, On Nature;29 Metrodorus, On Wealth;
Carneiscus, Philistas; Colotes, Against Plato's Euthydemus, Against Plato's Lysis; Polystra­
tus, On Irrational Contempt [for Popular Opinion], On Philosophy; Demetrius the Lacon­
ian, On Geometry, On Poems, On Some of Epicurus' Opinions, On Some Investigations in­
to Daily Living, Against the Problems of Polyaenus; Philodemus, On Lives(?) and Morals,
On Epicurus, On Piety, On Death, On Gods, On Vices, On Flattery, On Music, On Conver­
sation, On Anger, On Frankness of Speech, On Wealth, On Poems, On Rhetoric, On Signs,
On the Good King according to Homer, Pragmateiai, On the Stoics, On Gratitude, Syn­
taxis of Philosophers.

Some few Stoic works have also been found, as might be expected of a library that was
formed by an Epicurean writer, who would want to be well informed of his philosophical
rivals; indeed one wonders why more Stoic texts have not been found. Known so far are
Chrysippus, Logical Investigations, On Providence; and Zeno, Craterus.30

A few Latin works have also been found. In addition to the epic verses on (at least in the
preserved columns) the battle of Actium (see note 9), scraps of Lucretius and Ennius have
recently been published, as well as Money Lender by the comic poet Caecilius Statius.

It is not easy within the limits of this chapter to assess the contribution this library has
made to current knowledge of Hellenistic philosophy.31 Surely, though, one of its principal
contributions has been in the area of ancient literary theory, which in the past had to
jump from Aristotle's Poetics and Rhetoric to Horace's Ars Poetica. With the eventual pub­
lication of several books of Philodemus's On Poems (largely with that of book 5 by Jensen
[1923]), Philodemus's regular practice of rehearsing his opponents' theories before criti­
cizing them laid bare an entirely new world of literary theories, much of which had been
silently absorbed by Latin authors, such as Horace in his literary epistles.32

The relatively small number of Latin texts raises the question of the nature of the original
collection. Was it primarily the working library of an Epicurean philosopher, most likely
Philodemus himself, as suggested by the number of (p. 312) duplicated rolls?33
Philodemus, however, died well over a century before the eruption of Mount Vesuvius,
and this library was located in the largest villa in Herculaneum, which surely belonged to
a Roman who would have had some favorite Latin readings handy, to say nothing of nec­
essary household archives. More Latin papyri may yet be found among the still unopened
rolls, but the laws of probability predict that the breakdown of these will reflect that of
those already opened.

Page 9 of 19
The Special Case of Herculaneum

Two possibilities suggest themselves: first, that along with other portable goods the fa­
vorite books of the owner of the villa in 79 CE were removed before the pyroclastic low
arrived; second, that the so-called Latin library remains unexcavated in the villa. Support­
ing the first possibility is that the signs of hasty removal of objects includes even the
books, some of which were found not in the rooms and shelves where others were found
but in the nearby hallways. The owner or his major domo might well have engaged in a
form of triage in which the books considered most valuable (the texts wished for by those
people disappointed by Philodemus) were saved. The historical irony is that had they not
been “saved,” that is, taken away from Vesuvius, they would in fact have been saved by
the tufa that has preserved the unique copies of the philosophy library.

Supporting the second possibility is that all the papyri were found quite close to a corner
of the house that was never excavated. That is, Weber's famous plan shows an empty area
along the upper right-hand corner of an atrium, where the books were found. Atria, how­
ever, are surrounded by rooms and thus cannot form the outer edge of the villa, as a care­
less glance at the plan might indicate. Moreover, recent excavations have revealed two
previously unknown stories below the level on which the statuary and papyri were found.
Here too there is the possibility of a Latin library, which Romans tended to keep separate
from their Greek volumes. In this chapter it is appropriate to do no more than mention
the current debate between those who favor further excavations (largely papyrologists)
and those who argue for maintaining the current state of excavations, which are deterio­
rating from exposure to sunlight, air, and pollution. Indeed, one well-known archaeologist
has been quoted as saying, “I am almost indifferent on the subject of the papyri.” The bat­
tle will play itself out in the coming years.

Any future papyrus finds would also help define the nature of the villa's library. At the
present, too many possibilities present themselves. The technical nature of these treatis­
es, including annotated and duplicate copies, strongly suggests the working library of an
Epicurean philosopher. The preponderance of works written by Philodemus further sug­
gests that it was Philodemus's own library. How such a working library made its way from
Philodemus's modest dwelling (as he himself describes it in an epigram) to the grandest
building in Herculaneum is most easily answered if we accept the view that the villa was
owned by a descendant of the Piso who has several secure links with Philodemus.34 That
is, leaving out the necessary provisos, Philodemus dies before Piso, who receives
Philodemus's library into his (p. 313) villa (already decorated with busts of Epicurus and
other philosophers, to say nothing of the famous jumping porcellino (famous enough to
maintain this name, even though porcellina would be more accurate) with its inescapable
Epicurean coloring. Piso's heirs add nothing to this philosophical library, which suggests
that their interest did not match that of Piso Caesoninus.35 They would, either out of piety
or simple neglect, have held on to the collection during the several generations between
Piso's death, ca. 40 BCE, and the eruption of Vesuvius in 79 CE.

The collection of Greek philosophical texts thus forms a well-defined library that was
quite likely gathered by Philodemus. The palaeographical investigations of Guglielmo
Cavallo (1983) have shown that it comprises rolls written as early as the late second cen­

Page 10 of 19
The Special Case of Herculaneum

tury BCE (ibid., 51), which form Cavallo's “gruppo A,” seven rolls (in nine P.Herc.
numbers), each containing one or another book of Epicurus's On Nature (ibid., 28–29). If,
as we have been imagining is the most likely case, they belonged to Philodemus, he would
have obtained them in Athens during his study there and carried them along with him on
his travels (which may have included Rhodes, Alexandria, and Himera) before arriving in
Herculaneum. All in all Cavallo identifies seventeen groups (A–R [no J]), plus a number of
“scritture varie,” which do not fit easily into any of the groups. They are distinguished on
palaeographical grounds and arranged in assumed chronological order. Cavallo further
identifies thirty-four different scribes (Anonimi I–XXXIV). Since all the hands are profes­
sional, none can be the once-sought-for “mano di Filodemo.” The last groups show the in­
fluence of Latin writing, not surprising in a society where Latin was the predominant lan­
guage. Even group R, however, seems to date to the end of the first century BCE and the
beginning of the next century, which is consistent with the picture painted earlier, that
the philosophical library was essentially formed over the working life of Philodemus,
whose death, on the basis of circumstantial evidence alone, is regularly placed in the
neighborhood of 40 BCE, at about age seventy, although he may well have lived for anoth­
er decade. Someone would have added a few works by other Epicureans (there are no
works by Philodemus in group R), but essentially the collection was closed at the begin­
ning of the millennium.

At present, no new rolls are subjected to any technique of unrolling, whether by machine,
mercury, gas, or steam. Progress, however, is still being made on already opened texts,
even those that have already been published. In part these advances are due to research
in the archives that leads to the recognition that disparately numbered pieces were once
part of the same roll, and in part, binocular microscopes have enabled us to make better
readings. Far more impressive than the latter, however, are the readings made possible in
recent years by the use of multispectral imaging (MSI), which can distinguish black let­
ters from their almost equally black background when the sheets are photographed
through filters in the infrared range. Thus, some sheets that are completely black to the
eye in ordinary light, even through a microscope, can now be read as easily as uncharred
papyri. (MSI has also proved of value in reading the charred Derveni and Petra papyri.)

The future holds even greater promise. Computer-assisted tomography, developed

(p. 314)

to view the interior details of living bodies, will be able virtually to unwind a roll by distin­
guishing each layer from its sovro-and sottoposti, with none of the physical loss inherent
in almost all earlier unrolling and scorzature. However slowly this high-tech version of a
Piaggio machine proceeds, we now expect that all eight hundred or so unopened rolls will
be revealing their secrets. What texts can we expect? A well-known scholar in this field is
anecdotally reported regularly to have asked the nonresponsive Epicurean gods for “non
piu di Filodemo.” His spirit is likely to be disappointed.

Page 11 of 19
The Special Case of Herculaneum

Bibliography
The Herculaneum papyri are so extraordinarily well documented that even their bibliog­
raphy needs an introductory paragraph. For a physical description of the papyri them­
selves, see note 28. A briefer but more up-to-date guide to current editions is given at
http://www.herculaneum.ox.ac.uk/books, which also cites editions of passages gath­
ered from the papyri under useful rubrics (e.g., Angeli's edition [1994] of Epicurus's frag­
mentary epistles). This site also has a link to the home page of the Herculaneum Society,
which contains information about Web sites that depict both the papyri themselves and
disegni. Mcllwaine (1988, 2000) offers a bibliography that includes all aspects of Hercula­
neum. The most comprehensive and continuing source of new editions, as well as articles
on the history of the collection, is Cronache Ercolanesi (CErc). Moreover, the back pages
of CErc regularly publish work in progress, thus providing a guide to future bibliography
on the papyri. The best overall introduction to both the papyri and their history is Capas­
so (1991), but much original work in the archives continues to be done, much of it by,
alone and together, Blank and Longo Auricchio; see, for example, Blank and Longo Auric­
chio (2004). Other useful bibliographical aids are Dorandi (1989), an index to all the
philosophers (and other thinkers) named in the papyri, and Delattre (1996).

Angeli, A. 1994. “Lo svoglimento dei papiri carbonizzati.” In Il Rotolo librario: Fabbri­
cazione, restauro, organizzazione interna, ed. M. Capasso, 37–103. Galatina: Congedo.

Arrighetti, G. 1971. “LʼOpera Sulla Natura di Epicuro.” CErc 1: 41–56.

Asmis, E. 1990. “The Poetic Theory of the Stoic ‘Aristo.’ ” Apeiron 23: 147–201.

———. 1992a. “Crates on Poetic Criticism.” Phoenix 46: 138–169.

———. 1992b. “Neoptolemus and the Classification of Poetry.” Classical Philology 87:
206–231.

Bayardi, O. A. (1752). Prodromo delle antichità dʼErcolano. 5 vols. Naples: Stamperia

Reale.

Blank, D., and F. Longo Auricchio. 2004. “Inventari antichi dei papiri ercolanesi.”
(p. 318)

CErc 34: 39–152.

Broggiato, M. (2001). Cratete di Mallo: I frammenti. La Spezia, Italy: Agorà.

Camodeca, G., and G. Del Mastro. 2002. “I papiri documentari ercolanesi (P.Herc. MAN):
Relazione preliminare.” CErc 32: 281–296.

Capasso, M. 1986. “Carlo Maria Rosini e i papiri ercolanesi.” In Carlo Maria Rosini (1748–
1836): Un umanista flegreo fra due secoli, ed. S. Cerasuolo, M. Capasso, and A.
DʼAmbrosio, 129–192. Pozzuoli, Italy: Azienda autonoma di cura, soggiorno e turismo di
Pozzuoli.

———. 1989. “Primo supplemento al Catalogo dei Papiri Ercolanesi.” CErc 19: 193–264.
Page 12 of 19
The Special Case of Herculaneum

———. (1991). Manuale di papirologia ercolanese. Galatina, Italy: Congedo.

———, ed. (1994). Il Rotolo librario: Fabbricazione, restauro, organizzazione interna.

Galatina, Italy: Congedo.

———, ed. 1995. “Le tavolette della villa ercolanese dei papiri.” In his Volumen: Aspetti
della tipologia del rotolo librario antico, 111–117. Naples: Procaccini.

———, ed. (1997). Bicentenario della morte di Antonio Piaggio: Raccolta di Studi.
Papyrologica Lupiensia 5. Galatina, Italy: Congedo.

Cavallo, G. (1983). Libri scritture scribi a Ercolano: Introduzione allo studio dei materiali
greci. CErc 13 (Suppl.). Naples: Gaetano Macchiaroli.

Cerasuolo, S., M. Capasso, and A. DʼAmbrosio, eds. (1986). Carlo Maria Rosini (1748–
1836): Un umanista flegreo fra due secoli. Pozzuoli, Italy: Azienda autonoma di cura, sog­
giorno e turismo di Pozzuoli.

Comparetti, D., and G. De Petra. (1972). La Villa Ercolanese dei Pisoni: I suoi monumenti
e la sua biblioteca. Turin: E. Loescher, 1883; repr., Naples: Centro Internazionale per lo
Studio dei Papiri Ercolanesi.

Deiss, J. J. (1985). Herculaneum, Italy's Buried Treasure, rev. and updated ed. New York:
Harper & Row.

Del Mastro, G. 2000. “Secondo supplemento al Catalogo dei Papiri Ercolanesi.” CErc 30:
157–242.

Delattre, D. 1996. “Les Mentions de titres dʼœuvres dans les livres de Philodème.” CErc
26: 143–168.

Diels, H. (1970). Philodemos. Über die Götter: erstes und drittes Buch. Berlin: Verlag der
Königl. Akademie der Wissenschaften, 1916, 1917; repr. Leipzig: Zentralantiquariat der
Deutschen Demokratischen Republik.

Dorandi, T. 1989. “Testimonianze ercolanesi.” In Corpus dei papiri filosofici greci e latini,
Parte 1, Autori noti. Vol. 1, 1–78. Florence: Leo S. Olschki.

———. (1991). Filodemo, Storia dei Filosofi: Platone e lʼacademia (P.Herc. 1021 e 164).
Naples: Bibliopolis.

———. (1993). Theodor Gomperz: Eine Auswahl herkulanischer kleiner Schriften, 1864–
1909. Leiden: Brill.

———. 2001. “Supplemento ai Supplementi al Catalogo dei Papiri Ercolanesi.” ZPE 135:
45–49.

Drummond, W., and R. Walpole. (1810). Herculanensia. London: T. Cadell & W. Davies.

Page 13 of 19
The Special Case of Herculaneum

Gaiser, K. (1988). Philodems Academica: Die Berichte uber Platon und die Alte Akademie
in zwei herkulanensischen Papyri. Stuttgart—Bad Cannstatt: Frommann-Holzboog.

Garuti, G. (1958). C. Rabirius: Bellum Actiacum e papyro Herculanensi 817. Bologna: N.

Zanichelli.

Gigante, M. 1979a. Catalogo dei papiri ercolanesi. Naples: Bibliopolis.

(p. 319) ———. 1979b. Civiltà delle forme letterarie nellʼantica Pompei. Naples: Bibliopo­
lis.

———. (1988). La Germania e i papiri ercolanesi. Heidelberg: Carl Winter Universitätsver­

lag.

———. (1995). Philodemus in Italy: The Books from Herculaneum, trans. D. Obbink. Ann
Arbor: University of Michigan Press.

———. (2001). Die deutsche Forschung uber die herkulanensischen Papyri in den letzten
drei Jahrzehnten. Nachrichten der Akad. d. Wissenschaften in Göttingen. Philol.-hist. Kl.
Nr. 11. Göttingen: Vandenhoeck & Ruprecht.

Hayter, J. (1811). A Report upon the Herculaneum Manuscripts in a Second Letter, Ad­
dressed, by Permission, to His Royal Highness the Prince Regent. London: Richard
Phillips.

Janko, R. (2000). Philodemus: On Poems, Book One. Oxford: Oxford University Press.

Jensen, C. (1923). Philodemos über die Gedichte, Fünftes Buch. Berlin: Weidmann.

Longo Auricchio, F., and M. Capasso. 1980. “Nuove accessioni al dossier Piaggio.” In Con­
tributi alla Storia della Officina dei Papiri Ercolanesi, 17–59. Naples: Industria TipograW­
ca Artistica.

Mangoni, C. (1993). Filodemo: Il quinto libro della Poetica (P.Herc. 1425 e 1538). Naples:
Bibliopolis.

Mansi, M. G. 1997. “Per un profilo di Camillo Paderni.” In Bicentenario della morte di An­
tonio Piaggio: Raccolta di Studi. Papyrologica Lupiensia 5, ed. M. Capasso, 77–108.
Galatina, Italy: Congedo.

Mattusch, C. C. (2005). The Villa dei Papiri at Herculaneum: Life and Afterlife of a Sculp­
ture Collection. Los Angeles: J. Paul Getty Museum.

Mcllwaine, I. C. (1988). Herculaneum: A Guide to Printed Sources. 2 vols. Naples: Bib­

liopolis.

———. 1990. “Herculaneum: A Guide to Printed Sources: A Supplement.” CErc 20: 87–
128.

Page 14 of 19
The Special Case of Herculaneum

Murr, C. G. von. (1806). Philodem von der Musik. Berlin: H. Frölich.

Obbink, D. 1996. Philodemus: On Piety. Part 1: Critical Text with Commentary. Oxford:
Oxford University Press.

Parslow, C. C. (1995). Rediscovering Antiquity: Karl Weber and the Excavation of Hercula­
neum, Pompeii, and Stabiae. Cambridge: Cambridge University Press.

Schutz, C. G. (1795). Proposita in Philodemi ∏∊ρì μουσικῆς librum IV. Jena: Ex Officina
Goepferdtii.

Sedley, D. 1973. “Epicurus, On Nature, Book XXVIII.” CErc 3: 5–83.

———. 1974. “The Structure of Epicurus' On Nature.” CErc 4: 89–92.

Sickler, F., and T. Tyrwhitt. (1817). Herculaneum Rolls: Correspondence relative to a

Proposition made by Dr. Sickler of Hildburghausen, upon the Subject of their Develop­
ment. London: J. Barfield.

Sider, D. 1997. The Epigrams of Philodemos: Introduction, Text, and Commentary. New
York: Oxford University Press.

Southey, R. (1873). Life of Nelson. London: G. Bell and Sons.

Störmer, F. C. et al. 1990. “Ink in Herculaneum.” CErc 20: 183.

Winckelmann, J. J. (1771). Critical Account of the Situation and Destruction by the First
Eruptions of Mount Vesuvius of Herculaneum, Pompeii, and Stabia … London: T. Carnan
and F. Newbery.

Wojcik, M. R. (1986). La Villa dei Papiri ad Ercolano: Contributo alla ricostruzione

dellʼideologia della nobilitas tardorepubblicana. Rome: LʼErma di Bretschneider.

Notes:

(1.) These texts will be our exclusive concern. A documentary archive was found else­
where in Herculaneum; cf. Camodeca and Del Mastro (2002). Other writing was pre­
served on wax tablets and in the form of graffiti; for the former see Capasso (1995), and
for the latter see Gigante (1979b).

(2.) For the sculpture of the villa see Mattusch (2005); for its architecture see Wojcik
(1986). A still useful volume on all aspects of the Villa, including the papyri is Comparetti
and De Petra (1972). Note also Mcllwaine (1988,1990). The journal Cronache Ercolanesi,
although devoted primarily to the subject of the papyri, remains a source of information
about the site of ancient Herculaneum in general. The question of ownership of the villa
need not be discussed here; cf. Capasso (1991, 43–64).

Page 15 of 19
The Special Case of Herculaneum

(3.) I am grateful to Valerio Papaccio, the architect at Pompeii and Herculaneum, for in­
formation on this matter. The Romans prayed to the personified Mephitis to avert pesti­
lential exhalations of all sorts. The name Mephistopheles, invented for sixteenth-century
tellings of the Faust legend, doubtless owes its origin to this Latin word.

(4.) Letter of May 22, 1752, Archivio di Stato, Napoli. Fondo Casa Reale Antica 1539, Inc.,
44, translated by C. Parslow (1995, 91).

(5.) Later Paderni was to become the king's curator of antiquities and then, more formal­
ly, director of the Museum Herculanense, part of the royal palace in Portici, where the pa­
pyri and other antiquities were displayed. A number of these wall paintings are of interest
here because they portray papyrus rolls (some of them with nonsense letters), inkwells,
pens, capsae, and sillyboi, as well as people reading from papyrus rolls. For Paderni, see
Comparetti and De Petra (1972, 238–250) and Mansi (1997). Today almost all the papyri,
along with those disegni (see below) not taken to Oxford, are housed in the Officina dei
Papiri as part of the Biblioteca Nazionale, located in the Palazzo Reale di Napoli.
Napoleon took a few papyri to Paris (P.Herc.Paris; they have since been returned to
Naples), others went to Oxford, and one piece is in Copenhagen (P.Herc.Haun.). Images of
those Oxford papyri that were not destroyed by Sickler's attempt at unrolling (discussed
later) are available online; see the introduction to the bibliography. Naples, therefore, has
always been the chief place of study for these papyri, although modern imaging tech­
niques now allow much valuable work to be done at a distance.

(6.) Philosophical Transactions 48(1) (1754) and subsequently reprinted in Comparetti

and De Petra (1972, 238).

(7.) That is, October 19, 1752, for which date the contemporary excavation report reads:
“mi fu consegnato n. 4 volumi di papiri, il tutto trovato al pozzo di Ceci”; cf. Comparetti
and De Petra (1972, 155).

(8.) On the composition of the ink cf. Störmer (1990).

(9.) Interestingly, although the vast majority of the rolls found in the villa contain Greek
texts, Paderni chose to quote from one of the few Latin papyri. Although this particular
fragment has been lost, it can reasonably be identified as coming from the Latin epic po­
em that describes, at least in the extant remains, the battle of Actium, P.Herc. 817, first
published in the second series of Herculanensium Voluminum quae supersunt (1809). If
this is true, the dots here printed on the line represent the raised interpuncts of this pa­
pyrus. Cf. Garuti (1958, xv, 69). None of the various attempts to assign an author to this
poem has won wide acceptance.

(10.) The findspots are discussed later.

(11.) Hayter (1811, 30). A modern papyrologist would find this early description of Greek
papyri fascinating reading. Apart from its firsthand account of the ambitions and easily
bruised egos (including Hayter's own) of the men charged with opening the papyri, there
is also the attraction of reading of “chasms” in the papyri, which Hayter himself later ex­
Page 16 of 19
The Special Case of Herculaneum

perimentally calls “lacunae” (Hayter in fact puts quotation marks around the word). While
Hayter consistently refers to the texts as “manuscripts,” he does take note of the word
the Italians used: “papiro.”

(12.) See Janko (2000, 86–114, fig. 2). Similarly, note Obbink's account of the several of
the inventoried pieces that he used to edit Philodemus's On Piety (Obbink 1996, 37–53),
which also describes Obbink's discovery (made independently by Daniel Delattre while
working on Philodemus's De Musica) that, since the layers had been numbered as they
were removed (and drawn) from the inside of the roll outward, they were, after this pro­
cedure had been forgotten, thereafter read in this order, even though the layers closest to
the middle are those at the conclusion of the text on any particular roll. The rereading of
the papyri according to the original order has come to be known as the Obbink-Delattre
method. The situation is made even worse when, as often happened, the papyrus roll had
been cut down the middle by Paderni, so that the reconstruction of the text calls for alter­
nating between two or more P.Herc. numbers.

(13.) Part of Piaggio's memoir, written between 1769 and 1771: Societa Napoletana di
Storia Patria Ms. 31-C-21. Cf. Parslow (1995, 103–106).

(14.) Modern scholarship in whatever language has maintained many of these now-techni­
cal terms such as scorza and midollo; others are given as appropriate.

(15.) Those that remain in Naples are designated N; those that (see later discussion) were
removed to Oxford are designated O. Thus, N and O are treated in apparatus critici as
primary witnesses. Furthermore, a new critical sign, a sublinear asterisk (a͙), was intro­
duced into edited Herculaneum texts to indicate those letters the editor believed had
been misread by the disegnatore. Thus, for example, Obbink (1996, 184) on De Pietate
1133 prints τῶι͙ where N has των.

(16.) Winckelmann (1771,111); Capasso (1986, 135–137).

(17.) See Drummond and Walpole (1810). Unfortunately many of these did not survive the
“development” attempted by Friedrich Sickler, whose initial communications with the
Englishman Thomas Tyrwhitt on the terms on which he would work are available in Sick­
ler and Tyrwhitt (1817); see Comparetti and De Petra (1972, 80–81) for a somewhat horri­
fied account of Sickler's “Dummheit.”

(18.) On Piaggio, see Longo Auricchio and Capasso (1980).

(19.) His manuscript account of Paderni's activity is well described and quoted by Parslow
(1995, 103–106).

(20.) Subscriptiones often contain column and/or line totals and occasionally kollemata
counts as well.

(21.) Anonymous in Comparetti and De Petra (1972, 245–246), reprinted from the London
Philosophical Transactions (1795), which credits the account to “a learned Gentleman of

Page 17 of 19
The Special Case of Herculaneum

Naples.” An even earlier and more detailed account of Piaggio's machine is too long to
quote here: Winckelmann (1771, 112–114). Angeli (1994) describes all these early at­
tempts at unrolling, as well as later ones.

(22.) Winckelmann (1771, 116–117). For writings on art by the fourth-century BCE
Pamphilus, cf. Suda s.v. Πάμφιλος Ἀμφιπολίτης.

(23.) On Rosini, see Cerasuolo, Capasso, and DʼAmbrosio (1986). Publication of this series
reached volume eleven in 1855. (Volume seven was never published; volume five ap­
peared in two parts).

(24.) VH3, Collectio Tertia, consisted of only one volume and was published in 1914, short­
ly before World War I.

(25.) Lord Nelson was actively engaged in warfare with Napoleon's forces throughout this
entire period, much of the time in the Mediterranean, where he often gave support to the
kingdom of Naples, indeed sometimes to an extent greater than that desired by the Eng­
lish crown, led as he was by his infatuation for Queen Maria Carolina (and for Emma
Hamilton, who may have been encouraged in this affair by the queen).

(26.) Most notably in the Gentleman's Magazine, which between 1804 and 1820 pub­
lished thirteen accounts of the new discoveries.

(27.) Note, among other things, the many kleine Schriften of Theodor Gomperz dedicated
to these texts, only some of which were collected in Dorandi (1993). For a survey of more
recent German scholarship, see Gigante (1988), supplemented by Gigante (2001).

(28.) For the P.Herc. numbers of each title, physical description, Greek titles, book num­
bers, and column and line numbers, see Gigante (1979a, 44–55), which also contains a
complete bibliography of each, the last of which is supplemented by Capasso (1989) and
Del Mastro (2000); see also Dorandi (2001). The following list does not include a number
of substantially preserved works whose titles scholars have only (however reasonably)
guessed at, such as P.Herc. 831, Demetrius Peri Meteorismou, or whose usual title in
scholarly literature derives from the content rather than from the subscriptio, such as
P.Herc. 1424, Philodemus Oeconomicus, where in fact the subscriptio has On Vices and
Their Corresponding Virtues. Nor does this list correlate the various other titles that
seem to belong to this last complex rubric, which in addition to Oeconomicus also com­
prises Flattery, [Arrogance], and [Greed].

(29.) Books 11, 14, 15, 25, 28, 34, and an unknown book (on time); on this work see Ar­
righetti (1971) and Sedley (1973, 1974).

(30.) In this general context one has to distinguish between Zeno of Citium (Philodemus's
Epicurean teacher) and the Stoic Zeno of Sidon.

(31.) A brief overview appears in Gigante (1995, 15–48).

Page 18 of 19
The Special Case of Herculaneum

(32.) Jensen (1923) has been superseded by Mangoni (1993). Reconstructing the views of
some of Philodemus's opponents are the articles by Asmis (1990, 1992a, 1992b). On
Crates, see Broggiato (2001).

(33.) For example, Gaiser (1988) and Dorandi (1991) are each editions of P.Herc. 1021
and 164, parts of Philodemus's Syntaxis of Philosophers, which overlap somewhat where
Plato's Academy is treated.

(34.) The links between Philodemus and Piso Caesoninus are (1) Cicero's blaming Piso's
moral failings in large part on his weakness for the hedonism of Epicureanism, which he
learned from Philodemus (In Pis. 68–72); (2) Philodemus's dedicating his On the Good
King according to Homer to Piso; and (3) Philodemus's addressing Piso directly in one of
his epigrams, in which he invites him to take part in an Epicurean celebration. See Sider
(1997, 5–8). Nonetheless, despite what might have been expected, there is no literary, pa­
pyrological, or epigraphical evidence that associates any Piso with the town of Hercula­
neum.

(35.) This is not to suggest that they were any the less cultured; Piso Caesoninus's son L.
Calpurnius Piso Frugi Pontifex and his two sons are the most likely dedicatees of Horace's
Ars Poetica. On this reconstruction of the library's history, Piso may have simply inherited
a library so technical that he himself did not read anything more than the Good King,
which was dedicated to him (and is also relatively undemanding).

David Sider

David Sider is Professor of Classics, New York University.

Page 19 of 19
Education in the Papyri

Education in the Papyri

Raffaella Cribiqre
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0014

Abstract and Keywords

This article focuses on Greek education during the roughly ten centuries between the
conquest of Egypt by Alexander of Macedon and the Arab conquest. Egypt has offered a
large quantity of educational material that permits us to glimpse the everyday, unexcep­
tional practices of schooling and to observe certain details. This educational material is
extant on papyrus, ostraca, wooden and waxed tablets, and, more rarely, parchment. The
recent discovery in Alexandria of eighteen or more classrooms (auditoria) used in late an­
tiquity for higher education is tantalizing. In this case, the literary tradition converges
with the archaeological findings to spotlight a formal school setting used by grammari­
ans, sophists, and teachers of philosophy. There are three divisions of levels of schooling:
basic reading, writing, and numeracy; grammatical and orthographical knowledge of the
language; and perfecting oral and written expression.

Keywords: Greek education, Alexander of Macedon, Arab conquest, Egypt, Alexandria, auditoria, grammatical
knowledge

THE last twenty years have witnessed a renewed interest in literacy and education in the
Greek, Roman, and Byzantine worlds. Previously, H.-I. Marrou's history of education
(1975; first edition 1948) and the study of Roman education by S. Bonner in 1977 were
the authoritative works in this field. Both authors used the papyri to a limited extent and
only to confirm the ancient literary accounts of education. They believed the evidence
from Egypt reflected a pallid image of the highly literary practices in Greece and Rome.
In the past two and a half decades, some of the new research, which has focused on the
role of education and orality in classical Athens (e.g., Thomas 1992) and on literacy in the
ancient world (Harris 1989), has taken the papyri into limited account. In 1996 Raffaella
Cribiore produced a rigorous study of school exercises in Hellenistic, Roman, and early
Byzantine Egypt that included a catalogue of exercises and extensive photographic docu­
mentation. The book on education by Teresa Morgan (1998) was based on this body of
material.1 Cribiore complemented her previous study with another, published in 2001,
that also took into account school texts and rhetorical exercises on papyrus.

Page 1 of 19
Education in the Papyri

In what follows, I focus on Greek education during the roughly ten centuries between the
conquest of Egypt by Alexander of Macedon and the Arab conquest. One might question
the legitimacy of covering this vast period as a continuum since it has long been recog­
nized that, from the socioeconomic point of view, Greek and Roman Egypt differed signifi­
cantly. Did education undergo any changes during these ten centuries? Can one be indict­
ed for adopting a methodology similar to that of earlier scholars, who placed evidence in­
to the neat categories derived from the literary (p. 321) sources? Some changes did in­
deed occur, but they neither warrant a different periodization nor significantly affect the
essentially “frozen” quality of education in Egypt as in other Roman provinces (Cribiore
2001b, 8–9).

While literary and anecdotal traditions alone reveal what we know of Greek education
elsewhere, Egypt has offered a large quantity of educational material that permits us to
glimpse the everyday, unexceptional practices of schooling and to observe certain details.
The sands of Egypt have preserved school exercises written by students and teachers and
some texts that were used in schools. This educational material is extant on papyrus, os­
traca, wooden and waxed tablets, and, more rarely, parchment. In addition, information
about ancient schools (and learning environments generally) emerges from the papyri
and from findspots of exercises and other archaeological material (ibid., 15–44). When re­
ferring to “schools” in antiquity, we must be open to all scenarios because of the diversity
and frequent lack of formality in schooling not only in villages but also in urban environ­
ments. While the papyri have transmitted the names of a considerable number of teach­
ers who are identified only by their professional title (Cribiore 1996, 161–170), direct ref­
erences to schools are more infrequent because of the lack of formal settings. Thus, the
recent discovery in Alexandria of eighteen or more classrooms (auditoria) used in late an­
tiquity for higher education is tantalizing. In this case, the literary tradition2 converges
with the archaeological findings to spotlight a formal school setting used by grammari­
ans, sophists, and teachers of philosophy.

The literary sources indicate a strict division of levels of schooling. This remains largely
valid, provided that one recalls that there were no fixed age limits for admission to (or
graduation from) a certain level and that education at the primary level depended much
on circumstances. The aims of the first stage were to teach basic reading, writing, and
numeracy. The second-level teacher, the grammarian, trained students to read literary
texts (particularly the poets) fluently, and reinforced grammatical and orthographical
knowledge of the language. In schools of rhetoric, young men of the elite read prose (the
orators and historians in particular), continued to study some poetry, and perfected their
oral and written expression. These three stages formed what the ancients called the
enkyklios paideia, that is, the “complete education,” which enveloped those privileged
young men who had access to it until the end.3

Page 2 of 19
Education in the Papyri

Identifying a School Exercise

By “school exercise,” I refer to students' work written in school or for school at any of the
three educational levels and to teachers' preparations for their classes. In the category
“school texts” I include works of literature and commentaries that appear to have been
used in school contexts. Several factors often simultaneously contribute to identifying a
school exercise: writing materials; features such as lines, (p. 322) decorations, punctua­
tion, and lectional signs; and types of textual material. These characteristics might point
to an exercise even in the absence of indications from palaeography, but one can reach
virtual certainty when the hand is a clumsy “school hand” or an accomplished “teacher's
hand.”

Writing Materials Used in School

The writing materials of school exercises do not differ significantly from those generally
used in Egypt (which are covered in chapter 1), but the ways students employed them
show some peculiarities (Cribiore 1996, 57–74). Parchment was used sparingly, unlike pa­
pyrus, ostraca, and tablets. Both students and teachers wrote most of their exercises on
papyrus. Students did not have much chance of getting large, unused papyrus of good
quality. School papyri are often thick and rough and of mediocre if not altogether poor
quality; they show marks of damage, such as missing fibers and clumsy attempts to repair
them with patches. To have a papyrus for an exercise, a student might wash of writing or
cut an unwritten part from a larger written piece. Students at elementary levels did not
need much writing space and could make use of blank areas on previously used papyri. At
higher levels of education, however, students wrote their work across the fibers on the
back (verso) of papyri whose fronts (recto) already bore writing. Unlike beginning writ­
ers, they did not need to follow the horizontal fibers as guidelines and were able to write
on a relatively less smooth surface.

Ostraca, both sherds of broken pottery and pieces of limestone, were ideal for short exer­
cises. Convenience rather than cost per se dictated their use: They were so readily avail­
able that one could overcome the relative disadvantage of their uneven surface. Students
used them more at elementary levels than in later stages, and teachers wrote on them
model alphabets that could easily circulate in class. The visible correlation between the
size of an ostracon and the length of an exercise indicates that schoolchildren selected
them according to the writing space needed. Not all ostraca with literary texts necessari­
ly originated in school contexts. In most cases, though, this material points to an exer­
cise: Scholars had no problem getting hold of papyrus. In remote places such as the quar­
ry settlement of Mons Claudianus and the Roman military praesidia along the road of
Myos Hormos, ostraca were the principal writing material used for education and every­
day matters. The good, literary level of the school ostraca found in the former fort
prompted their editor to posit the existence of a schoolmaster who instructed children of
military personnel (O.Claud. 1179–189, and II 409–416). Ostraca found in Krokodilô and
Maximianon also testify to some kind of basic instruction, but their level is below that at­
tested in Mons Claudianus (Cuvigny 2003). It is impossible to know whether those who
Page 3 of 19
Education in the Papyri

wrote them were children or illiterate adults and whether the instruction was imparted
by a schoolmaster or by another adult with some education.

In comparison to other materials, tablets, which were made of wood, were considerably
more expensive because wood was scarce in Egypt. No tablets from (p. 323) the Ptolemaic
period are extant, and more tablets survive from the early Byzantine period than from the
Roman age. Both wooden tablets and waxed tablets were used for educational purposes,
either as individual pieces or joined together (up to ten tablets) to form “notebooks.” Stu­
dents used pen and ink to write on the wooden tablets, which could be covered with a
coating that made the surface smoother and allowed the writing to be washed off to some
extent. Waxed tablets were hollowed out, leaving a narrow rim around the edges. This
concavity was filled with wax, which was inscribed with a pointed stylus and erased with
its spatula-shaped back. Since they could easily be reused, these tablets were popular at
elementary levels, where the need to erase was frequent. Teachers and more experienced
students wrote calligraphic exercises and grammar on wooden tablets.

Other Distinguishing Features

Special features such as lines, borders, and decorations of various kinds help identify a
school exercise (Cribiore 1996, 75–96). Some of these had the practical purpose of divid­
ing or highlighting sections, but others served only to embellish schoolwork. The common
practice of writing words without separation (scriptio continua) made it necessary to em­
ploy reading and writing aids in the form of spaces, dots, and oblique strokes that divided
syllables and/or words. Words in lists were often divided into syllables, and these same di­
visions are visible in passages of authors that still offered learners a challenge. Teachers
found these divisions useful in the passages they wrote out for their classes. Models of
this kind (e.g., ibid., nos. 292, 296, and 342) were most often written on tablets that
might have been hung or circulated in the classroom. Since exercises, particularly in the
Roman and Byzantine periods, presented poetic texts in continuous lines, without re­
specting colometry (superficially indistinguishable from prose), marks were often used to
separate verses (Cribiore 1992). Finally, mistakes provide a strong argument for identify­
ing a particular document as schoolwork. Slips of the pen and phonetic mistakes are also
present in professional copies but often abound in exercises. Morphological errors in
grammatical exercises distinguish them from copies written by grammarians, and syntac­
tic mistakes indicate students' compositions.

School Hands and Teachers' Hands

The vast majority of “school hands,” that is, the writing of learners, display obvious fea­
tures caused by low speed and immaturity of handwriting (Cribiore 1996, 102–118).
Large size, irregular alignment and margins, varying inclination of letters, and clumsy let­
ter forms are evident to the palaeographer. It is possible to isolate four different types of
hand according to their writing experience: the “zero-grade hand” is that of the complete
novice; the “alphabetic hand” can be trusted for not more than (p. 324) alphabets and is
unable to bear the burden of longer texts; the “evolving hand” does a good amount of

Page 4 of 19
Education in the Papyri

writing and is moderately fluent but still displays a coarse and uneven look; finally, the
“rapid hand” is that of the older student and cannot identify an exercise in the absence of
other characteristics.

Teachers' hands may exhibit the large size that is characteristic of students' hands (in­
deed, they may be even larger) but show fluency, regularity, strength, and excellent legi­
bility (ibid., 97–102, 121–128). They are not as rigid and formal as “book hands” but dis­
play a considerable degree of beauty, that is, an attractive evenness and precision of the
strokes. “Teachers write letters of great beauty for the children to imitate,” wrote John
Chrysostom in the fourth century CE (MPG 59.385.56). The hands of models from Graeco-
Roman Egypt are graceful and elegant even though their style is often informal. Models
are attested in both literary and visual sources. In the absence of erasable blackboards
and schoolbooks with words separated, they were of great help both at the elementary
and the grammatical levels. Students could easily handle small ostraca with models of al­
phabets without much risk of damaging them, but large and heavy ostraca inscribed by
teachers might have been displayed in the classroom (e.g., Cribiore 1996, 319; O.Claud. II
415).

Types of Textual Material

Some examples point directly to schoolwork, but the papyrologist must also use other cri­
teria of distinction, particularly for textual materials at advanced levels of education. Al­
phabets and the repetition of individual letters can represent teachers' models and stu­
dents' and apprentice scribes' practice. Much is still unknown about the education of
scribes, such as whether they followed a regular course of study for some time or en­
rolled in scribal schools from the beginning (for scribes of Christian texts, see Haines-
Eitzen 2000). Scribes needed to have specialized training at a certain point to learn dif­
ferent styles of writing and acquire the complex terminology used in legal and bureau­
cratic documents.4 Since knowing the proper letter sequence was presumably a prerequi­
site for their specialized training, exercises that reinforced that ability were assigned to
pupils at elementary levels. Combining letters into syllables and mastering syllabaries
was the next hurdle for these learners. Syllabaries were more or less complete and elabo­
rate combinations of consonants and vowels; they exist in the form of teachers' models
and students' exercises, full of mistakes and imprecision. They were important in the
teaching of reading and writing (see Cribiore 1996, nos. 78–97). Authors such as Quintil­
ian (1.1.30) and Plato (Polit. 278b) emphasized the necessity of mastering all of the com­
binations because they enabled a learner to proceed to words, phrases such as maxims,
sayings and single verses, short passages, and, finally, long passages of authors.5

Lists of words also exist as models and students' exercises.6 The vast majority display
words either divided into syllables or arranged in groups according to (p. 325) the number
of syllables and were intended to teach reading and writing. Word lists, however, do not
necessarily pertain to an elementary level of learning or automatically indicate school
contexts, and the papyrologist needs to distinguish carefully. Some lists arranged by
theme, such as mythological and heroic genealogies taken from literary works, belong to

Page 5 of 19
Education in the Papyri

more advanced levels of education (e.g., Cribiore 1996, no. 390). When students were
ready to go beyond single words, their writing assignments started with limited amounts
of prose and verse and evolved progressively into extensive passages of authors. Elemen­
tary students wrote maxims, sayings of famous men, and short excerpts from poetry. Stu­
dents of grammarians copied or wrote (from dictation) long passages, mostly from the po­
ets. An important caveat: Identifying an exercise at this point becomes more problematic
since advanced students usually possessed a trained (rapid) hand that is virtually undis­
tinguishable from that of other educated people and scholars.

Knowledge of textual materials can still help us to identify schoolwork at higher levels of
education. Compositions with mistakes of morphology and syntax unmistakably belong to
school contexts. When they contain paraphrases and summaries of Homeric episodes and
books, however, we must take into account other distinguishing characteristics in order
to tell them apart from professional copies (ibid., nos. 344–357). The same is true in the
case of scholia minora, Homeric commentaries that consist of lists of words taken from
the Homeric text (lemmata) and accompanied by the corresponding words in a more cur­
rent form of Greek (glosses). In the past these commentaries were automatically consid­
ered the products of school activity, but the general educated public also needed such
“translations.” Thus scholia minora circulated as private copies and professionally pro­
duced books written in formal hands. While the exercises Cribiore included (ibid., nos.
325–343) seem to have originated in school settings, and undoubtedly many more copies
were used in educational contexts, it is impossible to tell them apart from those consulted
by the general public. Grammar was also a prominent subject in secondary education.
Students read and copied parts of grammatical handbooks (technai) and engaged in mor­
phological exercises of declension and conjugation (ibid., nos. 358–378).

When a young man entered a school of rhetoric, his hand was fairly trained, his spelling
more secure, and as a rule he indulged less in decorations of any sort, so that it is more
difficult to recognize students' work at this stage than before.7 In addition, since the
teaching of rhetoric was preeminent in Alexandria but only a limited number of papyri
from there have been preserved, the body of rhetorical exercises at the papyrologist's dis­
posal is limited. Textual material greatly helps us to understand how rhetoric was gener­
ally taught and practiced in Egypt, but one is hard put to distinguish the work of students
and teachers from that of amateurish practitioners of rhetoric. Nevertheless, a substan­
tial number of rhetorical exercises—different from professional texts and orations actual­
ly delivered—are extant. They range from preliminary exercises (progymnasmata) to
whole declamations (meletai).

(p. 326) Books Used in School

The teachers' models functioned as books for copying and consultation leading up to the
initial studies under the grammarian, when a student still needed help in decoding words.
The fourth-century Christian writer Basil of Caesarea (Homilia dicta tempore famis et sic­
citatis 67c) mentions some students breaking in anger the tablets that belonged to their

Page 6 of 19
Education in the Papyri

teacher. It is reasonable to suppose that many of the tablets found in Egypt were the
property of teachers who lent them to students. One might conjure up a scenario in which
the ownership of tablets and some degree of literacy were the simple prerequisites for a
primary teacher to set up a school. Yet, papyrus books (biblia) that presented material in
simplified ways also existed. The literary and papyrological sources indicate that students
used books at higher levels of education. A few letters on papyrus are illuminating in this
respect since they casually allude to schoolbooks of various kinds (P.Giss. 85; SB III 7268;
P.Oxy. III 531). Identifying such texts, however, is not easy. Even though there was not a
large market for books with special features (because the models adequately covered
those needs), some exceptions exist (Cribiore 2001b, 137–143). In a few papyri the writ­
ten text shows spaces between words and a very legible handwriting (e.g., P.Ryl. III 486).
In another papyrus with a text on the labors of Heracles written with gaps between the
words, colorful illustrations indicate that it probably addressed a young audience (P.Oxy.
XXII 2331). Since Homer was heavily present in ancient education, most examples of
school texts can be found among Homeric papyri that exhibit dots or vertical dashes to
separate the words (e.g., MPER n.s. III 3) and/or an unusual wealth of accents and other
lectional signs. This is particularly true when the accents were added, sometimes clumsi­
ly, by the hand of a student who was doing an exercise in accentuation (P.Lond.Lit. 28). As
noted earlier, texts with scholia minora and grammatical handbooks were in use in the
grammarian's classroom.

Learning According to the Papyri

Approaching Literacy

The school exercises and the few school texts from Graeco-Roman Egypt are little more
than precious outlines of what went on in an ancient classroom. Education was largely
oral and thus is mostly out of our reach. Yet, starting from the extant remains and taking
into account a few vivid papyrus letters written by parents and students, archaeological
findings, and the information that the ancient writers handed down, we can hope to fill in
the picture to a large extent. In considering (p. 327) teaching methods, we must keep in
mind that bilingualism (or diglossia) constituted a hurdle for students at every level of ed­
ucation. In Egypt, as in other provinces of the empire, many individuals, who in their dai­
ly lives functioned in the indigenous language, learned Greek in school. Those who were
exposed to Greek at home were in a privileged condition for learning but had to confront
the fact that the koinê Greek of their daily life differed significantly from the Attic Greek
they encountered in school. Learning methods and teaching aids had to take this reality
into account and remedy the lack of books that could easily be consulted, as well as the
absence of tables of contents, indices, library catalogues, and the like.

One way to cope with these disadvantages was to enforce an extremely thorough mastery
of the alphabet, which became as flexible as the numerical order and was used as both a
mnemonic device and an organizational tool. Teachers discouraged rote memory of the al­
phabet and made up exercises that consisted of following other alphabetical sequences:
Page 7 of 19
Education in the Papyri

skipping a fixed number of letters, proceeding from the bottom up, and pronouncing
tongue twisters made up of alphabets in scrambled order. In addition, the maxims that
served as copying exercises sometimes formed alphabetical acrostics; the syllabaries in­
culcated ways to combine letters; and words were listed in alphabetical order, which was
used as a mnemonic device. Strengthening the memory was a constant concern. More ad­
vanced students not only memorized texts word for word but also practiced mnemotech­
niques in rhetorical school (Blum 1969; Small 1997).

The papyri contain examples of the rigid teaching method attested by the literary sources
that taught reading and writing by means of building blocks (letters, syllables, words,
sentences, and passages). Yet the school exercises from Egypt indicate that this sequence
was not universally followed. In order for students to practice handwriting, teachers
made them write their names and copy verses and texts of very limited extent as soon as
they learned their letters (e.g., Cribiore 1996, nos. 136, 160, 202, 383, 403). These stu­
dents could not read what they had copied but proceeded blindly, committing every sort
of mistake and omission (Cribiore 2001b, 167–172). This copying method apparently be­
came more popular in the late Roman and Byzantine periods and sometimes coexisted
with the traditional method. It was particularly useful to those who were in school for on­
ly a short time because they could thus acquire a limited literacy that enabled them to
perform in a society in which most people were somewhat familiar with reading and writ­
ing (Hanson 1991; Bowman 1991). The painfully written subscriptions of the so-called
slow writers and an example such as that of Petaus, a town clerk who could not read but
passed himself off as literate, are eye-openers (Youtie 1966).

The cultural content of an elementary education was very limited: Some maxims and say­
ings (sometimes by Isocrates and Menander) and a few verses of Homer and Euripides
were powerful symbols of literacy. The elementary package also included numerical liter­
acy. Since the letters of the alphabet, with the addition of three more signs, functioned as
numbers, it was always assumed that numerical (p. 328) operations were part of the first
phases of instruction. The mathematical exercises found in Egypt, however, allow us to
make a few distinctions (Cribiore 2001b, 180–183). Students rarely wrote down additions,
which they might have recited aloud, and engaged in written multiplications and frac­
tions only when their handwriting was fairly proficient. Many tables of fractions exist, but
most of them are capably written and were either hung in offices to facilitate computation
or used as teachers' models. Elementary students mostly learned mathematical opera­
tions by memorizing these tables. Extensive and advanced mathematical work was part of
the teaching of specialized schools.8

Christian education was closely modeled on pagan Greek formation and progressed
through the same stages. Besides studying the traditional classic authors, however, stu­
dents also copied and learned by heart the Psalms and passages from the Scriptures (e.g.,
Cribiore 1996, no. 403). From the third century CE on, Coptic schooling also used reli­
gious texts for practice and, like Greek education, entirely ignored Pharaonic culture
(Cribiore 1999). Students likely learned the Greek and Coptic alphabets at the same time,
but it is difficult to ascertain whether the teaching of the two languages proceeded simul­

Page 8 of 19
Education in the Papyri

taneously as the student advanced. A characteristic of Coptic education was to teach be­
ginners to write the opening and formulaic parts of letters, therefore addressing practical
needs. Epistolary texts were not used as copybooks in Greek school contexts. Since there
are no evident remains of advanced Coptic exercises, the highly rhetorical style of au­
thors such as Shenoute and Besa in all probability originated from their exposure to pa­
tristic literature and Greek rhetorical education.

I have referred so far to “students” and “teachers” generically, but education was open
not only to males. Girls also had access to primary instruction, but boys were the vast ma­
jority. The disproportion became more pronounced in grammar schools, though a number
of girls of the upper class also attended, as certain papyrus letters sent by women attest
(Cribiore 2001b, 74–101; Bagnall and Cribiore 2006). Rhetorical training, the last stage of
the enkyklios paideia, was in any case closed to female students because it was envi­
sioned as preparation for public and political life. Low-level teaching was not entirely in
the hands of male teachers. The literary sources disclose little about women teachers and
completely disregard female primary teachers, but the papyri indicate the existence of a
few of them. Even though, like their male counterparts, these teachers appear by their ti­
tle (deskalos or deskalê) in documents that do not reveal anything about their didactic ac­
tivity (e.g., P.Mich. VIII 464; P.Mich. II 123; and BGU I 332), their presence per se is a sig­
nificant testimony that education in antiquity was not entirely out of women's reach.

Strengthening the Language

Students who entered the grammarian's class developed and ampliied knowledge previ­
ously acquired. This common educational principle reached paradoxical (p. 329) propor­
tions in the ancient world. The grammarian's students already knew that the elements of
knowledge fitted precisely into a grid and were tightly connected like the links of a chain,
and they soon realized that multiple connections existed with the previous level. With
their divisions and lectional signs, the teachers' models provided necessary help to the in­
experienced student who read slowly, by syllables and words. They formed a transition to
the texts in general circulation, which were less “user friendly.” Education proceeded in a
circular fashion, so that the more advanced students revisited texts they had previously
encountered. Their reading now was more fluent, and the questions they had to answer
about a text were more thorough and diverse, but most of the authors they dealt with
were those they had met before.

On the whole, the school papyri agree with the literary sources with regard to the au­
thors studied under the grammarian, but they are less helpful in identifying the reading
list of an advanced student whose hand was experienced (Cribiore 2001b, 192–204). This,
for instance, is the reason that Hesiod does not frequently appear among the school pa­
pyri, although the educational writers attest to his presence in the schoolroom, and the
extant papyri that preserve his works are quite numerous.9 Homer was the author that
students came to know in detail, and the thousand or so extant Homeric papyri confirm
his popularity among the cultivated public. Both students and the general public vastly
preferred the Iliad to the Odyssey. The grammarians read the first six books of the Iliad in

Page 9 of 19
Education in the Papyri

detail, and their more advanced students went through the whole work.10 Ancient teach­
ers always concentrated on the beginning of an author's work; thus, the first two books in
particular were the subject of a meticulous analysis. Euripides was a major presence in
education, whereas other tragedians were overlooked. He was linguistically more accessi­
ble, and his most rhetorical plays (Phoenissae, Orestes, Hecuba, Medea, and Alcestis) con­
tinued to occupy students in a school of rhetoric. The popularity of Phoenissae was uncon­
tested from the time pupils wrote maxims to improve their penmanship until they en­
gaged in composition exercises (Cribiore 2001a). The maxims of Menander also accompa­
nied students from the beginning to the end, up to the time when they learned to develop
them with the rhetor, and Menander's plays, copied with many mistakes and corruptions
by students and teachers alike, appear among the P.Bodmer. Menander's monostichoi
continued to enjoy a vast popularity in the Byzantine period, but his comedies lost favor.
Teachers at advanced levels preferred Aristophanes, who was more interesting linguisti­
cally. Thus Zuntz (1975) has shown that marginal notes in late papyri of Aristophanes de­
rive from schoolbooks (e.g., P.Oxy. VI 856). The presence of other poets in the schoolroom
is more difficult to verify. Some works by Callimachus, Theognis, Sappho, Hipponax, and
Pindar surface but very sporadically because of the usual difficulties in identifying ad­
vanced schoolwork.11

Did the grammarians concentrate exclusively on poetry? Undoubtedly this was their tradi­
tional area of expertise, but they also taught reading, writing, and (p. 330) grammar with
the aid of fables and some Isocrates. One again sees the wide application of the educa­
tional principle of making a pupil revisit the same texts at subsequent levels. Present at
all stages of education, fables were the basis of the first rhetorical exercises. Likewise, el­
ementary students copied short excerpts from the Cyprian Orations of Isocrates (Ad De­
monicum, Ad Nicoclem, Nicocles); read more extensive passages under the grammarian
with the help of professionally produced books; and, with the rhetor, developed maxims
from these orations. A book of tablets containing these three speeches, summaries, and
lexical notes squeezed into the margins is the product of the grammarian's schoolroom
(Worp and Rijksbaron 1997).

Isocrates remained enormously popular at all times. Among seven Byzantine school
tablets, which date mostly from the seventh century and preserve Greek and Coptic texts,
one tablet dated to 470 CE displays a model and a copy from the Ad Demonicum and a list
of months (Duttenhöfer 1997). Before writing down the model, the teacher exhorted the
pupil, “Pay attention. I wrote in nice letters.” Likewise, a passage from Nicocles, written
from dictation in the sixth century on the back of a protocol, testifies to the continuous at­
traction Isocrates exercised (Lundon and Messeri 2000). Because of the heavily gnomic
(that is, moralistic and didactic) content of these three orations, they were also very well
liked by the general public: They make up about half of the papyri of Isocrates. While an­
cient authors suggest that students at this stage paid more attention to reading literature
than to writing, the papyrological sources indicate that they practiced epistolary writing
(as some letters sent to families show) and later perfected their skills with the rhetor
(Cribiore 2001b, 215–219).

Page 10 of 19
Education in the Papyri

The papyri allow us to enter the grammarian's schoolroom and glance at all aspects of its
activity. Since Homer was the author par excellence, let us follow the grammarian's
teaching by focusing on some of the Homeric exercises. As I have already said, metrics
was an area of fundamental interest, so students had to do exercises in accentuation. In a
Roman papyrus, a student wrote down only the first half of each line of Odyssey 9.122–
150, producing two crowded columns separated by a thick, rough line. In an exercise that
probably also involved memorization, the student used the papyrus very economically to
show his knowledge of the lines up to the caesura (Cribiore 1996, no. 291).12 A grammari­
an had to elucidate all of the words in the Homeric text, thereby producing “historical
notes” (historiai), that is, details on the mythological matters, persons, places, and events
mentioned. Two fragments from a third-century papyrus preserve a student's exercise:
names of Achaean heroes from the Iliad, together with their fathers and mothers and a
list of gods with their genealogy (P.Oxy. LXV 4460). The content, the evolving hand, the
decorated title, and the long lines that separate the sections mark this as a product of the
grammarian's schoolroom.

Another evolving hand wrote on the verso of a Roman account (P.Oxy. LVI 3829). The stu­
dent in question committed several mistakes and corrected some of (p. 331) his clumsy let­
ters by rewriting them above the line. This papyrus spotlights a series of school activities.
It contains the end of a catechism (questions and answers) that listed the characters of
the Iliad; a narrative concerning the events leading to the Trojan War (including the judg­
ment of Paris); Iliad 1.1; and a summary of that book. The first section with erôtêmata
allows us to perceive the oral side of the grammarian's teaching. A student had to demon­
strate his knowledge by answering various questions, such as on the identity of Hector's
advisers or on the seers who appear in the Iliad. The question-and-answer format was ex­
tremely popular in late antiquity and in the Middle Ages to systematize knowledge in vari­
ous fields, such as medicine and grammar, into easily assimilated parcels. It seems to
have originated as a pedagogical tool in the Roman schoolroom.

The grammarian's practice of explaining all historical and geographical details often went
to extremes. His students learned much information that bordered on the useless and
paid attention to the minutiae in a text rather than to the meaning and themes of the
whole. But in one area the grammarian's pedantic focus on details was necessary. The lin­
guistic background of the students made elucidation of unfamiliar Homeric vocabulary
imperative. Students who already had trouble with the Attic Greek they encountered in
school needed more help in decoding Homeric terminology. Scholia minora to Homer pro­
vided an elementary commentary that they had to consult and copy. It is not surprising
that these are more numerous for the Iliad than for the Odyssey since teachers and the
general public were more interested in the former. The grammarian focused on the first
book of the Iliad, for which he provided glosses that covered the text almost in its entire­
ty. An exercise with scholia minora to a few lines of book one carries us into the midst of
the classroom and shows how students at every level read according to syllables and not
to whole words (Cribiore 1996, no. 339). A student apparently first copied the whole col­
umn of the lemmata and then the glosses in a parallel column. He was following a model

Page 11 of 19
Education in the Papyri

and wrote down only the first syllable of each gloss and interrupted his work after five
lines.

Once the class had assimilated the vocabulary of the first book of the Iliad, the grammari­
an could zero in on other difficult terms in a variety of books, thereby providing a sparser
commentary. Consider, for example, a fragment of a roll with Iliad 5.24 written by a
teacher: The glosses accompany a text reproduced in its entirety and provide an easily
consulted Homeric dictionary (ibid., no. 330). In compiling their vocabularies to Homer,
grammarians used glossographical material that had an ancient, scholarly origin, but the
erudite scholia vetera that appear in some Homeric manuscripts have a different scope
and tone. Grammarians felt free to modify and integrate that ancient exegesis; as a re­
sult, the scholia found in Egypt also provide their personal contributions. This elementary
glossographical material was assimilated into a Byzantine compilation, the so-called D-
scholia, which offered every kind of Homeric exegesis, such as paraphrases and sum­
maries of Homeric books, mythographical material, and inquiries (zêtêmata) (p. 332) into
certain questions. Mythographical notes are found in a number of school papyri (e.g.,
ibid., no. 183). Montanari (1995) has shown that teachers consulted an ancient commen­
tary on mythological subjects, usually called Mythographus Homericus, which circulated
in scholarly and scholastic circles.

The technical aspect of grammar had very likely always been a part of the grammarian's
teaching, but the fact that no grammatical exercises and texts have survived from the
Hellenistic period cannot be only the result of chance. The Alexandrian scholars worked
on systematization of parts of grammar, but only from the beginning of the Roman period
did grammar become codified into a body of knowledge that was transmitted separately.
It is not a coincidence that texts containing scholia minora appeared in the same period
and not before. Until then the need to classify grammatical terms and forms was not felt
with the same urgency, and the understanding of the Homeric text was less deficient. Lin­
guistic tools were increasingly more necessary to approach the ancient authors and grasp
the nuances of poetry. Grammatical manuals that treated the parts of speech started to
circulate (Wouters 1979), but their influence on schoolwork seems to have been limited.
Up to the fourth century, the papyri that preserve parts of handbooks used in school dif­
fer from the Technê of Dionysius Thrax, a grammarian who wrote around 100 BCE (see,
e.g., Cribiore 1996, nos. 358, 359, 362, 368, 371, 373). The authenticity and dating of the
body of this text as it was transmitted are still under scrutiny (e.g., Law and Sluiter
1995), but it is clear that it became the standard school text only from the fifth century
on. In the previous period, grammarians had to rely on the work of their predecessors
and adapt it to their pedagogical needs. A papyrus from the first century, which provides
definitions of genders, numbers, and types of nouns, well exemplifies this trend (PSI inv.
505; Di Benedetto 1957). This manual was either a cheap professional copy or a copy
made by a teacher and follows the question-and-answer format throughout.

Page 12 of 19
Education in the Papyri

Learning Rhetoric

The training of a young man of the elite who entered a school of rhetoric relied heavily on
what he had learned with the grammarian. Knowledge at this stage was organized ac­
cording to the customary inflexible order in links of progressive difficulty, each connected
to the one before and to the next (Cribiore 2001b, 220–244). Thus progymnasmata
expanded the knowledge of poetry already acquired and concentrated on Homer and the
Iliad in particular.13 In practicing exercises such as those of praise (encomium), blame
(psogos), impersonation (êthopoiia), and description (ekphrasis), a student improved his
skill in writing, expression, and observation. At a higher level of expertise, he could incor­
porate some progymnasmata into the composition of declamations (meletai) on delibera­
tive and forensic subjects. Theon of Alexandria composed a handbook of preliminary exer­
cises in the (p. 333) first century (Patillon 1997), and in fourth-century Antioch, the sophist
Libanius wrote progymnasmata as models for his students (Foerster 1963, vol. 8).

While most of the exercises in these and later collections correspond to those found in the
papyri, students in Egypt practiced especially êthopoiiai and encomia. It is important to
note, however, that most of the exercises on papyrus are in verse (epic hexameters and
iambics) even though the various literary collections contain examples only in prose as
preparation for rhetorical discourse. It seems that rhetorical training not only was based
upon previous knowledge of the poets but also reinforced and expanded students' poetic
skills. The student who wrote an exercise of impersonation followed the text of a certain
author but lingered on a mythological or literary figure's reaction to a specific event, re­
porting, for instance, the words of Phoinix reproaching Achilles or those of the latter at
the point of death.14 An êthopoiia did not require much originality, but it was very useful
to practice the êthos of several characters, a skill a student needed because he would
never appear in his own persona in historical and argumentative declamations.

Encomia found in Egypt are very similar to those that appear in rhetorical textbooks and
did not focus only on mythological subjects. Consider, for example, the encomium of the
fig, which was supposedly the favorite fruit of Hermes, the god of rhetoric, and sports a
heavily decorated title (P.Oxy. XVII 2084), or the praise of the horse in another papyrus
also found at Oxyrhynchos (P.Oxy. LXVIII 4647). Praise of various persons and gods was a
traditional subject. A Roman exercise preserves an encomium of Dionysus that extols him
on the occasion of a certain celebration. Its irregular, rapid hand and the general presen­
tation point to a teacher's model rather than to the composition of a rhetor that was actu­
ally delivered (P.Köln VII 286). As Libanius shows (e.g., Ep. 63), speeches of praise were
useful to the student who stopped at this level of rhetorical education and could use this
skill to compose panegyrics of officials. Rhetors and poet-grammarians often engaged in
similar exercises and competed in public contests and festivities (e.g., P.Oxy. VII 1015).

Young men who continued in their education faced rigorous training. The theory of “is­
sue” (stasis), which governed the composition of declamations, was a formidable, de­
manding system. In the preliminary exercises, a student had used building blocks to con­
struct relatively short pieces, but now he had to follow painstakingly detailed instructions

Page 13 of 19
Education in the Papyri

to learn to construct arguments and cases. In spite of limitations, this system helped de­
velop verbal skills and the capability of analyzing the pros and cons of a case. The tradi­
tional Roman division of declamations into suasoriae (on imaginative, deliberative
themes) and controversiae (fictitious legal cases) was not followed in Greek education,
where declamations were distinguished as either historical or fictive. In Egypt, in any
case, historical declamations were much more common, and examples of fictive meletai
are rare.15 The random chance of survival and the preservation of only a few Alexandrian
papyri might be responsible for that, but one should not rule out the possibility that
(p. 334) sophists in Egypt taught rhetoric mainly through literary texts (historical and ora­

torical) and avoided fictitious legal themes. A young man who wished to practice law
could learn on the job or go to a school of Roman law either in Alexandria or abroad, par­
ticularly to the renowned school of Berytus (Beirut).

In the late Roman period, a change in the training of the advocates occurred, and the ed­
ucation acquired in a school of rhetoric became increasingly insufficient. Most young men
who aspired to jobs in the administration opted to learn Roman law. A knowledge of some
Latin, therefore, became mandatory, and the so-called Latin school exercises found in
Egypt are evidence of this. They consist of bilingual glossaries and bilingual lists of au­
thors. Most of the glossaries are transliterated; that is, they are written entirely with the
Greek script (Kramer 1983, 2001). The vast majority of the bilingual word lists are from
the Aeneid and show either the whole text or isolated words rendered in Greek.16 A few
bilingual books also contain texts of Cicero, Juvenal, Sallust, and Terence.17 The script of
these glossaries and lists consists of either fluent, cursive hands (those of teachers per­
haps or of lesser scriptoria) or the formal hands of professionally produced books
(Cribiore 2003–04). It seems, therefore, that these were not exercises written by students
but were more or less formally produced books on which they practiced their reading
skills. It is likely that a veneer of Latin was sufficient to enter a school of Roman law.

In conclusion, the contributions of the papyri to our knowledge of education are many.
They beautifully illustrate what the ancient educational writers have handed down re­
garding the methods and stages of teaching and at the same time allow us to glimpse real
students and teachers whose actual work was preserved. They are evidence that iron
rules systematized knowledge and a strict curriculum governed the various stages of edu­
cation; when its traces are faint, it is only because of the difficulty of identifying the work
of advanced students. But the papyri also provide important correctives to the literary
sources. They show that elementary training served both the student who was going to
continue his schooling and the one who dropped out and that education, for all its rigidity,
never lost touch with reality. Likewise, it appears that grammatical education not only ex­
posed students to the poets and rules of grammar but also gave them the ability to ex­
press themselves in letters sent to family members, a skill that they perfected at later
stages. While modern educational theorists have maintained that poetry was exclusively
the province of the grammarian and that the rhetor taught only prose (e.g., Marrou 1975,
296), the papyri provide a more nuanced view of the transition between the two levels
and prove that poetry continued to interest the student of rhetoric. The numerous rhetori­
cal exercises in verse point neither to an eccentric phenomenon nor to the exclusive
Page 14 of 19
Education in the Papyri

predilection of the Egyptians for poetry but confirm that poetry was cultivated every­
where at high levels of education, particularly in late antiquity, though only the sands of
Egypt offer us remnants of actual school contexts.18

Bibliography
Abbenes, J. G. J., S. R. Slings, and I. Sluiter, eds. (1995). Greek Literary Theory after Aris­
totle: A Collection of Papers in Honour of D. M. Schenkeveld. Amsterdam: VU University
Press.

Bagnall, R. S., and R. Cribiore 2006. Women's Letters from Ancient Egypt, 300 BC–AD
800. Ann Arbor: University of Michigan Press.

Blum, H. (1969). Die antike Mnemotechnik. Hildesheim: Georg Olms.

Bonner, S. F. (1977). Education in Ancient Rome: From the Elder Cato to the Younger
Pliny. Berkeley: University of California Press.

Bowman, A. K. 1991. “Literacy in the Roman Empire: Mass and Mode.” In Literacy in the
Roman World, 119–131. Ann Arbor: Journal of Roman Archaeology.

Cameron, Alan. 1965. “Wandering Poets: A Literary Movement in Byzantine Egypt.” His­
toria 14: 470–509.

———. 2004. “Poetry and Literary Culture in Late Antiquity.” In Approaching Late Antiqui­
ty: The Transformation from Early to Late Empire, ed. S. Swain and M. Edwards, 327–
354. Oxford: Oxford University Press.

Cribiore, R. 1992. “The Happy Farmer: A Student Composition from Roman Egypt.” GRBS
33: 247–263.

———. (1996). Writing, Teachers, and Students in Graeco-Roman Egypt. Atlanta: Scholars
Press.

———. 1999. “Greek and Coptic Education in Late Antique Egypt.” In Ägypten und Nubi­
en in spätantiker und christlicher Zeit, ed. S. Emmel et al., 279–286. Wiesbaden: Re­
ichert.

———. 2001a. “The Grammarian's Choice: The Popularity of Euripides' Phoenissae in Hel­
lenistic and Roman Education.” In Education in Greek and Roman Antiquity, ed. Y. L. Too,
241–259. Leiden: Brill.

———. 2001b. Gymnastics of the Mind: Greek Education in Hellenistic and Roman Egypt.
Princeton: Princeton University Press.

———. 2003/04 “Latin Literacy in Egypt.” KODAI: Journal of Ancient History 13/14
(2003/04) [2008]: 111–118.

Page 15 of 19
Education in the Papyri

Cuvigny, H., ed. 2003. La route de Myos Hormos: Lʼarmee romaine dans le désert oriental
dʼÉgypte. 2 vols. Cairo: Institut francais dʼarcheologie orientale.

Di Benedetto, V. 1957. “Frammento grammaticale.” Annali Pisa 2(26): 180–185.

Di Bitonto-Kasser, A. 1998. ”P.Bodmer LI recto: Esercizio di divisione sillabica.” Museum

Helveticum 55: 112–118.

Duttenhöfer, R. 1997. “Schultexte auf byzantinischen Holtztafeln: Isokrates, Pros De­

monikon §17.” Akten 21. Papyrologenkongresses, 244–250. Stuttgart: Teubner.

Emmel, S., M. Krause, S. G. Richter, and S. Schaten, eds. (1999). Ägypten und Nubien in
spätantiker und christlicher Zeit. Wiesbaden: Reichert.

(p. 337) Foerster, R. 1963. Libanius, Opera. 12 vols. 1903–1927; repr. Hildesheim: Olms.

Gaebel, R. E. 1970. “The Greek Word-Lists to Vergil and Cicero.” Bulletin of the John Ry­
lands Library (Manchester): 284–325.

Haines-Eitzen, K. (2000). Guardians of Letters: Literacy, Power, and the Transmitters of

Early Christian Literature. New York: Oxford University Press.

Hanson, A. E. 1991. “Ancient Illiteracy.” In Literacy in the Roman World, 159–198. Ann Ar­
bor: Journal of Roman Archaeology.

Harris, W. V. (1989). Ancient Literacy. Cambridge, Mass.: Harvard University Press.

Kramer, J. (1983). Glossaria bilinguia in papyris et membranis reperta. Bonn: Pap. Texte
Abh. 30. Bonn: R. Habelt.

———. (2001). Glossaria bilinguia altera. Munich: K.G. Saur.

Kugener, M.A., ed. (1903). Zacharias Scholasticus, Vita Severi. Patrologia Orientalis II,
fasc. 1. Turnhout: Brepols.

Law, V., and I. Sluiter, eds. (1995). Dionysius Thrax and the Techne Grammatike. Münster:
Nodus Publikationen.

Leuven Database of Ancient Books (LDAB). http://ldab.arts.kuleuven.ac.be/.

Lundon, J. 2004. “The Scholia Minora in Homerum: An alphabetical List.” http://

www.uni-koeln.de/phil-fak/ifa/NRWakademie/Lundon/ScholiaMinora.htm.

———, and G. Messeri. 2000. “A Passage of Isocrates on the Back of a Protocol (P.Vindob
G 39977).” ZPE 132: 125–131.

Marrou, H.-I. (1975). Histoire de lʼeducation dans lʼantiquite, 7th ed. Paris: Seuil.

Page 16 of 19
Education in the Papyri

McNamee, K. 1994. “School Notes.” In Proceedings of the Twentieth International Con­

gress of Papyrologists, ed. A. Bulow-Jacobsen, 177–182. Copenhagen: Museum Tuscu­
lanum Press.

Montanari, F. 1995. “The Mythographus Homericus” In Greek Literary Theory after Aris­
totle: A Collection of Papers in Honour of D.M. Schenkeveld, ed. J. Abbenes, S. Slings, and
I. Sluiter, 135–172. Amsterdam: VU University Press.

Morgan, T. (1998). Literate Education in the Hellenistic and Roman Worlds. Cambridge:
Cambridge University Press.

Patillon, M. (1997). Aelius Théon: Progymnasmata. Paris: Les Belles Lettres.

Small, J. P. (1997). Waxed Tablets of the Mind: Cognitive Studies of Memory and Literacy
in Classical Antiquity. New York: Routledge.

Swain, S., and M. Edwards, eds. (2004). Approaching Late Antiquity: The Transformation
from Early to Late Empire. Oxford: Oxford University Press.

Thomas, R. (1992). Literacy and Orality in Ancient Greece. Cambridge: Cambridge Uni­
versity Press.

Too, Y. L., ed. (2001). Education in Greek and Roman Antiquity. Leiden: Brill.

Worp, K. A., and A. Rijksbaron. (1997). The Kellis Isocrates Codex. Oxford: Oxbow.

Wouters, A. (1979). The Grammatical Papyri from Graeco-Roman Egypt: Contributions to

the Study of the “Ars Grammatica” in Antiquity. Brussels: Koninklijke Academie voor
Wetenschappen, Letteren en Schone Kunsten van België, Klasse der Letteren.

Youtie, H. C. 1966. “Pétaus, fils de Pétaus, ou le scribe qui ne savait pas écrire.” CdÉ 41:
127–143 = Scriptiunculae II (Amsterdam 1973), 677–695.

Zintzen, C. (1967). Damascii vitae Isidori reliquiae. Hildesheim: Olms.

Zuntz, G. (1975). Die Aristophanes-Scholien der Papyri. Berlin: Seitz.

Notes:

(1.) Morgan (1998) needs to be used with some caution, especially for matters concerning
higher education. In what follows, I identify exercises by the numbers included in
Cribiore (1996) and refer to full bibliographic references only for items that have ap­
peared subsequently.

(2.) Especially Zacharias Scholasticus and Damascius; see Kugener (1903) and Zintzen
(1967).

(3.) I cover neither philosophy, which was outside this circle, nor specialized areas of
knowledge such as higher mathematics, geometry, and astronomy.

Page 17 of 19
Education in the Papyri

(4.) Apprentice scribes did most of the exercises in copying parts of personal and official
letters that are included in MPER n.s. XV. For scribal practice at a higher level, see, for
example, P.Köln VII 298, and P.Oxy. LXVIII 4668.

(5.) In describing the steps involved in the teaching of reading and writing, modern histo­
rians rigidly follow the accounts of ancient authors such as Augustine, De ordine 2.7(24);
Ambrose, De Abraham 1.30; Manilius, Astronomica 2.755–764; Jerome, Ep. 107.4 and
128.1; and Gregory of Nyssa, De beneficentia 5–13.

(6.) To the lists in Cribiore (1996, nos. 98–128), add Di Bitonto-Kasser (1998), a model
that was reused for cartonnage.

(7.) Cribiore (1996), whose main interest was to show the various stages in learning to
write, included only a few exercises at this level that showed a deficient hand. Morgan
(1998) relied only on these; consequently, her treatment of rhetoric in Egypt is incom­
plete. Cribiore (2001b, 220–244) covered rhetoric more fully.

(8.) Apprentice scribes practiced multiplications and fractions extensively, as many exer­
cises in MPER n.s. XV show. They also wrote down tables of weights, measures, and the
like; see, for example, P.Köln VIII 352.

(9.) On his presence in school, see Cribiore (1996, no. 386) (Hesiod as copybook), ibid.,
no. 390 (in a list of words), and P.Oxy. LXI 4099 (mythographic lists). See also P.Oxy. XXIII
2355, a school text. According to the LDAB, the papyri of Hesiod number 155.

(10.) See two recently published school papyri with Iliad 24, BASP 41 (2004): 46–50.

(11.) See Cribiore (1996, nos. 303, 379, 590, 234, 235, 237, 247) and McNamee (1994).

(12.) See a similar exercise on Iliad 24, BASP 41 (2004): 46–50, and on mythological hexa­
meters with dots for metrics, P.Köln VIII 328.

(13.) See Cribiore (1996, 344–357). See also, for example, PSI XIII 1303; P.Harr. I 4; P.Ryl.
III 487; O.Bodl. II 2171; P.Kooln VII 286; and P.Oxy. LXVIII 4647.

(14.) Cf. G. A. Gerhard and O. Crusius, Melanges Nicole (Geneva 1905), 615–624, and
P.Cair.Masp. III 67316v.

(15.) On rhetorical treatises found in Egypt see, for example, P.Oxy. III 410; LIII 3708; and
P.Yale II 106. Examples of historical declamations are BKT VII 4–13; P.Oxy. XXIV 2400; VI
858; II 216; XV 1799; XLV 3235 and 3236; P.Yale II 105; P.Lond.Lit. 139; and P.Hib. I 15.
Examples of fictive declamations are P.Lond.Lit. 138; P.Oxy. III 471; and P.Hamb. III 163
(in Latin). The themes of the historical declamations found in Egypt are largely from
Athenian history in the period of Demosthenes and following the death of Alexander.

(16.) Gaebel (1970). See Aeneid LDAB 4146, 4149, 4154, 4155, 4156, 4160, 4146, and
4162; and LDAB 4159 (Georgics).

Page 18 of 19
Education in the Papyri

(17.) Cicero is represented in four papyri, LDAB 554, 556, 559 (passages from orations in
Catilinam), and 558 (Divinatio in Q. Caecilium). See also LDAB 2559 (Juvenal); 3875 and
3877 (Sallust); 3982 and 3983 (Terence).

(18.) Cf. Eunapius, VS 10.7.10–13, 493. See Cameron (1965, 2004).

Raffaella Cribiqre

Raffaella Cribiore is Professor of Classics, New York University.

Page 19 of 19
Mathematics, Science, and Medicine in the Papyri

Mathematics, Science, and Medicine in the Papyri

Alexander Jones
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0015

Abstract and Keywords

The sciences that are significantly attested among the papyri are mathematics, medicine,
astronomy, and astrology. Medicine and astrology were professions with numerous practi­
tioners who possessed collections of reference texts and whose activity involved produc­
ing written documents. Astronomy had a central role in weather prediction, calendrics,
and, above all, astrology. Mathematics had its uses in calculation and mensuration, and
applied and theoretical mathematics were often components of technical and liberal edu­
cation, respectively. These were all subjects that were cultivated for external ends at least
as much as for pure intellectual satisfaction. The most obvious difference between the ev­
idence of the medieval textual tradition, which was heavily influenced by the intellectual
elite of the later Roman and Byzantine periods, and that of the papyri is that much of the
papyrological record has a direct bearing on practice and applications.

Keywords: ancient sciences, papyri, medicine, astronomy, astrology, technical education, liberal education

THE contribution of papyrology to our present understanding of ancient science is large

but very unevenly distributed. First of all, in antiquity no overarching concept existed of a
field of study coextensive with the common modern meaning of “science.” When we speak
of science in antiquity, we generally mean two different things: first, a handful of more or
less autonomous traditions or disciplines that dealt with healing, time reckoning, the in­
terpretation of ominous astronomical phenomena, and so on in both the Near Eastern and
classical civilizations; and second, a range of subjects—meteorology and zoology are ex­
amples—that did not become established as disciplines but were usually subsumed by
Graeco-Roman authors within the broad heading of natural philosophy. In this chapter we
are concerned only with the scientific disciplines that were recognized as such in antiqui­
ty.

Perhaps the first lesson that the papyri can teach us about ancient science comes from re­
alizing that the areas where they fall silent are not entirely accidental. Thus, optics (the
study of visual appearances), harmonic theory, and theoretical mechanics are scarcely
present in the papyri, although each of these sciences had an abundant literature in Hel­
Page 1 of 20
Mathematics, Science, and Medicine in the Papyri

lenistic and Roman times and was well represented in the medieval transmission of Greek
scientific texts. This fact probably indicates that these were (p. 339) academic specialties
studied almost exclusively for their own sake in a few intellectual centers (Alexandria, for
instance, but not Oxyrhynchus). In many fields the community of research scientists and
advanced students equipped to understand their work was always small; thus, according
to one plausible estimate, the number of creative mathematicians in the Graeco-Roman
world was never so much as a hundred (Netz 2002, 202–205).

The sciences that are significantly attested among the papyri are mathematics, medicine,
astronomy, and astrology. Again we can see why this should be so. Medicine and astrolo­
gy were professions with numerous practitioners who possessed collections of reference
texts and whose activity involved producing written documents. Astronomy, though not a
profession in its own right, had a central role in weather prediction, calendrics, and,
above all, astrology. Mathematics, too, had its uses in calculation and mensuration, and
applied and theoretical mathematics were often components of technical and liberal edu­
cation, respectively. In other words, these were all subjects that were cultivated for exter­
nal ends at least as much as for pure intellectual satisfaction. The most obvious differ­
ence between the evidence of the medieval textual tradition, which was heavily influ­
enced by the intellectual elite of the later Roman and Byzantine periods, and that of the
papyri is that much of the papyrological record has a direct bearing on practice and appli­
cations.

Mathematics
Scribal Mathematics

Two broad styles of doing mathematics can be distinguished in the ancient world. One
was concerned with finding numerical values of quantities or magnitudes in set situa­
tions, usually (although not always) expressed in terms of real-world objects; the other,
with finding and proving properties of mathematical entities such as whole numbers or
circles, usually (but not always) treating them as abstractions. We customarily associate
the label “Greek mathematics” with the latter style while associating the former with
Egypt and Mesopotamia, but both were in fact practiced side by side in the Graeco-Ro­
man world, and there was a certain amount of sharing of methods and results between
them. To take a single instance, the theorem that the surface of a sphere has four times
the area of its largest circular section, which Archimedes discovered and proved rigorous­
ly in On the Sphere and Cylinder, could subsequently appear without demonstration as a
mensurational formula in practically oriented texts such as Hero's Metrica. The appropri­
ate distinction to make is thus not between Greek and Near Eastern but between academ­
ic and scribal mathematics.

The medieval manuscript tradition was highly selective. It favored the abstract,
(p. 340)

proof-oriented academic style through the survival of writings by a small number of au­
thors, most notably Euclid, Archimedes, and Apollonius, who were active in the third and

Page 2 of 20
Mathematics, Science, and Medicine in the Papyri

early second centuries BCE, along with a few rather less eminent figures such as Pappus,
who lived in late antiquity. Such texts in the quantity-oriented scribal style as were trans­
mitted (e.g., the Arithmetical of Diophantus and some of the works attributed to Hero of
Alexandria) belonged to the more literary part of this tradition most closely related to the
proof-oriented style. By contrast, the papyri confront us repeatedly with its more practi­
cal side, the kind of mathematics that was useful—or thought to be so—for everyday pur­
poses. For pre-Hellenistic Egypt, moreover, papyri are our principal source of information
on mathematics.

Numbers in Greek papyri were almost always written in the Ionian (alphabetic) notation,
which used ten as its base but was not a place value notation; thus, having learned that B
plus E equals Z (2 + 5 = 7) does not make it immediately obvious that K plus N equals O
(20 + 50 = 70). Egyptian scripts used an analogous non-place-value decimal notation,
though the individual symbols were not alphabetic. A component of scribal training in all
periods was to learn how to perform arithmetical operations on numbers written in these
notations. Numerical tables were extensively used in both the teaching and practice of
arithmetic (Fowler 1999, 234–240, and see 268–276 for an inventory of papyri).

The most basic task was to learn the sums and products of pairs of units, tens, and hun­
dreds. We have about twenty specimens from the Graeco-Roman period—on papyrus, as
well as on wooden or waxed tablets—of addition and multiplication tables. A multiplica­
tion table for the number 7, for example, might give the results of multiplying 7 by 1, 2, 3,
… 9, 10, 20, 30, … 90, 100, 200, 300, … 900, 1,000, 2,000, 3,000, … 9,000, and 10,000.
Performing arithmetical operations by means of tables would, of course, have been slow­
er and more cumbersome than relying on memory; one was probably expected to memo­
rize these tables rather than retain them as ready reference.

Division presented problems that were more vexatious for the scribe (though more inter­
esting for the modern mathematician) because of the way that fractions were expressed,
which was different from both our modern fractions, like 3/5, and our modern decimals,
like 0.6. This notation, which was originally Egyptian but later adapted to the Greek Ionic
numerals, required any fraction to be broken down into the sum of a series of distinct, so-
called unit fractions, that is, ½, ⅓ ¼… (but a special symbol for ⅔ was also available).
Thus, the result of dividing three by five, or six by ten, would be written as a symbol rep­
resenting a half followed by the numeral for 10 (I), both marked as fractions by a raised
stroke, because 3/5 = 1/2 +1/10. Because it is by no means a trivial problem to find a set
of unit fractions that add up to a given quotient, tables were made that listed the results
of dividing, say, 10 into 1, 2, 3, … 9, 10, 20, 30, … 90 (and so forth). Such tables are about
twice as common as addition and multiplication tables not because one had to do (p. 341)
division more often than the other operations but probably because the appropriate com­
binations of unit fractions were harder to commit to memory.

Aside from tables, the commonest form that mathematical texts take in the papyri is as
collections of problems. In Egypt a tradition of problem texts can be traced back to a half
dozen or so mathematical texts preserved from the first half of the second millennium

Page 3 of 20
Mathematics, Science, and Medicine in the Papyri

BCE, among which are two substantial and important hieratic rolls known as the Moscow
and Rhind mathematical papyri (Gillings 1972; Chace et al. 1927–1929; Struve 1930). Fol­
lowing these, there is a long gap in the documentation, after which it resumes in the Hel­
lenistic period and continues to the Byzantine. A regular feature of problem texts,
whether early or late and regardless of the language in which they are expressed, is that
they illustrate methods of solution by specific numerical examples rather than prescrib­
ing procedures in general terms, and no argument is provided to justify the procedures.
The simplest problems and, one surmises, the most practically useful are metrological
conversions and the calculation of rectangular areas and volumes, where the handling of
the chaotic systems of units of measure (see chapter 8) is again as much a part of the les­
son as the principle that area or volume is the product of the linear dimensions.

The problem texts are algorithmic: a step-by-step procedure is carried out on a particular
set of data so that the student can learn the pattern and apply the same steps to different
data. Texts practically never present a method in general terms and say, for instance, “to
get the area of any rectangle, you multiply the length by the width”; instead, one might
have a series of problems that deal with specific rectangular fields with stated numbers
for the length and width. (Much of elementary mathematics is still learned this way.)
Complex problems are distinguished from simple ones only by the number of steps to be
performed. Geometrical problem texts are accompanied—intermittently in the Pharaonic
papyri, regularly in the Graeco-Roman—by diagrams that schematically display a particu­
lar situation and are often labeled with numerals that represent the various dimensions.

As an illustration of a problem text we may take an early specimen, Problem 44 of the

Rhind papyrus, which asks the reader to calculate the volume of grain in a cubical box,
the length, width, and depth of which are all 10 cubits. The text instructs us to multiply
10 by 10, getting 100, and then to multiply this by 10, getting 1,000. Then we are told to
add half again of this result, getting 1,500, which, the text says, is the capacity of the box
in a unit called khar. Finally, we are told to take a twentieth of this figure, that is, 75, and
this is the capacity of the box in a unit called a quadruple hekat. Here the strictly geomet­
rical aspect of the problem—finding the number of cubic cubits the box contains—is dealt
with in the first two multiplications, followed by the metrological aspect of expressing this
quantity in conventional units of capacity. Essentially the same procedure is followed for
Problem 41, which asks for the volume of grain in a cylindrical container with a diameter
of 9 and a height of 10, except that the different shape requires a different formula for
the volume in cubic cubits: One subtracts a ninth from the diameter, (p. 342) multiplies
this first by itself and then by the height. Typically for this genre of text, no indication is
given that the formula is an approximation. The style of writing is in general very terse;
much of the explaining must have been oral.

Alongside such practically useful formulae, one finds throughout the tradition problems
that, although expressed in terms of real-world objects, can only be regarded as puzzles
or exercises since one could not easily imagine analogous situations arising outside of the
classroom. Thus in the Rhind papyrus the genuinely practical Problem 63 (to divide seven
hundred loaves among four workers according to specified pay scales) immediately fol­

Page 4 of 20
Mathematics, Science, and Medicine in the Papyri

lows a puzzle in which one is asked to determine the quantities of gold, silver, and lead in
a bag from the price paid for it, all of this subject to certain unreal conditions that give
just the right information to make the problem soluble. In later papyri, where the reper­
toire of mathematical techniques is enlarged relative to the pharaonic texts, probably un­
der Babylonian or Greek influence, the new methods give rise to more and more unreal
situations. A third-century-BCE demotic papyrus, P.Cairo J.E. 89127−30 + 89137−43
(Parker 1972, 41), contains a problem (no. 33) in which a square piece of land of known
area is miraculously rearranged as a circle, and we are asked to find its diameter: The so­
lution is in part an exercise in finding an approximate square root, a technique probably
borrowed from Mesopotamia. A Greek papyrus from the first or second century CE, P.Chic.
3 (Goodspeed 1898), consists of problems in which one has to find the areas of irregularly
shaped plots of land that (conveniently) can always be broken down into a few right-an­
gled triangles so that the solution is obtainable through the theorem of Pythagoras. One
could cite comparable instances from many manuscripts. The emphasis on impractical
problems in the mathematical papyri raises interesting questions about the goals and ra­
tionale of the mathematical education of scribes in antiquity.

It goes without saying that documentary papyri contain immense quantities of numerical
data, some of which resulted from calculations. Recent studies of pharaonic Egyptian
mathematics (Imhausen 2003a, 2003b) have for the first time brought such evidence to
bear on the interpretation of “administrative mathematics” as found in the problem texts,
showing how extensively the framework of the problems is determined by the profession­
al practice of scribes and accountants. The Graeco-Roman material deserves to be sub­
jected to this kind of analysis.

Academic Mathematics

Compared to the problem texts, the known papyri that contain Greek mathematics in the
proof-oriented tradition are very few indeed. Most attention to date has been given to the
papyri that include parts of the most famous of all Greek mathematical treatises, Euclid's
Elements. Four have been published so far, and more certainly lurk in the major collec­
tions (Fowler 1999, 209–217). All the known papyri of the Elements date from the interval
between the late first and the third centuries CE, thus at least (p. 343) three centuries af­
ter Euclid's time. By itself this fact would be dangerous as a basis for hypotheses, but it
turns out that the earliest definite references to Euclid's Elements in Graeco-Roman au­
thors are from about the same time as the earliest of the papyri, so it seems likely that
the book's adoption as a standard school text was, surprisingly, this late. It was likely part
of the process whereby deductive geometry became a curricular subject taught more
widely if also more superficially than it had been in the Hellenistic period. Three of the
four Euclidean papyri certainly do not derive from complete copies but present excerpts,
apparently intended for some didactic purpose. The fourth, P.Fay. 9, might have belonged
to a full text but presents significant variants—including the omission or displacement of
an entire theorem—with respect to the medieval textual tradition. We are still far from

Page 5 of 20
Mathematics, Science, and Medicine in the Papyri

knowing how faithfully the Elements as it has come down to us represents what Euclid
wrote and how much our text owes to editors of the Roman period and late antiquity.

Deductive mathematical treatises discussed a small range of objects (e.g., points, lines,
circles) in a highly standardized idiom, but their arguments develop on a large scale.
While the distinctive terminology should make it easy in principle to identify a small frag­
ment of a manuscript as belonging to this genre, one would generally need a substantial
amount of text to take the identification further. Still, it is remarkable that no papyrus
fragment has yet been shown to belong to any identifiable work of the Greek mathemati­
cians except the Elements, not even a scrap of the most renowned of them, Archimedes.
This is in strong contrast to the prominence of the mathematical literature among the
secular texts preserved in Byzantine manuscripts or translated into Arabic around the
ninth century. Again we seem to be seeing evidence that serious study of academic math­
ematics was concentrated in just a few centers. Perversely, the most advanced mathemat­
ics preserved in a contemporary document from antiquity is a series of geometrical theo­
rems, with accompanying diagrams, written on a set of ostraca found at the distinctly re­
mote site of Elephantine (Mau and Müller 1960). These texts were written in the third
century BCE, that is, roughly contemporary with or barely later than Euclid himself, and
they concern the construction of the regular polyhedra inside a sphere, which is the sub­
ject matter of the culminating book 13 of the Elements.

Astronomy and Astrology

The Hellenistic Period

Leaving aside a few brief or incidental references to astronomical phenomena, chiefly of

calendrical interest, no astronomical texts survive on Egyptian papyri (p. 344) older than
the Hellenistic period. For that matter, few astronomical papyri in any language are pre­
served from the beginning of Greek rule until the second half of the first century BCE, but
they include some of exceptional importance. One reason for this is that our documenta­
tion of early Greek astronomy consists largely of reports and adaptations in texts of a lat­
er period. It is unlikely that it will ever be possible to write a history of ancient astronomy
that depends entirely on contemporary sources, but by their help one can get some sense
of the infidelities, selectiveness, and biases of the later testimony. The Hellenistic astro­
nomical papyri are quite diverse in genre and content, though certain themes recur, no­
tably an emphasis on annually recurring phenomena of the sun and stars, the relation­
ships between the sun and moon, and the regulation of calendars. Another notable fea­
ture is the presence, side by side, of specifically Egyptian elements with others that were
typical of the generality of Greek astronomy.

One early witness to this fusion of Greek and Egyptian material is P.Hib. I 27, which on in­
ternal and external grounds can be dated with certainty to about 300 BCE. The main body
of the text is a weather parapêgma, a variety of document highly characteristic of Greek
astronomy. A parapêgma is a list of dates of astronomical events that recur at intervals of

Page 6 of 20
Mathematics, Science, and Medicine in the Papyri

one solar year and dates of weather patterns that were also believed to recur annually.
The astronomical events included the solstices and equinoxes, and the dates when stars
and constellations were first seen to rise or set before dawn or after dusk; apparently,
these were believed to be signs, perhaps even causes, of the more regular weather
changes. Since they attempted to delineate a connection between the heavens and cer­
tain aspects of the immediate human environment, the parapêgmata came to be sub­
sumed under the category of astrology by certain Roman-period writers such as Ptolemy.

It is commonly supposed that astronomers as early as the fifth century BCE produced
parapêgmata, but the Hibeh parapêgma is the oldest one extant. It is also the earliest dat­
able Greek text that clearly refers to the signs of the zodiac (i.e., the division of the zodiac
into equal twelfths as distinct from the zodiacal constellations). Unlike the later Hellenis­
tic examples preserved through the manuscript tradition or as inscriptions, the Hibeh
parapêgma incorporates the dates of many religious festivals—both Greek and Egyptian,
some local to Sais—and stages of the Nile's flood cycle, and all the dates are specified ac­
cording to the Egyptian calendar, which in the short term (on the order of a decade or
two) could function as an ersatz solar calendar. A prefatory letter purports that the con­
tents of the papyrus were taught to the author, a man who had lived in the Saite nome for
five years, by a “wise man in Sais,” who apparently was an Egyptian, but the presence of
phrases that are repeated verbatim in the “Eudoxus” papyrus (discussed later) suggests
that this claimed pedigree is fictitious.

The Egyptian calendar proved to be Egypt's most important and enduring contribution to
the development of astronomy; its strictly uniform months and years had obvious advan­
tages over the Greek lunar calendars for accurate time- reckoning. (p. 345) Early Greek ef­
forts to coordinate the two kinds of calendar are reflected in P.Ryl.IV 589, an early sec­
ond-century-BCE account of debts to which has been appended a list of Egyptian calendar
dates that correspond to the first days of lunar months. The months in question are ac­
cording to the Egyptian reckoning, beginning with the morning disappearance of the
waning moon rather than the evening appearance of the new moon crescent, and the
twenty-five-year cycle by which the dates were determined also seems to have been
known in Egypt before the Ptolemaic period. A small scrap of text from third-century-BCE
Gurob, P.Petrie III 134, attests to Greek interest in the decans, constellations traditionally
used in Egypt for time-reckoning at night, and the problem of correlating the decans with
the Egyptian calendar.

The most famous of all astronomical papyri, P.Par. 1 (first half of second century BCE), is
commonly known as the “Eudoxus papyrus,” or Ars Eudoxi (Blass 1887). This is a render­
ing of the words EϒΔOΞOϒ TEXNH, “art of Eudoxus,” which are spelled out by the initial
letters of an acrostic iambic poem on the back of the roll and which may or may not have
been composed by the same individual as the text on the front. The lengths of the poem's
twelve lines also manage to encode the structure of the Egyptian calendar year's months
and days. The name “Eudoxus” here stands as a metonym for astronomy—the great con­

Page 7 of 20
Mathematics, Science, and Medicine in the Papyri

temporary of Plato had nothing to do with the composition of the papyrus, although he is
one of several authorities cited in it.

The preserved part of the roll, almost two meters long, contains on its front the final
twenty-four columns of an astronomical treatise with interspersed figures; not counting
geometrical diagrams, this is the oldest known illustrated Greek manuscript. The figures,
which are obscure and for the most part have little or no relation to the text, combine fea­
tures of Greek cosmology (circles that represent a spherical earth, sun, and moon and the
zodiac) and Egyptian imagery (scarab, Horus, scorpion). Though laid out as prose, the
treatise was originally composed at least partly in verse. The purported author is identi­
fied in the final column as one Leptines. A wide range of topics is discussed at an elemen­
tary level: the most obvious periodicities of the sun, moon, and planets; the risings and
settings of the fixed stars; an eight-year cycle for a lunar calendar (but not the more accu­
rate nineteen-year cycle, though this was certainly familiar in the Hellenistic period); a
crude geocentric cosmology; the causes of the moon's phases and eclipses; and the rela­
tive sizes of the sun, moon, and earth. As a whole the work does not seem to be an ade­
quate index of what astronomers were doing in the early second century, but it is proba­
bly a fair portrayal of what an educated layperson knew about the subject.

An early first-century-BCE papyrus from Abusir al Melek, P.Berl. 13146 +13147, is the
first known witness to the kind of astronomy that would become predominant in the pa­
pyri—and indeed in the Graeco-Roman world—during the Roman period (Neugebauer,
Parker, and Zauzich 1981). Unlike the foregoing texts, it is in (p. 346) demotic Egyptian,
though there are indications that it was adapted from a fuller model in Greek. The back
has a text that describes a procedure for determining Egyptian calendar dates of solstices
and equinoxes, but the important text is on the front, a list of predictions of the dates and
circumstances of lunar eclipses during the years 85–74 BCE. (The papyrus was actually
written after these dates and incorporates some details apparently derived from observa­
tions.) These are older than any other extant Greek or Egyptian predictions of complex
astronomical phenomena. The papyrus does not reveal the purpose of the eclipse predic­
tions, but almost certainly they were connected with the—ultimately Mesopotamian—tra­
dition of interpreting eclipses as astrologically ominous events, so this may also be the
oldest astronomical papyrus motivated by astrology.

The Roman Period

Astrology, as a coherent theory that systematically describes the effects of the instanta­
neous state of the heavens on human lives and characters seems to have come into exis­
tence about 100 BCE, probably in Egypt (Pingree 1997). It spread rapidly throughout the
Graeco-Roman world during the following century, thereby creating a demand for astro­
nomical data and in particular the locations of the heavenly bodies relative to the zodiac
and to the local horizon at any given date and time. The abundance of Roman-period as­
trological and astronomical papyri implies that astrologers existed in large numbers

Page 8 of 20
Mathematics, Science, and Medicine in the Papyri

throughout Egypt; yet curiously they are almost invisible in documentary texts. A passing
mention of a “one-eyed astrologer [astrologos]” in P.Oxy. LXI 4126 appears to be unique.

The most common manifestation of the bond between astrology and astronomy was the
personal horoscope, a document that states the birthdate of an individual and lists the zo­
diacal locations of the sun, the moon, and the five planets known in antiquity, as well as
the point of the zodiac that was rising on the horizon at the date and time of birth. Horo­
scopes were obviously astrological in purpose—pronouncements about the life of the indi­
vidual were based on it—but purely astronomical in content. Astrologers used the astro­
nomical data in a horoscope as the basis for making forecasts about the individual whose
birth was recorded in it, but this interpretation was never put into writing.

The authors of treatises on astrology sometimes incorporated genuine horoscopes in their

works to illustrate techniques of interpretation. One second-century textbook that sur­
vived through the medieval tradition, the Anthologies of Vettius Valens, contains about a
hundred horoscopes from the author's files, but most works limited themselves to one or
two, relying for the remaining part on contrived textbook examples. Horoscopes on pa­
pyrus, by way of contrast, are authentic records from consultations with astrologers. They
are among the more frequently encountered personal documents. When O. Neugebauer
and H. B. van (p. 347) Hoesen published the first comprehensive collection of Greek horo­
scopes from all sources (Neugebauer and van Hoesen 1959), they included roughly fifty
horoscopic papyri and ostraca. The number of extant specimens—mostly Greek, a handful
demotic—is now approaching two hundred (see especially Baccani 1992 and P.Oxy.Astron.
4236–4300a in Jones 1999a). They were evidently produced by numerous horoscope cast­
ers working in many localities over a long period and hence give a more reliable picture
of the activity of astrologers (valid at least for Egypt, if not for the whole Roman Empire)
than the “literary” sources.

The format of most horoscopes from all periods is highly standardized; P. Oxy. Astron.
4264 (Jones 1999a, 2.402–403), for an unnamed individual born on the morning of April
18, 300 CE, is a typical example:

Good fortune. Nativity.

Year 16 of Diocletian,
Pharmouthi 23, hour 4 of day.
Ascendant and Mars in Gemini.
Saturn in Leo.
Moon in Virgo.
Jupiter in Capricorn.
Sun, Mercury, and Venus in Aries.
Good luck!

Thus we have, in order, a brief introductory formula (optional), the name of the individual (op­
tional, and omitted in this example), the person's birthdate and time of birth, a list of the seven
heavenly bodies and the ascendant point (hôroskopos) with their positions in the zodiac at the
time of birth, and a brief closing formula (optional). For the positions most horoscopes, like this
one, give only the name of the relevant sign of the zodiac. Occasionally one finds more precise

Page 9 of 20
Mathematics, Science, and Medicine in the Papyri

locations in degrees or degrees and minutes counted from the beginning of the sign (each sign is
thirty degrees long). The typical medium was a scrap or ticket of papyrus, perhaps reused, and
there is considerable variation in the quality of hands and standard of orthography. A few horo­
scopes are more elaborate, stretching out the information by means of a formulaic text that pro­
vides a full paragraph for each heavenly body, and such horoscopes—surely reserved for the
wealthier customers—are written out in bookhands on rolls that must have been two to three
meters long when complete.
The papyrus horoscopes always gave the birthdate of the person in question in this form:
regnal year, Egyptian month and day, and hour of day or night. Even when the date for­
mula is damaged or lost, it can usually be reconstructed from the astronomical data. This
is not because the stated positions of the heavenly bodies are usually very accurate (they
are typically correct only to within five or ten (p. 348) degrees, and larger errors are not
uncommon). Rather, because each planet has a different periodicity for its revolution
around the zodiac, the combined information derived from a set of individually imprecise
positions can suffice to narrow down the possible dates within the historically and palaeo­
graphically admissible interval to a unique dating accurate to within one day (Jones
1999a, 1.47–52).

The document itself might have been produced anywhere from within days of the birth­
date (supposing it was obtained on behalf of a newborn child) to at most several decades
later since only in theoretical writings would there be an interest in finding the horo­
scopes of deceased people. The earliest known horoscope, a demotic ostracon, is for
someone born in 38 BCE, and the latest, a Greek papyrus, is for someone born in 508 CE.
In between, the numbers of horoscopes peak between 150 and 300, although this pattern
probably reflects only the relatively high survival rate of papyri of all kinds during that
period. The general impression is that, once horoscopic astrology became established in
Egypt in the second half of the first century BCE, it remained steadily popular until the
end of the fourth century, when a marked decline began (ibid., 1.5–6).

In the ancient debates about the validity of astrology, one sometimes encounters a carica­
ture of a horoscope caster as someone who watches the sky, waiting to record the places
of the heavenly bodies at the instant when he hears a signal that a child has been born.
The reality as exposed in the papyri is not at all like that. The astronomical information in
horoscopes and other astrologically oriented documents was not obtained by observation
but rather by calculation from various kinds of theory. The majority of the astrologers
likely knew little technical astronomy, but they had written resources, especially tables,
that supplied the data they needed.

Many fragments of astrologers' tables are extant (in addition to P.Oxy.Astron. 4148–4235
in ibid., see 1.301–307 for a checklist of astronomical papyri published elsewhere, among
which are nearly fifty tables). The tables consist mostly of numbers, sometimes accompa­
nied by names of the signs of the zodiac (when the information consists of the positions of
a heavenly body) or names of months (when the information consists of dates), but both
zodiacal signs and months are also sometimes coded as numbers running from one to
twelve. All these data are often laid out in a tabular grid of black or red lines. In many ta­
bles the successive rows represent steps of progressing time (e.g., successive days or
Page 10 of 20
Mathematics, Science, and Medicine in the Papyri

years), while the various columns contain different categories of data associated with
those dates. One of the techniques for identifying and analyzing astronomical tables is to
look for mathematical patterns in the numbers as one reads down a column or set of
columns, for example, determining whether the numbers increase or decrease by con­
stant or roughly constant amounts and, if so, by how much. Each of the heavenly bodies
has characteristic patterns of motion that manifest themselves in such numerical pat­
terns.

The tables also tend to conform to certain established formats (Jones 1999b). The most
frequently attested variety of astronomical table, the sign-entry almanac, (p. 349) is a ta­
ble of the dates when, according to computation, each planet crosses from one sign of the
zodiac to a neighboring sign. We have examples of sign-entry almanacs covering intervals
from as early as the late first century BCE and as late as the fourth century CE. Such a ta­
ble, which could extend over several decades of planetary motion, would make it trivially
easy to find the zodiacal signs occupied by each planet on a given date, though it would
be of no use for obtaining more precise positions in degrees. Another common type of ta­
ble, the ephemeris, gives computed positions for all seven heavenly bodies for every sin­
gle day in a succession of months and years. The known examples are spread over an
even wider temporal span, from the first century BCE to the fifth century CE. The princi­
pal application of ephemerides was probably not in casting horoscopes, however, but in
the so-called catarchic branch of astrology, which professed to determine whether any
particular day was auspicious or inauspicious for different activities. Two or three other
standardized formats of almanac are attested, in which computed planetary positions are
given at regular intervals. With the sole exception of the ephemerides, which are men­
tioned in a few astrological texts dating from late antiquity, none of these varieties of ta­
ble is preserved or discussed in the literature preserved through the medieval manuscript
tradition.

The first questions that naturally occur to a historian of astronomy are, how were these
almanacs and ephemerides calculated, and what kind of theories of planetary motion un­
derlie them? Unfortunately, it is generally difficult or impossible to work backward from
the computed results in such tables to the methods of computation, especially given the
fragmentary condition of the papyri. On the other hand, we know of other varieties of ta­
bles (and a few instructional texts that describe how to make or use them) that have a
closer connection to the astronomical theories. These are analogous in function to the as­
tronomical tables preserved by the medieval tradition, which are chiefly those incorporat­
ed in two second-century-CE works by Ptolemy: his treatise Mathematical Syntaxis (now
usually called by its medieval nickname, the Almagest), and the Handy Tables. Like
Ptolemy's tables, the ones on papyrus would have been employed only by the more tech­
nically competent and sophisticated astrologers to find precise positions of the heavenly
bodies, the kind of data recorded in the more elaborate horoscopes. The discovery of
these “primary” tables is among the most important developments in the study of ancient
science in the last generation; it has cast light on a part of Greek astronomy, the exis­
tence of which is scarcely hinted at in the medieval tradition.

Page 11 of 20
Mathematics, Science, and Medicine in the Papyri

Before such tables were encountered, what we knew about the theories of motion of the
heavenly bodies current during the Roman period was practically limited to the works of
Ptolemy. It was assumed that Ptolemy's approach in the Almagest was representative of
his contemporaries and predecessors, at least as far back as Hipparchus (second century
BCE). Ptolemy's astronomy is geocentric, and its basic principle is that the heavenly bod­
ies move around a stationary earth along (p. 350) geometrical paths compounded out of
circular motions. For example, a planet might be assumed to revolve with uniform speed
around a relatively small circle called an epicycle, which itself revolves uniformly around
the earth, thereby making it appear that the planet varies in speed and even periodically
reverses the direction of its motion through the zodiac. The object of the theory was to es­
tablish the sizes, speeds, and positions of the various revolutions consistent with observa­
tions. Tables based on such geometrical theories or models use what we would now call
trigonometrical functions to represent the way that an observer on earth sees these circu­
lar motions. Largely through the Almagest's influence, Islamic, Byzantine, and western
European astronomy continued along similar lines until the seventeenth century.

Soon after the decipherment of Mesopotamian cuneiform script in the late nineteenth
century, it was discovered that astronomers in Babylonia during the last four centuries
BCE had developed planetary theories that operated on entirely different principles.
Babylonian astronomy employed sophisticated combinations built up from simple arith­
metical sequences but without trigonometry to reproduce the observed patterns of astro­
nomical phenomena. Babylonia turned out to be the source of certain conventions of
Greek astronomy, such as the division of the zodiac into twelve signs of thirty degrees
each; the representation of fractions as sixtieths (“minutes”), sixtieths of sixtieths (“sec­
onds”), and so forth; and a handful of specific significant numbers such as a highly accu­
rate estimate of the average length of the lunar month. Nonetheless, the dependence of
the Greek tradition on the Babylonian did not appear likely to extend beyond these impor­
tant but elementary borrowings (Neugebauer 1975, 2.601–607).

Since the 1980s, about twenty Greek papyri have come to light that reveal that, from the
first through the fourth centuries CE, a substantial part of the most complex Babylonian
arithmetical theories were used in Egypt in forms adapted to accommodate the Egyptian
calendar instead of the Babylonian lunar calendar but otherwise practically unchanged.
The first of these “Graeco-Babylonian” papyri (and still perhaps the most astonishing)
turned up in a private collection and was irst published by Neugebauer, who fully recog­
nized its revolutionary significance (Neugebauer 1988; see also Jones 1997). It is a frag­
ment of a very elaborate kind of computation of the circumstances of new or full moons,
familiar from cuneiform tablets of the last three centuries BCE and known as “system B.”
Other tables found among the Oxyrhynchus papyri proved to be computations of dates
and zodiacal positions of planetary phenomena, such as first visibilities and stationary
points, in which the same methods of calculation were used that we find in the cuneiform
tablets; the majority of the Babylonian planetary models are now identifiable in Greek pa­
pyri (P.Oxy.Astron. 4152–4161 in Jones 1999a). The Babylonian methods seem in fact to
have been more widely employed than methods based on theories of circular motion until
the third century CE. After that time, Ptolemy's astronomical tables began to compete vig­
Page 12 of 20
Mathematics, Science, and Medicine in the Papyri

orously with them, as is attested by several fragments of (p. 351) papyrus manuscripts of
his Handy Tables, but the Babylonian methods seem to disappear from the record only af­
ter the fourth century—yet there is scarcely a hint of knowledge of them in any of the lit­
erature from the medieval tradition. Historians are only beginning to explore the conse­
quences of this new picture of rival traditions and to speculate about the channels of
transmission.

In comparison to the mass of known astronomical tables and texts related to prediction
and astrological applications (now approaching two hundred papyri), the papyri that con­
tain genuine theoretical texts constitute a mere handful (Jones 2003). For the most part,
the astrologers of Roman Egypt did not devote much space in their libraries to astronomi­
cal theory; it is noteworthy that no fragment even of the Almagest, the work in which
Ptolemy presented the empirical reasoning behind his predictive tables, has shown up on
papyrus. Nonetheless, we do possess two disconnected passages from a treatise compara­
ble to Ptolemy's but written a generation earlier, about 105 CE, in P.Oxy.Astron. 4133 and
PSI XV 1490, but these, though of enormous historical interest, are altogether exception­
al in the papyrus record. The passage in P.Oxy.Astron. 4133 discusses an analysis of a pair
of observations of the position of Jupiter relative to certain stars in the constellation Can­
cer, one made by unidentified astronomers in 241 BCE and the other by the author him­
self in 105 CE. These are the only known dated astronomical observations preserved in a
Greek source earlier than Ptolemy. The other passage concerns the construction of tables
for calculating the sun's zodiacal position on the basis of a theory of the sun's motion
around the earth that was geometrical but significantly different from Ptolemy's.

Paralleling the situation with the horoscopes, Egyptian-language astronomical papyri are
vastly outnumbered by those in Greek, but a few demotic specimens are known of both
the “almanac” and “primary” varieties of table. Horoscope casting is documented in the
ostraca from the temple at Medinet Madi, in both Greek and demotic, and more circum­
stantial evidence exists that astrology was practiced in a bilingual temple setting at Teb­
tynis (Jones 1994). The Egyptian scribal traditions of the Roman period also preserved
texts relating to the older, nonmathematical and nonastrological astronomy of the
pharaonic period, most impressively in P.Carlsberg 1, an extensive commentary on astro­
nomical texts and pictures that independently survive in some New Kingdom royal tombs
(Neugebauer and Parker 1969).

Many strictly astrological papyri other than horoscopes have been published, and more
await study in the major collections (checklist in Baccani 1992, 32–34). They have not as
yet altered our understanding of the evolution of Graeco-Roman astrology as much as the
astronomical papyri have affected the historiography of astronomy—perhaps reflecting
the fact that the study of ancient astrology still is more preoccupied with the discovery
and editing of texts than with historical analysis of their contents. Yet the potential of the
papyri for improving our understanding of the early development of astrology is consider­
able since (p. 352) very little of the medievally transmitted literature is older than the sec­
ond century CE. A large portion of the astrological papyri derive from handbooks on the
interpretation of horoscopes, similar in character to much of the astrological literature

Page 13 of 20
Mathematics, Science, and Medicine in the Papyri

that survives through the medieval manuscript tradition, though specific textual overlaps
are rare. Several of the popular astrological handbooks, especially in the earliest period,
were written in verse, and papyri have made substantial contributions to the known cor­
pus of astrological verse, including notably several fragments in elegiac meter attribut­
able to the first-century author Anubio (Obbink 2004).

Special interest attaches to a small number of papyri, both Greek and demotic, that relate
to so-called general astrology. While horoscopic astrology focused on the individual, gen­
eral astrology attempted forecasts of events and circumstances pertaining to entire peo­
ples and countries, for example, by interpreting the appearance of eclipses or the posi­
tions of the planets at the annual first rising of Sirius. Several texts help to trace a histori­
cal connection between this branch of astrology and the much older Mesopotamian tradi­
tion of astral omens, documented in cuneiform texts as early as the Old Babylonian peri­
od (early second millennium BCE). Most impressively, the Roman-period demotic papyrus
P.Vind. 6278 + preserves Egyptian eclipse omen texts of a distinctly Babylonian flavor
that can be dated to approximately 500 BCE, which suggests that Egypt was an important
intermediary in the transmission of this lore even before the Hellenistic period (Parker
1959). The permeability of the language barrier between Egyptian and Greek is shown by
textual coincidences between omen interpretations based on the rising of Sirius in the de­
motic P.Cairo 31222 (Hughes 1951) and the Greek P.Oxy. LXV 4471.

Medicine
The Egyptian Tradition

One can draw obvious parallels between the role of papyri in the historiography of an­
cient medicine and in that of ancient mathematics. From the second millennium BCE we
have parts of several (mostly) hieratic medical manuscripts, including two substantial
rolls, and these are our principal sources of knowledge of Egyptian medicine (Westendorf
1999, 1.6–79). The Graeco-Roman period is represented by a few demotic medical manu­
scripts and considerably more written in Greek, and we also have a generous literature in
Greek and Latin transmitted through the medieval tradition. However, whereas the de­
motic and Greek mathematical problem texts contain essentially the same style of mathe­
matics, the language divide was (p. 353) more significant in medicine; thus, the demotic
medical papyri appear as a continuation of older Egyptian medicine operating in parallel
but scarcely interacting with the Greek medicine that was imported to Egypt during the
Hellenistic period. Moreover, Greek medical authors and texts identifiable from the me­
dieval tradition are much better represented in the papyri than the mathematical authors.

The dozen extant medical manuscripts from the second millennium are all instructional or
reference texts. None has a known author, and most can be shown on textual grounds to
be copies rather than autographs. Each is composed as a collection of shorter and more
or less independent sections (cf. again the mathematical problem texts) that tend to fol­
low standardized patterns according to their purpose. In particular, there was an estab­

Page 14 of 20
Mathematics, Science, and Medicine in the Papyri

lished way of prescribing a diagnosis and treatment, which in its full form comprised a
descriptive title (“treatment for such-and-such affliction”), a second-person conditional
that described details of the case (“If you investigate a man with such-and-such afflic­
tion…”), a diagnosis (“this is a case of…”), a verdict (“Then you will say: an affliction with
which I will struggle”), and a course of treatment. Such texts, as well as the structurally
simpler directions for medicines, account for the majority of the material in the papyri.
Large sections of a manuscript or perhaps an entire roll may be dedicated to a special­
ized subject. Thus, one of the oldest of the medical papyri, Ramesseum V (Twelfth Dy­
nasty, nineteenth century BCE), contains recipes for illnesses of the “vessels” (a term that
describes various stringlike or tubular entities, e.g., muscles), while the roughly contem­
porary Kahun papyrus is devoted to women's afflictions. Other manuscripts have more
eclectic contents.

The Edwin Smith and Ebers papyri, which are the largest and most historically significant
of the extant Egyptian medical manuscripts, may serve as illustrations of the varieties of
instructional text that a physician might possess or have access to. They both date from
the early New Kingdom (sixteenth century BCE); both were purchased at Luxor by the
American collector Edwin Smith in or before 1862, along with fragments of the Rhind
mathematical papyrus; it is possible that all three manuscripts were discovered in a sin­
gle library or archive. The Ebers papyrus is a calligraphic and massive roll, some twenty
meters long, containing nearly nine hundred text sections in a hundred columns of text on
the front side and a further eight columns on the back. It is an eclectic collection compris­
ing blocks of varying numbers of texts that share a common format and subject matter;
for example, one section consists of about a hundred recipes for ailments of the belly,
though most of the sections are much briefer. There are comparatively few of the formal
diagnosis-and-treatment texts. A remarkable feature, unique to this papyrus, is a theoreti­
cal section describing the connections that exist between the heart and the “vessels.” The
Smith papyrus is considerably shorter (not quite five meters long), and the seventeen
columns of text on its front are devoted to a collection of diagnosis-and-treatment texts
unified by their subject matter—the surgical treatment of wounds—and by a principle of
organization according to the part of the (p. 354) body affected, beginning with the head
and progressing downward. On the back, the same hand has written a short collection of
spells against epidemics and a few medical and cosmetic recipes.

The pharaonic medical papyri exhibit a broader chronological spread than the mathemati­
cal papyri, and although there is again a documentary gap between the second-millenni­
um Egyptian texts and those of the Graeco-Roman period, it is much shorter, extending
only about seven centuries and centered on the first half of the first millennium BCE. (The
oldest of the late manuscripts may in fact belong to the end of the Persian period.) The
most obvious questions raised by the Hellenistic- and Roman-period texts are, how much
continuity do they show with the pharaonic texts, and do they give any signs of interac­
tion between the indigenous medical knowledge and foreign (Near Eastern, then Greek)
medicines? At the present state of scholarship it appears that, at least as a literary tradi­
tion (and that is primarily what the papyri give evidence of), Hellenistic-Egyptian medi­
cine followed much the same lines as in the second millennium, adhering to the same
Page 15 of 20
Mathematics, Science, and Medicine in the Papyri

forms of expression for procedures and recipes. The corpus of texts is still small and frag­
mentary, however, and one would not be surprised if in due course incontrovertible in­
stances came to light of material transmitted from Greek to Egyptian or vice versa, paral­
leling what we know happened in mathematics, astronomy, and astral divination.

The Graeco-roman Tradition

Between two and three hundred Greek papyri with medical contents are currently known
(Andorlini 1993). The projected Corpus dei Papiri Greci di Medicina will include editions
or reeditions of all this material (Andorlini 1997, 2001, 2004). Additionally, documentary
papyri occasionally refer to subjects related in varying degrees to the practice of medi­
cine, and a number of papyri mention individual physicians (Sudhoff 1909, 254–275).
Physicians were thus much more publicly visible than astrologers; the comparatively
abundant epigraphical documentation of physicians tells much the same story for the
Mediterranean world as a whole.

Ephemeral medical documents, which are of course entirely missing from the corpus of
medical literature in the medieval tradition, abound in the form of small papyrus and
parchment fragments and ostraca. They are predominantly artifacts of the pharmacologi­
cal aspect of medical practice: labels of medicaments (the surviving specimens are all
from late antiquity), catalogues of products, prescriptions, and single recipes. While some
of these documents were surely issued to patients to deal with specific ailments, we also
find collections of recipes, typically grouped by the class of affliction or affected part of
the body, and these evidently served as reference texts. Most of these practically oriented
collections would not have been authorial, published treatises but rather personal and
traditional compilations.

Physicians apparently did possess manuscripts of the medical authorities of the

(p. 355)

past, and their tastes were not radically different from those of the scholars whose li­
braries fed into the medieval manuscript tradition. Among the roughly thirty known pa­
pyri that represent fragments of works that have otherwise survived, nearly two thirds
are from works in the Hippocratic corpus. In view of the fact that these include several
manuscripts of the Epidemics and the Letters, texts that can have had little practical ap­
plication, we can infer an interest in the scholarship of the profession detached from
everyday utility. Galen is the next most popular author, though lagging far behind Hip­
pocrates, and it is an interesting measure of how soon his fame spread to Egypt that we
find a copy of his De Placitis from Hermopolis, P.Münch II 43 + BKT IX 42, dated palaeo­
graphically to the beginning of the third century, thus remarkably close to the author's
lifetime (Hanson 1985).

More interest naturally adheres to texts that did not survive the medieval transmission.
Considering the mere quantity of text in them, what the papyri add to the body of “liter­
ary” medical treatises available to us is proportionately very small, but this is no ade­
quate measure of their historical value. A major problem that the selectivity of the me­
dieval manuscript tradition presents us is the complete absence of original medical writ­

Page 16 of 20
Mathematics, Science, and Medicine in the Papyri

ings from the Hellenistic period (except, probably, for one or two late items included in
the Hippocratic corpus). For what knowledge we have of the revolutionary anatomical
and physiological research programs carried out by Herophilus and Erasistratus in the
third century BCE, the subsequent rise of the medical sects, and the conflict between the
empiricists and their dogmatic opponents, we are heavily dependent on later reports,
above all in Galen's works. But Galen, though generous with his discussions and criti­
cisms of his predecessors, is perhaps not a wholly objective reporter.

Two examples suffice to illustrate the direct contribution of papyri to our evidence for
Hellenistic medical theory. A third-century BCE manuscript, fragments of which are now
in four different collections (P.Grenf.II 7b + P.Ryl.I 39 + P. Heid.inv. 401 + P.Hib.II 190),
contains a treatise on the physiology of vision that clearly reflects physical speculations
current at the time, for example, in the assumption of “pores” through which visual rays
emanate from the eye (Marganne 1994, 37–96); its authorship is disputed, but some con­
nection with the school of Erasistratus is plausible. In addition, P.Iand. V 82, from the first
century BCE, preserves parts of a treatise discussing the nomenclature of the female and
male reproductive organs, which again seems likely to have been composed by a member
of the Erasistratean school (Azzarello 2004).

The single most informative medical papyrus extant, however, known as the Anonymus
Londinensis (P.Lit.Lond. 165), is from the early Roman period, probably the first century
CE (Diels 1893; Manetti 1999). In part the importance of this manuscript is due to its ex­
ceptional preservation: Approximately 3½ meters of the roll, with thirty-nine columns of
text, survive more or less intact, together with smaller fragments. However, the treatise,
which according to current supposition is (p. 356) a draft autograph in an uncompleted
state, is also of exceptional intrinsic interest both as a source for early Greek medical the­
ory and as an exercise in scientific doxography. One long section, digesting a lost work as­
cribed to Aristotle, reviews and systematizes theories of the causes of disease attributed
to a number of named medical and philosophical authorities of the fifth and fourth cen­
turies BCE; this is followed by an inquiry into more specific physiological questions that
bring Hellenistic theories into play and attack the opinions of Erasistratus and his school.

Bibliography
Andorlini, I. 1993. “Lʼapporto dei papiri alla conoscenza della scienza medica antica.” Auf­
stieg und Niedergang der Römischen Welt: Geschichte und Kultur Roms im Spiegel der
neueren Forschung, ed. H. Temporini. Vol. 2.37.1, 458–562. New York: W. de Gruyter.

———, ed. 1997. “Specimina” per il Corpus dei Papiri Greci di Medicina: Atti dellʼIncontro
di studio, Firenze, 28–29 marzo 1996. Florence: Istituto papirologico G. Vitelli.

———, ed. 2001. Greek Medical Papyri. Vol. 1. Florence: Istituto papirologico G. Vitelli.

———, ed. 2004. Testi medici su papiro: Atti del Seminario di studio, Firenze, 3–4 giugno
2002. Florence: Istituto papirologico G. Vitelli.

Page 17 of 20
Mathematics, Science, and Medicine in the Papyri

Azzarello, G. 2004. “Pland V 82: Trattato sullʼapparato genitale e renale (?).” In Testi
medici su papiro: Atti del Seminario di studio, Firenze, 3–4 giugno 2002, ed. I. Andorlini,
237–256. Florence: Istituto papirologico G. Vitelli.

Baccani, D. 1992. Oroscopi greci: Documentazione papirologica. Messina, Italy: Sicania.

Blass, F. 1887. Eudoxi Ars Astronomica qualis in charta Aegyptiaca superest. Kiel; repr. in
ZPE 115: 79–101.

Chace, A. B., L. Bull, H. P. Manning, and R. C. Archibald, eds. 1927–1929. The Rhind
Mathematical Papyrus: British Museum 10057 and 10058. 2 vols. Oberlin, Ohio: Mathe­
matical Association of America.

Diels, H. 1893. Anonymi Londinensis ex Aristotelicis Iatricis Menoniis et aliis medicis

eclogae. Supplementum Aristotelicum 3.1. Berlin.

Fowler, D. 1999. The Mathematics of Plato's Academy: A New Reconstruction, 2d ed. Ox­
ford: Clarendon Press.

Gillings, R. J. 1972. Mathematics in the Time of the Pharaohs. Cambridge, Mass.: MIT
Press.

Goodspeed, E. J. 1898. “The Ayer Papyrus: A Mathematical Fragment.” American Journal

of Philology 19: 25–39.

Hanson, A. E. 1985. “Papyri of Medical Content.” Yale Classical Studies 27: 25–47.

Hughes, G. R. 1951. “A Demotic Astrological Text.” Journal of Near Eastern Studies 10:
256–261 and pl. x.

Imhausen, A. 2003a. Ägyptische Algorithmen: Eine Untersuchung zu den mittelägyptis­

chen mathematischen Aufgabentexten. Wiesbaden: Harrassowitz.

———. 2003b. Egyptian Mathematical Texts and Their Contexts. Science in Context: 16.

Jones, A. 1994. “The Place of Astronomy in Roman Egypt.” In The Sciences in Greco-Ro­
man Society, ed. T. D. Barnes, 25–51. Edmonton: Academic Printing and Publishing.

(p. 357) ———. 1997. “A Greek Papyrus Containing Babylonian Lunar Theory.” ZPE 119:
167–172.

———. 1999a. Astronomical Papyri from Oxyrhynchus. 2 vols. in 1. Philadelphia: Ameri­

can Philosophical Society.

———. 1999b. “A Classification of Astronomical Tables on Papyrus.” In Ancient Astronomy

and Celestial Divination, ed. N. M. Swerdlow, 299–340. Cambridge, Mass.: MIT Press.

———. 2003. “A Posy of Almagest Scholia.” Centaurus 45: 69–78.

Page 18 of 20
Mathematics, Science, and Medicine in the Papyri

Manetti, D. 1999. “ ‘Aristotle’ and the Role of Doxography in the Anonymus Londiniensis
(P.Br.Libr.inv. 137).” In Ancient Histories of Medicine: Essays in Medical Doxography and
Historiography in Classical Antiquity, ed. P. J. van der Eijk, 95–141. Leiden: Brill.

Marganne, M.-H. 1994. Lʼophtalmologie dans lʼÉgypte gréco-romaine dʼaprès les papyrus
littéraires grecs. Leiden: Brill.

Mau, J., and W. Müller. 1960. “Mathematische ostraka aus der Berliner Sammlung.”
Archiv für Papyrusforschung 17: 1–10.

Netz, R. 2002. “Greek Mathematicians: A Group Picture.” In Science and Mathematics in

Ancient Greek Culture, ed. C. J. Tuplin and T. E. Rihll, 196–216. New York: Oxford Univer­
sity Press.

Neugebauer, O. 1975. A History of Ancient Mathematical Astronomy. 3 vols. New York:

Springer.

———. 1988. “A Babylonian Lunar Ephemeris from Roman Egypt.” In A Scientific Human­
ist: Studies in Memory of Abraham Sachs, ed. E. Leichty, M. deJ. Ellis, and P. Gerardi,
301–304. Philadelphia: Samuel Noah Kramer Fund, University Museum.

———, and R. A. Parker. 1969. Egyptian Astronomical Texts. Vol. 3, Decans, Planets, Con­
stellations, and Zodiacs. Providence, R.I.: Published for Brown University Press by L.
Humphries, London.

Neugebauer, O., R. A. Parker, and K.-T. Zauzich. 1981. “A Demotic Lunar Eclipse Text of
the First Century B.C.” Proceedings of the American Philosophical Society 125: 312–327.

Neugebauer, O., and H. B. van Hoesen. 1959. Greek Horoscopes. Philadelphia: American
Philosophical Society.

Obbink, D. 2004. Anubio: Carmen astrologicum elegiacum. Leipzig: Saur.

Parker, R. A. 1959. A Vienna Demotic Papyrus on Eclipse- and Lunar-Omina. Providence,

R.I.: Brown University Press.

———. 1972. Demotic Mathematical Papyri. Providence, R.I.: Brown University Press.

Pingree, D. 1997. From Astral Omens to Astrology: From Babylon to Bikaner. Rome: Isti­
tuto italiano per lʼAfrica e lʼOriente.

Struve, W. W. 1930. “Mathematischer Papyrus des Staatlichen Museums der Schönen

Künste in Moskau.” Quellen und Studien zur Geschichte der Mathematik, Astronomie,
und Physik. Ser. A Vol. 1. Berlin: Springer.

Sudhoff, K. 1909. Ärztliches aus griechischen Papyrus-Urkunden: Bausteine zu einer

medizinischen Kulturgeschichte des Hellenismus. Leipzig: J. A. Barth.

Westendorf, W. 1999. Handbuch der altägyptischen Medizin. Vol. 1. Leiden: Brill.

Page 19 of 20
Mathematics, Science, and Medicine in the Papyri

Alexander Jones

Alexander Jones, Professor of the History of the Exact Sciences in Antiquity, Institute
for the Study of the Ancient World, New York University.

Page 20 of 20
The Range of Documentary Texts: Types and Categories

The Range of Documentary Texts: Types and Categories

Bernhard Palme
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0016

Abstract and Keywords

The diversity of texts and types of documents can easily create the impression that papy­
rological evidence offers a representative view on almost all aspects of life in antiquity.
But even in Egypt the geographical and chronological distribution of preserved papyri is
very uneven, concentrated on a few sites on the edges of ancient settlement. The present
state of editions differs greatly among linguistic groups. Conventionally the documentary
papyri and ostraca are divided into private and public. On the basis of historical and for­
mal features, private documents can be classified into private communications, records of
private legal transactions, accounts, and finances, and documents of piety and worship.
The second group contains documents concerning the interaction between the state and
individuals and pronouncements of the government and administration. Only the last
group was meant for public dissemination.

Keywords: documentary papyri, legal documents, papyrological evidence, Egypt, private documents

Why Some Things Are Documented and Others

Are Not
The diversity of texts and types of documents can easily create the impression that papy­
rological evidence offers a representative view on almost all aspects of life in antiquity.
But even in Egypt the geographical and chronological distribution of preserved papyri is
very uneven, concentrated on a few sites on the edges of ancient settlement. Much of the
material surviving from the Ptolemaic and Roman periods was produced and found in vil­
lages, while nearly all of the papyri from the Byzantine period come from cities, mainly
the metropoleis of the nomes.

Equally uneven is the spreading of preserved papyri over the centuries.1 The number of
Greek texts fluctuates from the beginning of papyrological documentation (about 300

Page 1 of 39
The Range of Documentary Texts: Types and Categories

BCE) until Greek was replaced first by Coptic (after 700 CE) and later on by Arabic as the
main language of Egypt (after 800 CE). The comparatively small number of papyri, wood­
en panels, and ostraca preserved from other parts of the Graeco-Roman world are
methodologically highly important because they (p. 359) allow us to compare and some­
times to correct our views based on the evidence from Egypt.2 Unfortunately, because
they are few and come from a handful of places, papyri from outside Egypt offer hardly
more than some selective spotlights (cf. chapter 20).

Besides the physical and climatic conditions of preservation, the preferences of papyrolo­
gists have a strong influence on the available material. The present state of editions dif­
fers greatly among linguistic groups. Compared to the roughly 650 volumes of Greek pa­
pyri published so far, other languages are poorly represented with barely 35 volumes with
Coptic documents, although extensive collections of Coptic texts remain unpublished. The
small circle of specialists in demotic and Arabic papyri have edited only about four thou­
sand fragments of demotic and around three thousand Arabic papyri.

Even with Greek papyri, editors tend to publish documents that offer numerous parallels,
and they have favored literary texts over documents, so that while the proportion of edit­
ed literary to documentary papyri is about 1:5, the actual proportion of preserved papyri
is probably closer to 1:50. Papyri from archives and dossiers (see chapter 10) are also
more likely to be edited than isolated texts.

Written Documents and Illiterate People

Even on an optimistic assessment, only a small percentage of the population of Graeco-

Roman Egypt was able to write and read, even among those who helped govern the vil­
lages. On the other hand, a considerable number of papyri and ostraca document efforts
to learn to write (see chapter 14). A large part of transactions and communications was
done orally and did not find any written expression. Even such far-reaching agreements
as marriages very often happened without any written contract. Whole areas of life were
never documented. For a document to be put into writing there must have been a need to
secure and memorialize information. As a consequence, written documentation in antique
cultures very often dealt with matters concerning power, wealth, or dependence. The ma­
jority of the papyrological evidence is thus connected with the justification and securing
of material rights or responsibilities between private people or between the state and the
individual.

The difficulty in giving a systematic overview of the documentary papyri is a product of

their extreme textual diversity. Compared to the Ptolemaic period, the spectrum of papyri
continually and considerably widened during the Roman period. It is hard to say whether
this tendency reflects a change in the common writing traditions or (probably) whether it
was the consequence of a more elaborated administration and changes in the legal sys­
tem.

Page 2 of 39
The Range of Documentary Texts: Types and Categories

Every classification of the documentary papyri becomes complex, as the model of Orsoli­
na Montevecchi, with thirteen main categories and dozens of minor categories, shows.3
However, this complexity may be anachronistic in trying to (p. 360) understand ancient
documents. Furthermore, the categories change depending on the choice of historic, ju­
ristical, or technical-formal features. Finally, different languages created different written
forms. The “papyrological habit” explained in the following pages primarily describes the
practices displayed in Greek documents, largely applicable to Coptic as well. Other cate­
gories might be useful for demotic or Arabic documents.4

Lost Documentation

It is impossible to determine the loss of private correspondence or accounts, but we may

estimate what was lost in the sector of official administration. Throughout the entire “pa­
pyrological millennium” Alexandria was the capital of Egypt. As residence of the Ptolema­
ic kings and seat of the highest officials, Alexandria accommodated not only the famous li­
brary but also the central archives of the court and headquarters of administration. Here
converged the accounts of the tax and finance administration, as well as the papers of le­
gal affairs and the records of diplomatic affairs. During the principate, Alexandria was
the seat of the governor and several financial procurators. In late antiquity it continued to
be the capital of the diocese. At all times it housed the archives of those governors, where
the papers of the individual officeholders were collected. Presumably since Claudius but
at the latest since the middle of the second century CE, the core of the archive consisted
of the official diaries (commentarii), which listed all of the activities of the prefects. Per­
haps also since the first century CE the extensive books containing copies of all of the
prefects' correspondence might have been stored there as well. Since the reign of Marcus
Aurelius or at the latest since the Severan emperors numerous petitions have also been
added to the archives. P.Yale I 61 tells us that, during a three-day period at a conventus of
the prefect Subatianus Aquila in 210 CE, more than 1804 petitions were handed in. This
may hint at the amount of the paperwork in the governor's headquarters. Furthermore,
the judicial decisions of the governors and other judges, notifications of birth from Roman
citizens, epikrisis records, and of course official correspondence with the central govern­
ment in Rome were collected as well.5

Almost all of this is lost, save for a few documents that returned to the chôra. If we had
just a small fraction of the contents of these archives, our knowledge of the Ptolemaic,
Roman, and early Byzantine administration, government, and diplomacy would have quite
a different base.

Similar archives were also kept in the offices of the local administration in the roughly
fifty districts (nomes), described in chapter 22.6 Not a single find has been made of a
nome archive that could give us a representative selection of all of the archival materials
collected there for centuries; only fragments remain. Occasionally old records were
thrown away or their blank reverse sides used as (p. 361) cheap paper for private use. To

Page 3 of 39
The Range of Documentary Texts: Types and Categories

such reuse we owe the survival of one example (P.Panop.Beatty 1–2) of the incoming and
outgoing correspondence of a strategos.

After the mêtropoleis received a city council during the reign of Septimius Severus, the
volume of public documentation increased still further, including the records of council
meetings. Once more, the surviving texts represent a very small fragment of the original
quantity.

Types of Documents
Conventionally the documentary papyri (and ostraca) are divided into private and public.
On the basis of historical and formal features, private documents can be classified into
private communications, records of private legal transactions, accounts, and finances,
and documents of piety and worship. The second group contains documents concerning
the interaction between the state and individuals and pronouncements of the government
and administration. Only the last group was meant for public dissemination. The formal
subdivisions should not be taken too strictly, as overlaps occur.

Private Documents

Private Communications
In a quite mobile society, written communication played an important role. Hundreds of
letters and reports survive, but their chronological distribution is unbalanced: About 550
letters from the Ptolemaic period have been edited, but half of these originate from the
Zenon archive. By contrast, about 1900 have been published from the first three cen­
turies of Roman rule. Whether there was less letter writing in the Ptolemaic period or the
hazards of survival are responsible for this situation, we do not know. The following cate­
gories can be classifed by contents:

Among the oldest Greek papyri, letters between private persons appear as a genre that
followed particular formal conventions. Every papyrological anthology contains a certain
number of private letters touching the modern reader through their immediacy and real­
ism.7 Remarkably, quite a few letters were written by women and even by children or
teenagers.8 Also in social terms, it seems as if a wide spectrum is represented. Although
most of the private correspondence comes as expected from the upper classes such as the
metropolitai or landowners and (later) in the better-educated parts of the Christian cler­
gy, it seems as if at least occasionally simple farmers and day laborers sent letters, too. It
often remains unclear (p. 362) whether the sender wrote the letters with his own hand or
dictated them to a professional scribe.

In most cases the letter reports that the sender is in good health. Sometimes letters
speak of small daily concerns, for instance, requests to send particular items. Some, how­
ever, were written in extraordinary situations like some letters from recruits in the Ro­
man fleet in Misenum reporting home to their families9 or communications about the
death of a relative (P.Fouad 75). The exterior design of private letters usually exhibits
Page 4 of 39
The Range of Documentary Texts: Types and Categories

fixed patterns. From the third century BCE to the fourth century CE, private letters used
a stylized phrasing, either “X to Y, greetings” or “To X from Y, greetings.” Additionally, a
whole series of stereotyped phrases was used, in particular in the introduction and the fi­
nal greetings. Dozens of private letters start with the phrase “Above all I wish you to be
healthy” and finish with the usual formula valetudinis: “I pray that you will be well for
many years.” Among the stereotyped elements we very often find greetings directed to
other members of the household and wishes that the children of the house will not be
touched by the evil eye.” At least the final formula of greetings should be written with the
sender's own hand if that person is literate. Because of the widespread clichés only a few
really emotional comments appear.10 Any search, for instance, for a love letter among the
papyri would be in vain. Reference to peculiar situations, events, and people make pri­
vate letters often difficult to understand for the present-day reader. From the fourth cen­
tury CE onward the habit developed of omitting the correspondent's name at the begin­
ning of the letter and instead writing the postal address on the reverse side.

A second category is business correspondence. The majority of these letters contain re­
ports to a superior or colleague or orders to a subordinate. Most of these are connected
with the management of large estates. The extensive correspondence of the Zenon and
Heroninos archives belongs largely to this category. As parts of more extensive archives
(which ideally also contain accounts and legal contracts), business letters can offer vari­
ous insights into forms of cultivation, structures of organization, labor systems, and eco­
nomic circumstances of estates. It is likely characteristic of the general economic and
productive structures in Egypt that records and correspondence regarding commercial
enterprises did not survive to a comparable extent.

Among the private letters stand out two groups that follow their own design and form and
come close to the literary genre of “open letters”: More than 100 letters of recommenda­
tion survive in Greek and a couple in Latin.11 They range from the third century BCE to
the sixth century CE; 55 of them come from the Ptolemaic period, and no fewer than 49 of
these from the Zenon archive. In contrast, only a dozen letters of condolence survive, dat­
ing from the first to the sixth or seventh century CE.12 It can hardly be a coincidence that
the consolatio established itself as a literary genre during just this period. Not a single ex­
ample survives from the Ptolemaic period. Surprisingly, perhaps, all of the condolence let­
ters express only (p. 363) moderate sorrow, sympathy, and encouragement. Their stereo­
typed style of writing also explains why there is no significant difference between the pa­
gan and the Christian letters of condolence.

Standardized formulas are also evident in the two dozen written invitations for dinner,13
which belong mainly to the second and third centuries CE. These are small in size (ca. 5
× 5 cm) and are formulated in short words following this pattern: name of the host—the
person invited—invitation to dinner—occasion of the invitation—place of the party—hour
and date. Because they were delivered personally, they do not mention any address.
Birthdays, weddings, epikriseis, or the awarding of honorary posts were occasions for
such dinners. From time to time they had a religious character, for instance, invitations to
the “klinê of Anoubis” (SB XX 14503) or Sarapis that was to take place in a private house

Page 5 of 39
The Range of Documentary Texts: Types and Categories

or a holy shrine. In the formulation of SB X10496 (third century CE), even the god himself
supposedly issues the dinner invitation. Similar invitations occur also for Christian feasts
(SB XVI 12980).

Finally, there is the large but heterogeneous group of private notes, lists, and
memoranda. Concerning their content, the spectrum reaches from short messages to the
listing of various ordinary, daily items (such as laundry lists) and records of income or ex­
penses (such as for a feast). Among these, lists of objects for dowries stand out. Very of­
ten only various items are listed, without a heading informing us of the purpose.

Private Legal Documents

A very large part of the texts written on papyrus serve to secure certain rights and
claims. At all times legal transactions were set down in writing to be protected from later
alterations. These documents of juridical relevance follow specific formulas and patterns,
the fundamentals of which are generally uniform, although they sometimes surprise us
with local (often nome-specific) particularities. Legal documents were usually composed
by professional scribes who had to know the correct terminology and the appropriate
clauses. Their work might have been made easier with sample texts like SB XX 15027.
Indispensable components of such legal documents were, in a strict sense, the fixing of
the business and its resulting obligation, the precise naming of the parties to the contract
(including the father's name, the origin of each party, and, during Roman times, also a
physical description), and the date and place of writing. Written contracts or agreements
had value as evidence in court, but for the legal force of a transaction, only the parties'
agreement was decisive. The establishment of a written document was not legally neces­
sary.

The evolution of the processes for drawing up contracts is discussed in detail in chapter
23. In general, we may suppose that copies remained in the hands of the contracting par­
ties, but in some procedures a copy was also placed in a trusted repository. In the Ptole­
maic period this was often the “contract keeper” (syngraphophylax), but another possibil­
ity existed, having contracts drawn up and (p. 364) registered by the professional notaries
(agoranomoi). From the third century BCE until at least the third century CE the state
notary's office offered contracting parties the option to register publicly their private le­
gal contracts. This was common and indeed compulsory after 146 BCE for demotic
records also (P.Par. 65).

Another form of official recording of private legal business developed from banking trans­
actions. While in Ptolemaic times such diagraphai were considered as additional evidence
for otherwise recorded business, beginning with Augustus (BGU IV 1184, 27 BCE), “inde­
pendent” diagraphai were valid “bank-notarial” documentation of the underlying legal
business as well. Bank diagraphai disappear before the end of the third century CE (the
latest independent diagraphê is SPP XX 74 from 276; the latest dependent one is M.Chr.
171, col. II, from 293).

Page 6 of 39
The Range of Documentary Texts: Types and Categories

The participation of state authorities and especially the entry of the contract into an offi­
cial register gave the agreement additional security and, from our point of view, created
another type of surviving documentary record. Such grapheion registers survive for the
Arsinoite villages Theogenis (CPR XIII, second century BCE), Kerkesoucha Orous, and
Tebtynis (P.Mich. II 121–128; P.Mich.V 238, first century CE). These registers were orga­
nized as the office's journal dʼentrée and recorded every contract in chronological order.
They usually give detailed descriptions of the type of business, the parties, the object of
contract, and the monetary value. One of them lists the 247 contracts (which came in
within four months) and summarizes their title and content: 136 homologiai (contracts
generally, sometimes specified as contract of sale or contract of dowry), fifty leases, twen­
ty-seven loans, seventeen employment contracts, and so on.

During the first century of Roman rule every nome had appropriate archives where pri­
vate and public papers were stored. In 72 CE at the latest (BGU 1184 = M. Chr. 202) the
archives were divided into a depository of public records (bibliothêkê dêmosiôn logôn)
and a depository of property records (bibliothêkê enktêseôn), where the records of own­
ership for land and real property (and slaves?) were kept. The central archives in Alexan­
dria —we do not know a lot about them—were further extended by Hadrian's time. The
Roman administration encouraged registration by declaring that nonnotarial contracts
would be accepted at the court only after an expensive “disclosure.” By the end of the
third century CE the official syngraphai (by state notaries), the bank diagraphai, and all
other types of legal documents except the cheirographon and the hypomnema
disappeared. In the next 150 years the state did not offer citizens any official authority to
certify and secure their private legal contracts. Only the emerging of private notary of­
fices at the beginning of the fifth century CE brought about a basic change. From now on
the tabellio, who was appointed by the state but acted as a trusted private person, com­
posed the documents.14 Such contracts note as usual the date and place of conclusion,
the parties of contract, the type of business, processing features, and the sum in ques­
tion. Beginning with the Severan emperors, additionally, the stipulation clause found its
way into Greek contracts. Below this agreement first the witnesses and then the tabellio
sign.

(p. 365) Legal Contracts

Specific types of contracts (e.g., marriages, sales, leases) required a set of specific regu­
lations. They remained similar over the whole papyrological timespan. But individual pur­
poses and local variations produce significant diversity in detail. Shifting conceptions of
the law reflect changes of mentality and of social and economic conditions. For this rea­
son each document should be studied in the context of the whole corpus of the parallel
texts.

Records of legal acts usually follow stereotyped forms. One must make a clear distinction
between the type of legal transaction and its written form in a specific style of record (see
the preceding section). Some styles of records were used mainly for specific types of
transactions, but there was no strict and universal correspondence. Every systematic
classification of the legal documents is in danger of anachronism, as modern legal points
Page 7 of 39
The Range of Documentary Texts: Types and Categories

of view differ from those of the ancients. For this reason we distingush only two large cat­
egories (although further juristic differentiation would be possible): legal transactions
that concern people and those concerning things.

Contracts Regarding Persons

In a society that consisted of strictly separated groups of different ethnic, legal, and fiscal
status, as Egypt was particularly down to the third century CE, the ascertainment and
proof of personal status was of central significance for every individual. The society of
Graeco-Roman Egypt was always divided between freemen and slaves. The legal (and fis­
cal) status of freeborn individuals was defined through the state. Although slaves formed
a small minority of the population (and terms for slaves in informal contexts are some­
times ambiguous), in legal contexts clarity was essential. Cases absolutely free of doubt
can be found in about 170 edited sales of a slave.15 Only 15 of them date from Ptolemaic
times, while most survive from the Roman period. Christianity did not stop either slavery
or the keeping of records on sales of slaves, the majority of whom were female. Often
they were unwanted children who had been exposed and then were picked up to be
raised as slaves. Some records of slave sales written outside Egypt show that a certain
importation of slaves was carried on at all times. Moreover, a few emancipations of slaves
(mostly authenticated in public by an agoranomos) are found throughout the papyrologi­
cal millennium.16 There were different legal types of emancipations such as those con­
nected with a specific legal act (P.Oxy. I 48–50) or with a donatio mortis causa (P.Stras. II
122) or issued by public proclamation through a herald (Jur.Pap. 7). Besides deeds of sale
and emancipations, there exist many other types of documents that occasionally illumi­
nate topics such as the escape of slaves, their scope of work activity, and their education
(C.Ptol.Sklav. 53–255).

For freeborn persons, their legal and fiscal status was defined by birth. Under certain cir­
cumstances there was a need for additional arrangements, laid down in a great variety of
documents. Underaged children were subordinate to the authority of the father, only
rarely to their mother's authority. Orphans had to be placed (p. 366) under the care of a
guardian. Requests for a guardian give us a good sense of the detailed prescribed steps
that had to be undertaken (e.g., P.Harr. I 68; M.Chr. 325). Appointments of guardians
(P.Ryl. II 120) and oaths on assuming guardianship (SB VI 9049) are the results of these
procedures. Comparable proceedings can be found in requests for tutors of women (kyrioi).17
According to Greek law, women were capable of owning property but did not enjoy any le­
gal capacity to act independently (in Egyptian law they did). Normally there was no need
for a record dealing with the assumption of a guardianship for a woman. But if no hus­
band or other male relative existed, an application for the appointment of a kyrios had to
be adressed to the strategos, exegetes, or prefect. Some applications ofthis kind survive
in the original, and many others are embedded as copies in other business documents.18
Under Roman law, women could act on the basis of the “right of three children” even
without a guardian, but for the confirmation of that right they had to send an application
to the prefect, too (e.g., P.Oxy. XII1467).

Page 8 of 39
The Range of Documentary Texts: Types and Categories

Occasionally other cases of acts concerning family affairs and status appear: Both Ptole­
maic and Roman papyri mention adoptions, but actual acts of adoption are not known be­
fore late antiquity.19 Single documents also give us an idea about exceptional measures,
such as the emancipation of a daughter from the patria potestas (ChLA XII 521) or dis­
ownment (apokêryxis) by a family (P.Cair.Masp. I 67097 verso D; III 67353; Jur.Pap. 11, all
sixth century CE).

A vast number of marriage contracts survive, although marriages without any written
record were fully valid and obviously common; some were later documented in writing
(e.g., SB VI 9264). More than 140 marriage contracts survive from the fourth century BCE
to the fourth century CE. In keeping with general patterns of preservation, half of these
date to the second century CE.20 The legal form and the design of the written record vary
greatly down to the Byzantine period, when only one type remained in use. Until the
fourth century CE a marriage was formed either by the act of ekdosis (“handing over” the
bride), recorded in a document called “a contract of living together,” or by the act of fur­
nishing a dowry. But other types can also be found. The oldest marriage contract (P.Eleph.
1, 310 BCE)—which is at the same time the oldest dated Greek papyrus overall—was con­
cluded in front of six witnesses who sealed the document. The bride and the bridegroom
both descended from different Greek poleis, and the bridegroom and the bride's father
are given the right to decide where the couple would live. This was a private agreement
which has neither been registered in any official way nor corresponded to the legal norms
of any state. A final clause establishes in particular that the contract is supposed to be
valid in every city. Intention was a decisive point for the choice of a certain type of record
and the legal framework. In 128 CE P.Yadin I 18 was drawn up in Maoza in the district of
Petra ( province of Arabia ) for a Jewish couple. It basically followed Greek legal tradi­
tions and was written in the Greek language because this made it possible to present the
record in front of a Roman court and secured the woman more property rights. Just a few
years earlier (122–125 CE), a (p. 367) contract for the second marriage of the widow Ba­
batha was drawn up at the same place but was in Aramaic and in the form of a Jewish ke­
tubba (P.Yadin II 10).21

Deeds of divorce were concluded individually or by mutual agreement and were also
shaped in various written forms according to local traditions. Second-century CE
contracts of divorce from the Arsinoite nome show extensive stereotyped formulas
(P.Corn. 52). In addition to the confirmation of separation, such contracts contain clauses
about the return of the dowry and permission to remarry.22 In an indirect way these acts
show us underlying conceptions of marriage and female rights, as well as general moral
values.

One matter that needed written form was the settlement of succession other than by in­
testacy. The following types of written arrangements often appear in papyri:23 Greek wills
(diathêkê) are formulated as one-sided statements intended to take effect in the event of
death and follow a stereotyped formula with few, mostly local, variations. Typically they
begin with these words: “(So and so) made the following dispositions in good mental
health.” Some special forms are joint declarations of husband and wife (e.g., M. Chr.

Page 9 of 39
The Range of Documentary Texts: Types and Categories

307), parental division inter liberos upon death (e.g., P.Mich. V 321, 322a), and agree­
ments concerning succession made between husband and wife in marriage contracts.

The meriteia (often called donatio mortis causa by scholars) may have developed from
Egyptian traditions. Nevertheless, it was also popular among the Greek elements of the
population. Apparently standard Greek wills were more common in metropoleis, meriteiai
in villages. We also have thirty Roman wills dating from the late first century to the early
fourth century CE. At first these were written in Latin, but beginning around the time of
Alexander Severus they appear also in Greek.24 Women use the same legal instruments as
men in this domain. Public recording of wills was common during the Ptolemaic period
and obligatory in the Roman. The opening of a testament had to follow certain rules (BGU
XIII 2244). It was not unusual to update a testament several times, but a written revoca­
tion was necessary to make previous wills ineVective.25

Contracts Concerning Property

Documents that deal with property, its acquisition, and various activities of related busi­
ness number in the thousands. In terms of quantity, this group represents the majority
not just in legal documents in a wider sense but also in documentary papyri in general.
These data are not biased: The wish to establish rights or obligations in writing was the
decisive reason for an extensive production of such texts. This section describes only
some of the basic types, each one of which could be subdivided into a series of subtypes
of acts arranged by content, form, and temporal (and often local) character.

It is hardly surprising that sales are one of the best represented transactions in papyri.
The seller transfers the property and the full right to a specific thing to the buyer and
confirms the receipt of the purchase price. The contracts of sale are formulated as
records of already settled transactions. During the Ptolemaic period the objective formula
“he sold” (seller)—“he bought” (buyer) was the most (p. 368) frequent, while later on the
subjective homology (“I acknowledge that I have sold and received the price”) was more
common. Objects of purchase are chiefly real estate, movable property, animals (mainly
donkeys), and slaves, who are occasionally described in detail. There had to be an official
registration (katagraphê) of the purchase of real estate and slaves. Many questions con­
cerning legal-historical or social and economic-historical areas (for instance, liability for
defects, status of business partners, development of prices) can be addressed to the
records of sales. A transfer of rights (similar to a sale) could be effectuated also by a deed
of gift, although this was less common (cf. P.Neph. 31, introduction). Furthermore, hous­
es, estates, and movable property, too, were objects of division contracts, which were
made mainly between relatives. Deeds of gift and divisions often contain clauses of revo­
cation. The preserved samples come mainly from Roman times.

Besides sales and gifts, other forms of assignment of rights existed. Although similar in
purpose, they differ strictly in terminology and legal concepts, as, for instance, does the
cession (parachoresis) of katoikic (“settler's”) land. Originally given by the Ptolemaic
kings to their soldiers on revocable tenure, katoikic land was still a special category of re­
al estate during the Roman period. Legally it was not seen as true private property, al­
Page 10 of 39
The Range of Documentary Texts: Types and Categories

though it was heritable and even transferable to women from the second century BCE on.
Such parcels of land (klêroi), like military accommodations, could not be sold, but it was
possible to transfer them for monetary “compensation.”

From Roman and increasingly from Byzantine times we have records of transactions that
straddle the boundary between sale and loan. Normally they deal with agricultural prod­
ucts (often wine) or products of craftsmanship. In a sale on delivery or loan of money with
repayment in kind the vendor receives his money immediately, although the product is to
be delivered at a set date in the future.26 Sometimes a precise statement of price or the
amount of the sold product is missing, which creates the impression that the sale on de­
livery hides a loan with excessive interest. The fact that the majority of the almost two
hundred actual known sales on delivery date from the sixth and seventh centuries CE and
involve a dependent rural population points in the same direction. But many times pre­
purchase of a product of the harvest might also account for these transactions. The sale
on credit is a similar transaction but is organized the other way around, as the product is
going to be paid for later but is delivered immediately after the sale. Nevertheless, the
purchase price is stated as “received,” and a contract of loan that deals with the relevant
sum was set up.27

In every century of the “papyrological millennium” the most numerous preserved con­
tracts are loans.28 The delivery of money or other valuables, as well as the fixing of inter­
est and (normally) a deadline for repayment, demanded a written security—and records
of loan were carefully preserved. Formally, loan contracts show both subjectively and ob­
jectively styled types, and their content can be separated into two basic groups: loans of
money and loans in kind. The latter often deal with seeds borrowed from a state-owned
granary, which normally had (p. 369) to be given back after the harvest with an additional
50 percent. Loans in money legally bore 24 percent interest during the Ptolemaic period
and 12 percent during the Roman. Numerous loans of money declare themselves as inter­
est-free loans, but they probably already included the interest in the capital sum. Ficti­
tious loans also appear, hiding other transactions. The duration of loans varied consider­
ably (from a few weeks to many years). In the Byzantine period the term of repayment of­
ten is not fixed but depends on the creditor's preference. This of course further weakened
the debtor's position. Loan contracts were usually made invalid after repayment by sim­
ply crossing them out, but repayment receipts were also common.

Contracts of deposit were another method of securing property rights. A custodian con­
firms that he has taken over certain objects (sometimes even people). He can use the ob­
jects but is obliged to give them back on the owner's demand. Some contracts of deposit
arouse the suspicion that other transactions are involved (e.g., a dowry for illegal mar­
riages of soldiers).29

Another extensive group is that of legal documents connected to gainful employment, like
lease, rent, work, and related business contracts. In antiquity, contracts of this kind were
uniformly called misthôsis in Greek. Contracts of lease dealing with agricultural estates
have survived by the hundreds, reflecting an important agricultural reality. Others involve

Page 11 of 39
The Range of Documentary Texts: Types and Categories

workshops, shops, and stores or parts of them.30 The contracts often describe in detail
the location and the condition of the piece of land, the work and tax obligations, the pay­
ment of the rent, and clauses of penalty. Some common characteristics are evident in the
various styles of lease contracts. During the Ptolemaic period, leases were almost always
designed as an objective statement: “(So and so) has leased to (so and so).” In the third
century BCE the six-witness act predominated, while in the second century contracts
were ratified by an agoranomos and later the objective homology (“acknowledges that he
has leased”). In the Roman period two different types developed. One is a contract form
that appeared either in narrative style (“has leased”) as an objective homologia or as a
private written cheirographon (“I have leased to you”), a form very common in the Arsi­
noite nome. The second type was expressed as an offer of lease (“I wish to lease”), even
though the deal had already been settled. In the Byzantine period the subjective homolo­
gy (“I acknowledge that I have leased”) prevailed, but one can observe a remarkably un­
even local distribution of the lease contracts, which might reflect different agrarian con­
ditions in various parts of the country. The records of leases concentrate heavily on the
Arsinoite, with many fewer appearing from the Oxyrhynchite and Hermopolite nomes.
The duration of leases was usually short (often until the next harvest, e.g., one year), but
a special form of hereditary leasehold, the emphyteusis, appeared in Byzantine times. The
emphyteusis created a long-term lease, often for a lifetime and even for several genera­
tions. In particular, churches and monasteries often handed their plots over on this
basis.31 In general, rental agreements follow the same patterns as leases. The only differ­
ence is that rental agreements deal with houses, rooms, or parts of them.32

Other contracts involving labor were also called misthôsis. Thus, contracts of em­
(p. 370)

ployment and work, apprenticeship, and transport, as well as the renting of animals, were
designed in a formal and terminological style similar to contracts of leases. A work con­
tract centers on the production of a certain object or performance of services, and pay­
ment of all or part of the wage is sometimes made in advance. Contracts of apprentice­
ship spell out the manner and skills of the job (most commonly weaving).33Nursing con­
tracts concern the nursing and care of infants, who were often rescued from exposure (C.
Pap.Gr. I). In some contracts a remarkable condition requires the nurse to procure a new
infant if one dies. Contracts of transport deal with the use of an animal or a ship and in­
clude detailed arrangements about loads, delivery dates, and destinations. Furthermore,
numerous shipping contracts are preserved, especially from Roman and Byzantine Egypt.
They contain the shipowner's acknowledgments of reception of cargo and arrangements
for transport, often for the shipping of tax grain.34 Most surviving contracts of transport
deal with trips within Egypt. However, the remarkable examples of a maritime loan for a
commercial journey to Somalia (SB III 7169, second century BCE) and another one to In­
dia (SB XVIII 13167, second century CE) show that comparable arrangements also exist­
ed for transport across the sea.

Eventually misthôseis were also linked to other legal transactions, especially loans. Most
common was the antichrêsis, a credit transaction in which capital and/ or interest were
not paid back but paid off through work or the use of an item. The latter mainly concerns

Page 12 of 39
The Range of Documentary Texts: Types and Categories

a house or an apartment (or a part of it), a garden, and so on that the creditor could use
in lieu of repayment. Such contracts do not necessarily reflect desperate debtors.35

Accounts
Another extensive group of private documents, described broadly as economic texts, in­
cludes accounts of all kinds, receipts, orders for payment, proofs of banking, lists of in­
coming or outgoing money or goods, transport lists, and much more. No strict “types of
records” developed; rather, a huge variety of texts exists. The content and form of these
documents were determined by individual habits and requirements.

As we expect in an agricultural economy, farm accounts are prominent in every century.

Usually these are lists that, with luck, have a heading giving their purpose and date. In
isolation they often seem unremarkable, but in groups and properly approached they can
inform us of the size and structure of both smaller and bigger estates, as well as the
methods of cultivation and management. An outstanding example is the Kellis agricultur­
al account book (P. Kell. IV 96). Unsurprisingly, such accounts originate in particular from
large estates (e.g. P.Oxy. XVI 1911 and LV 3804 from the Apions). Small farmers had nei­
ther the necessity nor the skill for doing such systematic accounts. Most of these ac­
counts are, in any case, mere lists of income and expenses that did not require more so­
phisticated analysis.

In the area of trade and handicraft production we find particularly short receipts
(p. 371)

and lists. Although the single documents—often written on ostraca—are again stereo­
typed, they offer very valuable historical evidence because of their volume. The following
examples illustrate the broad textual spectrum that is evident in lists, calculations, bills,
and so on: information about the affairs of private households, like a list of expenses for
food and fuel from the third century BCE (UPZII 158 A, cols. 1–2) or the monthly calcula­
tions by a cook in 185 or 215 CE (P.Oxy. I 108, cols. 1–2). Several inventories of books are
of cultural-historical significance (C.Liste Libri). A list of clothing belonging to Zenon
gives us an interesting insight into the wardrobe of a wealthy man (P.Cair.Zen.I 59092),
especially if compared to the list of clothes that the scholasticus Theophanes took with
him on his journey to Antioch nearly six hundred years later (P.Ryl.W 627, cols. 1–2).
Theophanes also left highly informative notes about the stops and expenses incurred on
his journey (Matthews 2006). An inventory of church property (Sel.Pap. I 192) from the
fifth or sixth century offers welcome information on the equipment used in early Christian
worship.

Small daily purchases were paid for in cash without any written record. Especially in Ro­
man times, however, it was also common to pay larger sums through written banker's or­
ders. Clearing and transfer involved both money and grain. The banker transferred sums
of money from one account to another; grain was transferred by the sitologoi, tax officials
who were also responsible for the state granaries, where private grain accounts were
kept as well. Tax payments in grain or payments to other private people could be trans­
ferred with written (and sealed) orders.

Page 13 of 39
The Range of Documentary Texts: Types and Categories

Belief and Superstition

Documentary papyri join the literary and semiliterary sources as important evidence of
religion and cult. The diversity of the textual evidence in this area does not easily fit into
a system.36 A series of Ptolemaic and Roman records directly concern temples, pagan
priesthood, and cult, such as a list of Egyptian temples in Kerkeosiris, which was written
by a village scribe in 115/114 BCE (P.Tebt. I 88); applications for the purchase of priestly
offices (e.g., P.Tebt. II 294 and 296); prohibitions against disturbing divine services; texts
for the sealing of sacriicial animals (BGU I 356), and much more. Tellingly, most Greek
texts concern the contact of priests with state authorities, while matters of cult and belief
or the temples' internal affairs are nearly absent in the Greek papyrological documenta­
tion; demotic texts ill part of that gap.

On the other hand, documents that reflect private piety, especially those connected with
cults surrounding death, are numerous. Mummification of various grades was practiced
until Christian late antiquity, and several documents deal with sending mummies to the
family grave (Sel. Pap. I 104; C.Pap.Hengstl 59); many receipts record the transport of
mummies, often by ship (P.Hamb. I 74). So called mummy labels were used to identify the
bodies; these labels are small wooden (p. 372) tablets with the name and age of the de­
ceased, as well as occasionally destination and notes on charges paid. Thousands of such
mummy tablets survive and provide information especially on onomastics and demogra­
phy.37 A remarkable tabulation of funeral expenses appears in SPP XXII 56.

Papyri are also welcome testimonies of popular piety and widespread superstition. Be­
sides magical texts and prayers (which belong to the semiliterary area), extensive magi­
cal textbooks (like PGM 11–15), instructions in religious methods (PGM II 13.234–244),
and numerous amulets (e.g., P.Princ. III 159, against fever) are found. The biggest homo­
geneous group in this domain are oracle questions. More than seventy examples, chrono­
logically spread over all centuries, have been published, and many more remain
unedited.38 Oracles occasionally have a political connection, like the oracle of Hermes
Trismegistos, which dealt with the suppression of a revolt in the Thebaid under Ptolemy
VI (O. Hor, pp. 1–6). From time to time oracles were also used in the search for justice.39
But mostly they concern private matters, chiefly reflecting distress or difficult decisions.
Many such documents, especially of the Wrst two centuries CE, are preserved from
Oxyrhynchus and Soknopaiou Nesos, where the local god, Soknopaios, was worshiped
(SB XVIII 14043–14050). The question concerning a marriage in W.Chr. 122 is an instruc­
tive, maybe even typical example. Another prominent group includes some 140 horo­
scopes, most of them dating from the second and third centuries CE. Christianity did not
end people's faith in magical methods; both amulets and oracular questions continue
down to the seventh century.40

Public Life

Page 14 of 39
The Range of Documentary Texts: Types and Categories

Political History
Documents with a direct reference to political events are relatively rare. No original
records of diplomatic correspondence or historical events survive. Nevertheless, some
types of papyri deal with historical situations and sometimes also with events of impact
on world history. All of them are purely accidental discoveries.

An official war report about an operation by the Ptolemaic fleet along the Cilician coast in
246 BCE (W.Chr. 1) provides a concrete episode from the Third Syrian War, between
Ptolemy III and the Seleucid empire, which is otherwise known only from anecdotal evi­
dence. A multilingual dossier highlights the Jewish-Syrian-Egyptian conflict, also called
the “war of scepters,” which lasted from 103 to 101 BCE (C.Jud. Syr.Eg.). A private letter
(P.Köln IV 186) refers to combat during (probably) the Sixth Syrian War, and a whole se­
ries of letters and official documents also shows the dynastic and regional conflicts within
Egypt during the rule of the later Ptolemies (e.g., W.Chr. 9–12). A piece of official corre­
spondence between subaltern officials shows arrangements for the visit of a Roman sena­
tor to Egyptian sanctuaries with holy crocodiles in 112 BCE, long before Egypt was actu­
ally annexed to the Roman Empire (P.Tebt. I 33). An epistula of Claudius to the Alexandri­
ans from November 41 (p. 373) CE shows the unstable situation in the capital of Egypt
(C.Pap.Jud. II 153). At the same time, this famous letter offers remarkable evidence for
the style of Roman government. Hadrian's accession to the throne is announced in P.Oxy.
LV 3781, dated August 117, and P.Giss.13 contains an invitation for the related celebra­
tion. Such documents are preserved because they were copied and sent to the nomes as
official announcements. In one instance, CPR XXIII 24, which is an official announcement
of news about imperial victories, sheds light on these procedures.

Among the “historical papyri” only the certificates of pagan sacrifices from the period of
the Decian persecution of Christians form a salient type of document. Decius enacted an
edict in September 249 that obligated all citizens of the empire to make sacrifice for the
gods. Forty-five original certificates from the year 250/251 CE have been published so
far,41 half from a single findspot (Theadelphia). As becomes obvious from their stereo­
typed formula, these certificates were indeed styled as requests (libelli) to a local com­
mission for confirmation of a food and drink sacrifice. One (W.Chr. 125) was issued for a
female priest serving the cult of the crocodile god Petesouchos; not even pagan priests
were excluded from the obligation. Apart from this extraordinary group the papyri tell us
next to nothing about universal measures like Diocletian's persecution of Christians, for
which one might expect massive effects, including written documentation in Egypt. A
sample of what once existed comes from P.Oxy. XXXIII 2673 (304 CE), in which a lector
makes a sworn statement that his church does not own any property.

More concrete are administrative regulations. The reforms of the Tetrarchs, for example,
are elucidated by several important sources. The edict of the prefect Aristius Optatus
from March 297 (P.Cair.Isid. 1) enforces a collateral measure for the introduction of a new
tax system. In addition, CPR XXIII 20 (298) preserves a fragmentary copy of Diocletian's
edict ordering a general census, and P.Panop.Beatty 1 and 2 (298 and 300) provide unique
testimony for Diocletian's campaign to the southern border of Egypt. On various occa­
Page 15 of 39
The Range of Documentary Texts: Types and Categories

sions historical events of far-reaching consequences are reflected in papyri amazingly

quickly, at least in an indirect way. For example, the profound change in the recruiting
system that occurred in 376 CE is visible in the papyri of the following year.42 The disas­
trous plague of 542, which killed about a third of the population of Constantinople, has
been credited with a 20 percent decline in tax rates in the Upper Egyptian village of
Aphrodito in the year 544 ce.43

Pronouncements of Government and Administration

Records of government administration and public life may nearly match private legal doc­
uments in their bulk. But those records and their types depend much more on the politi­
cal arrangements, the structures of administration, and the systems of taxation than con­
tracts do. All administrative documents were supposed to be stored in official archives.

Legislative, judicial, and administrative announcements issued by rulers them­

(p. 374)

selves survive in significant quantity. They express the intention of the government and
therefore are excellent historical sources, even if many of them deal with matters of sub­
ordinate importance. More than one hundred such regulations, prostagmata, by the Ptole­
maic kings are known (C.Ord.Ptol.), and a similarly impressive number of various orders
(epistles, edicts, rescripts, etc.) of the Roman emperors from Augustus to Gallienus are
preserved.44 Among the latter, the decrees of Caracalla (P.Giss. I 40) and the apokrimata
(legal decisions) of Septimius Severus (P. Col. VI 123) are especially important. In late an­
tiquity there are fewer imperial orders, although some extensive texts exist, for instance,
Justinian's rescript of the year 551 in P.Cair.Masp. I 67024. Also remarkable is a fragmen­
tary papyrus that carries some lines of Justinian's edict XIII, promulgated in 539 CE
(P.Oxy. LXIII 4400 and addendum). These imperial decrees are, of course, not original
documents but copies, whether public or private.

Edicts and orders (epistalmata) of high officials, in particular of the prefects, appear in
fair numbers.45 They contain mainly administrative regulations (such as Mettius Rufus's
edict on the archives, M.Chr. 192) or judicial decisions. Such sentences (in full or in ab­
stracts) were also quoted as a legal basis in later petitions. For instance, P.Vind.Tandem 1,
a copy of a letter by Ptolemy II (285–246 BCE) written in the middle of the third century
CE, demonstrates how long some legal decisions might be remembered and cited.

Compilations of legal instructions, made for official use, were probably supposed to pro­
vide a quick orientation to laws or to serve as a guideline for decisions. The interval be­
tween the original and the copy could again be considerable: The so-called (Egyptian) law
book of Hermopolis, the demotic original of which goes back to the third century BCE,
was still copied in a Greek translation in the second half of the second century CE (P.Oxy.
XLVI 3285). A passage of the tax law SB XVIII 13315 of 89 CE relates to a decree of
Tiberius. The so-called Gnomon of the Idios Logos is the famous collection of legal canons
and regulations for that procurator. In its opening the author (probably an earlier holder
of that office) indicates his intention to catalogue all relevant instructions since Augustus.
An extensive copy of the Gnomon in BGU V 1210 dates after 149 CE, but another, frag­

Page 16 of 39
The Range of Documentary Texts: Types and Categories

mentary copy in P.Oxy. XLII 3014 was written in the first century and therefore proves the
older age of the Gnomon.

Minutes and Proceedings

Apparently all civil service departments were told to keep official diaries in which every
action was written down.46 Extracts from these hypomnêmatismoi are frequently men­
tioned. Many original examples also survive, such as W.Chr. 41, a journal of a strategos of
the Ombite nome. Written by secretaries, they were published by public notice and later
put into archives.

Similarly, officials had to compile copybooks containing duplicates of their corre­

(p. 375)

spondence. Even the smallest departments generated paperwork, as is demonstrated by

two dozen journals written on potsherds at the small outpost called Krokodilô in the mid­
dle of the Eastern Desert (O. Krok. 1–5, 7, 24–40). An outstanding example is
P.Panop.Beatty 1, containing the outgoing letters of the Panopolite strategos, which are
virtually complete for September 298 CE. The text is of particular interest because at this
precise time preparations were being made for the arrival of Diocletian and his army.
There are sometimes as many as seventeen outgoing letters a day, produced by six writ­
ers, concerning correspondence with the president of the local town council, with the
head of the provincial financial administration, with strategoi of neighboring districts,
with military persons, and so on. Its counterpart, P.Panop.Beatty 2, lists the incoming offi­
cial correspondence of January and February 300. The largest number of letters was sent
by the procurator of the Lower Thebaid (fifty letters in two months), and often his com­
munications arrived on the same day in Panopolis, although the headquarters of the
procurator was about 120 miles away. Account books of the postal station at Takona,
north of Oxyrhynchus, give insight into the logistical efforts involved (P.Oxy. LX 4087 and
4088). These stations compiled records of the number and ranks of traveling officials and
messengers, as well as of the number of horses and the expenses for their food. The
postal service of the Ptolemaic kings was also well organized, as a journal of an interme­
diate posting station shows (Sel. Pap. II 397).

After Septimius Severus granted councils to Alexandria and the metropoleis, their admin­
istrative duties (and financial accountability) for the nomes continuously increased. Each
boulê kept the minutes of its meetings. Snatches of these detailed acts survive from
Oxyrhynchus and Hermopolis.47 Not only resolutions but also discussions, motions, com­
ments, and acclamations were recorded verbatim and therefore present an insight into
the routine meetings and business of the city councils, which consisted mainly of the in­
stallation of (liturgical) officials and financial administration. Proceedings of this kind
were widespread, for P.Yadin I 12 contains extracts from council minutes of Petra as early
as 124 CE.

Official Correspondence
Many papyri contain correspondence between officials. Higher authorities transmit or­
ders and instructions to lower levels, which had to report in turn to their superiors. Some­
times very detailed information about specific official issues was communicated, such as
Page 17 of 39
The Range of Documentary Texts: Types and Categories

that in the carbonized papyrus P.Bub. II 5, in which the eklogistes of the Boubastite nome
in the Nile delta posed questions to his strategoi concerning financial matters (tax
debts?). Another type of correspondence dealt with official inquiries, such as P.Oxy.
XXXIII 2665:Two bibliophylakes and bouleutai from Oxyrhynchus report that a con­
demned person whose property is supposed to (p. 376) be confiscated actually has no
property in their district. The official correspondence of the village clerk Petaus provides
a good impression of the content of such exchanges of letters (P.Petaus 10–25). The extent
of paperwork seems to have been huge in view of the fact that written communications
were required even between officials who worked in the same metropolis—maybe even
next door. Formalism made it necessary to file every single official act. An instructive in­
stance came to light in the correspondence of the basilikos grammateus Hephaistion
(alias Ammonios), who in the year 194 ce also administered the vacant post of strategos
of his district. Hephaistion, acting as vice strategos, sent to himself in his position as basi­
likos grammateus letters and instructions, correctly using the polite phrases and other el­
ements of official correspondence (SB XVIII 13175).

Even to illiterate people a document from an official chancery was recognizable because
of its layout and the use of a special, stylized handwriting distinct from the usual business
script. Only a few complete originals from chancelleries of the Ptolemaic kings or Roman
governors have survived. We have an imposing order with the personal handwriting of
Ptolemy X Alexander I (UPZI 106), perhaps one signed by Cleopatra VII (P.Bingen 45), one
of the prefect Subatianus Aquila (SB I 4639), and a letter of the Emperor Theodosius II
(ChLA XLVI 1392). Most instructive is a comparison of the order of Aquila in the year 209
with a letter of Fl. Iulius Ausonius, praeses Augustamnicae, in the year 342.48 The letters
show remarkable palaeographic and formal similarities despite their considerable chrono­
logical distance and their composition in different offices: the design of the greeting, the
position of the date and place of issue on the left margin, as well as the dating by consuls
at the bottom of the document. An additional feature for the authenticity of official docu­
ments was the seal, which, however, is usually lost today. “I have sealed” became the
standard expression of approval under official and private letters of business, accounts,
and tax receipts of all kinds.

The Administration of Justice

Relatively few normative sources for law have survived, apart from enactments of law by
the kings, emperors, and prefects. For instance, the dikaiômata of Alexandria (P.Hal. 1, af­
ter 259 BCE) provide information on the laws of the Greek poleis. The coexistence and
blending of different legal systems with their separate juridical institutions becomes man­
ifest in a vast quantity of documents that were produced in the administration of
justice.49 During the Ptolemaic period summonses (like P.Hib. I 30), official reports of
summons (BGU VIII 1773–1778), statements of witness (P.Heid. VIII 413–416 with a list of
comparable documents on p. 45 f.), and court proceedings (P.Petr. III 22a–g) inform us of
the procedures followed in courts of law. In particular, the reports of proceedings stand
out as a specific category of text: About thirty examples are extant from the Ptolemaic pe­
riod, and more than two hundred from the Roman period until about 300 CE.50 (p. 377)

Page 18 of 39
The Range of Documentary Texts: Types and Categories

Stenographers recorded trials verbatim in shorthand. Afterward they made a clean copy
and filed it after the head of the court had checked it. Because copies of official proceed­
ings were often used as evidence in court, the archived files were accessible to the pub­
lic. Until Diocletian's time, reports of proceedings were always written in Greek; after­
ward they sometimes had a framework of introductory formulas in Latin. More than fifty
such bilingual reports of proceedings from late antiquity are known.51 Whatever their ori­
gin, they show almost identical layout and the deliberate utilization of different sizes and
characters of handwriting, even in the military courts.52

Often connected to the preparations for trials are the so-called orders to arrest. Standard­
ized in formulation, these documents sometimes concern official inquiries after one or
more people for an institution or a court; sometimes they are practically arrest warrants.
The nearly twenty examples from the Ptolemaic period were produced by various officials
(P.Paramone 9, introd., 104–106); the more than forty orders to arrest from the Roman pe­
riod show a much stronger standardization in wording and the authorities involved. Until
nearly the mid-third century, when substantial administrative reforms took place, almost
all of the orders are produced by strategoi and address the archephodoi of a village. The
approximately twenty orders from the next four centuries are mostly sent by centurions,
beneficiarii or praepositi pagi. The recipients are village officials such as eirenarchai or
komarchai.53

Interaction Between State and Individual

Another extensive category of official documents comprises texts arising from contact be­
tween the authorities and the general population. All incoming declarations, petitions,
and so on (often in multiple copies) went into the state archives. Each record had a stipu­
lated format, and those received were pasted together into a roll and numbered. Any item
could thus be identified by box, roll, and column. About 330 CE these tomoi synkollêsimoi
disappear, a fact that indicates profound changes in the organization of the public
offices.54

Petitions to Authorities
It may be significant that petitions to officials are the commonest type of record except
tax receipts. More than a thousand survive from the entire “papyrological millennium.”
The petitions sought redress for abuses or help against injustice. Often they inaugurated
a lawsuit.

The enteuxis is the classical Ptolemaic form of petition. In the third century BCE the en­
teuxeis addressed the king directly (e.g., P.Enteux.; M.Chr. 8–16). The structure of the pe­
titions is constant: After the opening, in the usual letter form—“to king X … from NN (+
title)”—followed a description of the injustice suffered and the request for concrete mea­
sures to correct it. The entry ended with the formulaic greeting eutuchei (“farewell”),
which was used only for the king. (p. 378) Administrative orders (BGU III 1011, II 5ff.) ex­
plicitly urged the petitioners to make the petitions as short as possible. Nevertheless, it is
hard to imagine that all enteuxeis were handed over personally to the king. As a copy
went to the strategos, petitions likely went through several levels before the king—in the
Page 19 of 39
The Range of Documentary Texts: Types and Categories

best case—agreed and added written instructions. The petitioner himself had to ensure
that the instructions were carried out. Petitions to officials (strategoi, epistrategoi), called
hypomnêmata, differed from the enteuxeis more in linguistic details than anything else.
From the second century BCE on, the differences diminish.

Besides petitions, people handed in notifications (prosangelmata) to civil and military offi­
cials and tried to obtain protection without a judicial decision. People did not hope for a
judgment with formal legal force but for the realization of justice through an official au­
thority. This “bureaucratic justice” also appears on the local level, as shown by the appli­
cations to the garrison commander Dioskourides in the years 154–145 BCE
(P.Phrour.Diosk. 1–12).

During the Roman period petitions were addressed to all levels of the provincial adminis­
tration, from the local police officer through the exegetes at the urban level and the strat­
egos, as well as the basilikos grammateus at the district level, to the procurators of
equestrian rank (e.g., epistrategoi, iuridicus) and the prefect himself. At the annual con­
ventus there was a possibility of reaching the prefect also outside Alexandria. All judg­
ments were accessible to the public because they were required to be posted in Alexan­
dria and the relevant metropolis, which also made it possible to obtain private copies.
This practice is best shown in the example of the Apokrimata (P.Col.VI 123), a copy of the
judicial decisions of Septimius Severus concerning eleven petitions. During the Roman
period a separate group of petitions was addressed to centurions, decurions, and benefi­
ciarii. So far more than fifty texts of this kind have been published.55 The military appears
in similar roles in epigraphical and papyrological evidence for other parts of the empire,
especially the beneficiarii.

After Diocletian's reform, petitions had to be addressed to the governors of the sub­
provinces, or duces. In addition, from the fifth century on, local dignitaries (urban author­
ities, officiales of the governor's staff, owners of large estates) increasingly appear as re­
cipients. During every period petitions followed stereotyped formulas.56 They give exten­
sive insight into legal transactions and conceptions of law because of their detailed narra­
tions of disputes and the requests for concrete measures of relief.

Almost thirty papyri dealing with the medical inspection of sick or injured persons are
known. These records date from the late first to the fourth century CE. The doctors who
wrote such reports were called “official doctor” from the late second century on and were
commissioned by the municipality (P.Oxy. LVIII 3926). A private person who wanted an of­
ficial medical inspection had to apply to the strategos or (later) to the ekdikos or nyk­
tostrategos (e.g., P.Oxy. LI 3620). (p. 379) Reports on the examination of corpses show
that in Roman Egypt an autopsy was required in cases of violent death (P.Oxy. I 51;
P.Rein. II 92).

Routine Declarations to Authorities

The Roman administration of Egypt, with its strict hierarchy and precisely prescribed offi­
cial procedures, developed a series of very specific documents. The tightening and stan­

Page 20 of 39
The Range of Documentary Texts: Types and Categories

dardization of administrative channels, the compulsory public service that was estab­
lished and expanded from the mid-first century CE on, and the efforts to control the popu­
lation, personal status, financial circumstances, and place of residence at the same time
enforced the formalization of official paperwork. Many specific types of documents have
in common the fact that they developed during the first century and disappeared in the
second half of the third century. Uniformity in wording, layout, and (often) size is remark­
able, as most of these documents were drafted by private individuals and handed in to of­
ficials on certain occasions. Their skilled handwriting reveals that documents that were
addressed to the authorities were normally written by professional scribes, who were also
trained in the layout and wording of declarations and contracts.

Of essential significance for all inhabitants of Graeco-Roman Egypt (and the whole em­
pire) was their legal and fiscal status. Especially for the privileged classes it must have
been a major concern that their status be officially documented and thus easily trans­
ferred to the next generation. Taxes and services were distributed in different ways
among the various, more legally than ethnically defined groups—Romans, Alexandrians,
Hellenes, Aigyptioi. The worries about status and an appropriate registration in the files
of the fiscal authorities lie behind many declarations and applications. It may have been
helpful for preservation that such documents were carefully treated and handed down
over the generations—like present personal documents. A Latin certificate of birth (testa­
tio) that was drawn up for the soldier M. Lucretius Clemens in 127 on behalf of his son
Serenus was kept by his descendants for at least a century (P.Diog. 1).

Notifications of Birth

Nearly all of the approximately forty notifications of birth come from the privileged sec­
tors of the population, the metropolitan citizens of Antinoopolis and Roman citizens.57 The
earliest examples are from the period of the Julian-Claudian emperors, the latest from the
beginning of the fourth century; more than half were found in Oxyrhynchus. In most cas­
es the father, sometimes together with the mother (but in the Oxyrhynchite also the house
owner), announced to the village or metropolis secretary in the form of a hypomnema the
birth of a son or (occasionally) a daughter. Often two, three, or more years passed be­
tween birth and declaration, but this had no significance because tax liability started at
fourteen. The small number of birth notifications indicates that they can hardly have been
obligatory. Things were different only in Antinoopolis, where the (p. 380) announcement
had to be made within thirty days in order to gain special privileges. Moreover, declara­
tion within this period was obligatory for the legitimate children of Roman citizens and
had to be made in Alexandria, where it was registered and posted in public, while the fa­
ther received a certificate on a wax tablet. So far thirteen such declarations are known,
dating from 62 (CPLat. 148) to 242 CE (CPLat. 163). For illegitimate offspring (e.g., chil­
dren of soldiers before Septimius Severus), instead of such a declaration, a private-law
declaration had to be made in front of seven witnesses and recorded on a wax tablet.

Epikrisis

Page 21 of 39
The Range of Documentary Texts: Types and Categories

As another bureaucratic procedure to define legal status, the epikrisis of boys who had
completed their thirteenth year produced extensive papyrological documentation. This
check of personal and iscal status determined who would enjoy Roman or Alexandrian cit­
izenship and was therefore exempted from the poll tax. It also clarified who belonged to
the privileged population of the metropoleis and had to pay only half the rate, as well as
who was Egyptian and had to pay the full amount.58 For the epikrisis of the native popula­
tion the district authorities (strategoi and basilikoi grammateis) were responsible; for the
epikrisis of the local Roman citizens and their households (as well as, up to the middle of
the second century, of Alexandrians), the prefects.59 In general, the status of an individual
followed that of the parents, but a certain degree of social mobility was possible, for in­
stance, with the attainment of Roman citizenship through military service. Status was
verified by the presentation of older epikrisis documents or census declarations. It is not
uncommon at the beginning of the third century to find among the documents presented
epikrisis papers from ancestors written half a century earlier. Various types of records in­
form us of these developments: epikrisis declarations (e.g., P.Oxy. XII 1452), epikrisis
registers, and excerpts from epikrisis files (e.g., Doc.Eser.Rom. 93).

Hellenes and Priests

Certain social groups with special status had their own required status documents and
processes for scrutiny. Among the Hellenes, the more exclusive class of “those from the
gymnasium” stood out, especially in Oxyrhynchus. They underwent additional selective
training and examination. For instance, P.Mich. XIV 676, from 272 CE, shows a particular­
ly remarkable application for epikrisis for the gymnasial class. Here the presented geneal­
ogy goes back six generations on the paternal side and eight on the maternal. Also men­
tioned is a maternal ancestor who had passed the (gymnasial) epikrisis under Nero—more
than two hundred years earlier. Similar arguments occur in P.Oxy. XLVI 3279.

A second exclusive group comprised the Egyptian priestly families. If someone wanted to
obtain a priesthood, he was required to be circumcised, in addition to meeting certain ge­
nealogical and physical conditions. Corresponding applications for circumcision had to be
made to the strategos, at least from the middle of the (p. 381) second century CE on; this
may indicate the Roman government's increased control of temples. The disappearance of
the applications for circumcision (latest example: PSIV 454 from the year 320 CE) marks
the decline of the Egyptian cults.60

Census Declarations

Registration of the inhabitants and their property was the most important precondition
for a counting of both the fiscal objects and the fiscal subjects. Under the Ptolemies this
took place through annual self-declaration; in Roman times through the “house-to-house-
declaration,” adapting the model of the census of Roman citizens. Every head of a house­
hold, who was typically also the home owner, male or female, had to submit a written cen­
sus declaration to the district authorities every fourteen years. An enormous bureaucratic
effort from both the officials and the general population was required because all house­
hold heads had to present themselves to the authorities of their home area. Orders to
Page 22 of 39
The Range of Documentary Texts: Types and Categories

conduct a census generally came from the prefect and were obligatory for all peregrine
inhabitants. Augustus introduced a seven-year cycle in 11/10 BCE, but in 19/20 CE this
was changed to a fourteen-year cycle, still sufficient for collecting the data needed for the
poll tax. There must have been millions of declarations, as the census in this form was
carried out until 257/258 CE. More than three hundred edited census declarations show
not only how consistently documentary patterns were maintained for decades but also
how local peculiarities of formula developed.61

Before the registration began, the administration routinely ordered people to return to
their registered homes. The approximately one dozen copies of so-called reintegration
edicts (or references to such) that survive threatened to impose sanctions against those
who disobeyed and remained in the cities, particularly Alexandria.62 The unstated objec­
tive is the recovery of tax fugitives. Some reintegration edicts seem to have been precipi­
tated by urban disturbances. For instance, P.Giss. I 40, col. 2, is probably the edict men­
tioned by Cassius Dio 77, 23, 2 in connection with the massacre that Caracalla carried
out in Alexandria in 215 CE.

Notifications of Death

The almost one hundred notifications of death of a male relative edited so far also had a
fiscal purpose.63 Formally, these so-called notifications are applications made by relatives
to the local officials of the metropolis or the villages to list the deceased person in the
register of the departed and delete him from the roll of taxpayers. The declarant received
a duplicate with the official subscriptions as evidence, while the original was—often in
form of a tomos synkollêsimos—put into the archive. Notifications of death were imposed
under Augustus (C.Pap.Gr. II 1, 2/3CE) in connection with the poll tax. At the beginning of
the fourth century this type of record disappears. Surviving examples show again the
haphazard distribution of papyrological evidence: Although notifications of death must
have been required everywhere, about sixty exemplars come from the Arsinoite and twen­
ty-six from the Oxyrhynchite nome, while from other nomes only two or three have sur­
vived.

(p. 382) Services for the State

Egypt 's punctual and complete payment of taxes was vital to every government. Conse­
quently, papyri show a highly elaborate system of fiscal administration but provide much
more information about the local levels than about the higher, central offices.64 For every
period the papyrological evidence offers rich information about numerous officials con­
cerned with assessing and collecting the taxes payable in kind or in money, the settle­
ment of accounts, control and forwarding of taxes, recovery of outstanding debts, and
much more. With many thousand relevant texts, the fiscal administration is as central in
papyrological sources as it was in the people's lives.

Taxes

Page 23 of 39
The Range of Documentary Texts: Types and Categories

Most significant were the duties on land, which generally had to be paid in wheat. Under
both Ptolemaic and Roman rule they were largely calculated on the basis of the yield of
the land. The tax wheat was directly transferred from the threshing floor to the state gra­
nary. Here the sitologoi recorded the entry and acknowledged receipt;65 moreover, they
reported to the strategos both daily and monthly (P.Vind.Worp 4, introd., pp. 39–42; P.Dub.
4, introd., p. 20). Under the Ptolemies, vineyards and gardens were also burdened by a
special tax payable in money. Other taxes were levied on animals, slaves, and various oc­
cupations. State-controlled banks received the taxes for the government's account.66

Innumerable tax receipts document the poll taxes, which under Ptolemaic rule were
called syntaxis and salt tax and had to be paid by male Egyptians. The Romans imposed
the laographia on all male inhabitants between fourteen and sixty-five years of age, ex­
cept Romans and citizens of the Greek poleis. From the late first century CE on, tax col­
lection was carried out mainly by liturgical officials, with tax farming used only for indi­
rect taxes. In Diocletian's fiscal reform, Egypt was subjected to the general system of cap­
itatio-iugatio, which imposed a unified tax assessment that embraced both land and peo­
ple. In actuality, however, the taxes were collected mostly on the basis of landholdings,
still largely in grain but also partly in cash. Again, numerous tax receipts document pay­
ments for the annona militaris, intended to feed the army and the civil servants, as well as
for the grain collected for Rome and later Constantinople (embolê). According to
Justinian's edict XIII 8 (539 CE), eight million artabas (approximately 312 million liters) of
wheat left Egypt each year for Constantinople. Many transportation contracts with
shipowners and some extensive records concerning the office of the praefectus annonae
Alexandriae (P.Ryl.IV 652, P.Turner 45, SB XXIV 16261) illuminate the complicated control
and accounting system.

The specific taxes, fees, customs dues, and other duties were collected by different offi­
cials, with their own types of tax receipts, which vary also by locality. Within each type,
however, the uniformity of formula for several decades is remarkable. Absolutely essen­
tial components of each tax receipt are the taxpayer's name, the tax, the year of tax as­
sessment, the amount paid, and the tax collector's (p. 383) name and title. Tax receipts
were usually written on the cheapest material available—very often potsherds, on which
they survive by the thousands.67 Tax receipts on ostraca have been found throughout
Egypt and elsewhere.68

Customs and Tolls

Large numbers of receipts for customs duties testify to the intensive movement of goods
across provincial boundaries but even more to traffic across internal barriers, where vari­
ous tolls (mostly 1–2 percent) had to be paid.69 Like tax receipts, those for customs duties
are brief, stereotyped texts, often on ostraca. Hundreds of toll receipts written in the vil­
lages on the periphery of the Fayyum (e.g., Soknopaiou Nesos, Karanis, Philadelphia ) for
the caravan drivers record their imports and exports between the oases of the Western
Desert and the Nile valley. The transport animals (donkeys, camels), the goods, and their
loads are mentioned briefly.70 There was also a charge for security, the “desert guard.”

Page 24 of 39
The Range of Documentary Texts: Types and Categories

Toll receipts appear during the early first century and disappear soon after 260 CE as a
result of changes in the customs system.

Control of Property

Also well documented are measures to control the objects of taxation. Taxpayers' declara­
tions would be checked by official “visit” (episkepsis). Many texts dealing with land sur­
vey and some tax lists of enormous length (e.g., P.Col. II), compiled on the local level,
have survived. Particularly impressive are the long tax rolls of the Ptolemaic and Roman
periods from Kerkeosiris, Karanis, Philadelphia, and Theadelphia.71 Fiscal obligations
were recorded in registers such as P.Thmouis 1 (Mendesian nome, 170/171 CE). From
Byzantine times we have accounting vouchers issued on the basis of the general register,
especially from the tax office of Hermopolis (CPR XXIV, p. 45n30), as well as extensive
records of tax payments. Apparently officials by that time preferred to keep the records
not in rolls but in codices.72

Also related to the fiscal administration are registers of landowners and lists of taxpay­
ers, some of which are extensive. Such documents occasionally reflect social and econom­
ic peculiarities such as the extensive rural exodus from Philadelphia under Nero, which
became obvious from a register of missing tax debtors (P.Ryl. IV 595 from 57 CE), or the
social structure of landowners in the Hermopolite nome around the middle of the fourth
century CE by means of a land register (P Herm Landl. 1–3). From the period of Justinian
we have the “cadastre of Aphrodito” (SB XX 14669), which lists the property of urban
landowners, along with a tax register from the same village,73 which records one year's
tax payments in gold and copper coins, and a “budget” for the nome capital Antaiopolis
(SB XX 14494).

Owners of real property were required to report it by written property returns. Under
Ptolemaic rule, private land was mainly building plots and garden land (e.g., olive and
palm groves, vineyards). Real property declarations, known from the third century BCE,
then disappear (W.Chr. 221–224, and the list in P.Heid. VII, pp. 36 f.). In the second and
first centuries BCE, a notary had to document transactions (p. 384) involving land and to
record the contract in a register. The state could thus keep track of private real property
and ensure payment of the sales tax.

Individual property returns were periodically revised and supplemented by general decla­
rations, their volume increasing with the spread of private property under the empire.
State-owned land that was leased out also generated paperwork, and unattractive parcels
could be assigned for compulsory cultivation.74 In case of crop failure or if the area had
not been flooded by the Nile, the administration granted a reduction in taxes. This re­
quired a written declaration of uninundated land; some seventy examples of this type of
document from the period from 158 to 245 CE show that requests for tax reduction were
not uncommon.75

Page 25 of 39
The Range of Documentary Texts: Types and Categories

One also was required to declare any animals that one possessed, especially camels,
sheep, and goats, which were taxed. These declarations make us realize once again what
a small fraction of the immense volume of paperwork has come into our hands. So far
forty-three declarations concerning camels are known,76 all of them from the Arsinoite
and dated to the period between 122 and 217 CE. A single example from the year 330 CE
(P.Col.X 288), however, shows that, much later, similar declarations were also demanded
from time to time. Registrations of livestock were thought to have appeared only from the
second century CE on, until BGU XVI 2578–2587 proved their existence since the begin­
ning of Roman rule.77 More than seventy registrations of sheep and goats survive from
various nomes, dating from Augustus to the third century.

All of these types of declarations were gathered in the offices of the strategos and basi­
likos grammateus and stored in the archives. The declarations, reports, and accounts of
the taxation system also had to be forwarded to the central archive in Alexandria. Official
diaries of the strategos clearly show that the original documents went to Alexandria,
while the copies remained in the locality.

Irrigation Work

Since pharaonic times the Egyptians were required to do compulsory work to maintain
the irrigation system. In the papyri, work on dikes and ditches is well documented, espe­
cially after an early Roman reorganization. Every man of “Egyptian” status was obligated
to work five days a year on the irrigation and drainage canals and received a certificate
afterward. More than 170 of such penthemeros certificates remain from different villages
of the Arsinoite nome, dating from the reign of Claudius to that of Elagabalus.78 Several
other documents (for instance, an order to compile lists of workers [SB VIII 9925] or re­
ports by sitologoi on dike work [BGU XIII 2273–2275]) confirm our impression of a tight
organization. In other nomes comparable service was required with local variations; for
example, in the Oxyrhynchite nome this service was called the “five naubion,” reflecting
the quantity of earth (the naubion) each man had to remove from the canals. Irrigation
works were checked by “visits” and reports.79 The disappearance of work certificates af­
ter about 221 clearly indicates that the jobs were administered differently, because the ir­
rigation system still had to be cared for.

(p. 385) Liturgies

In the course of the first century CE the population of Egypt was increasingly burdened
with compulsory services (“liturgies,” Greek leitourgiai). Virtually all public jobs below
the level of the strategos and basilikos grammateus (e.g., collection of taxes, security ser­
vices) had to be done for a certain period (up to three years) without payment by those
deemed able to afford the service. Liturgies were assigned proportionally according to
property. From the second to the fourth century the liturgical service was an elaborate
system, which yielded its own specific types of files and documents (e.g., P.Leit. 1–14). In
special nominations to liturgy, various officials were required to present candidates and
argue for their nomination.80 Those officials who were responsible for nominations, like
the komogrammateus and the strategos, informed themselves about the property of the
Page 26 of 39
The Range of Documentary Texts: Types and Categories

chosen candidate at the bibliothêkê enktêsêon (P.Petaus 10 and 11, P.Flor. II 206–207). Af­
ter such a check, the liturgists were usually nominated by the strategos. Sometimes high-
ranking procuratores like an epistrategos or even the idios logos were also involved in
nominations (P.Oxy. XLIX 3508, for a dike overseer). The governors' guidelines for the cri­
teria for liturgical service were collected (SB VI 9050), and inadequate candidates were
refused.81 If necessary, a substitute had to be nominated.82 The liturgical official had to
swear an oath at his assumption of office.83 Many nominees tried to avoid liturgical ser­
vice, and protests against nomination or requests for exemption refer to disqualifying rea­
sons such as advanced age, insufficient property, double burden through other liturgies,
or membership in an exempt group of people.84 Repeatedly the governors took strict
steps against the flight of liturgists, like Petronius Mamertinus in 138 CE (P.Oslo III 79).
Papyri from the second and third centuries show that in several cases the proposed candi­
dates preferred to relinquish their property rather than accept a liturgical office (W.Chr.
402; SB XXVI 16526). Especially in the fourth century it seemed necessary to guarantee
liturgical services with sureties for the officeholders (e.g., PSI I 86). During the fifth cen­
tury the system obviously underwent some significant changes, including the option for a
nominated liturgist to pay somebody else to perform the service (CPR X 17–20).

Military
Military history is probably the area that benefits most from the documentary texts found
outside of Egypt: Latin ostraca from a camp in Libya provide military reports (O.
BuNjem); the wood panels from Vindolanda at Hadrian's Wall (T.Vindol. I–III) and similar
ones from Vindonissa (T.Vindon.) come from military bases and shed light on the official
and private agenda of soldiers (and their relatives); papyri from Dura Europos written in
the scriptorium of the cohors XX Palmyrenorum represent the most extensive textual as­
semblage from the army's internal administration. (See chapter 20 for a description of the
Dura papyri and other finds from the Near East.)

The Dura papyri are our richest source of rosters like the Feriale Duranum
(p. 386)

(Rom.Mil.Rec. 117); but the Egyptian papyri also provide similar documents. A compari­
son of the duty roster of legio III Cyrenaica in Rom.Mil.Rec. 9 (90–96 CE) with the roster
on O. Claud. II 308 (150 CE) reveals that in different units the files followed exactly the
same patterns. Ostraca—in particular those from the Eastern Desert—offer various exam­
ples of the daily, official correspondence, the circulars, and the reports that had to be pre­
pared by every commander of even the smallest guardpost (praesidium).85 Besides testi­
monies for the routine guard duties and receipts for food and wine, some exceptional doc­
uments exist, such as the report about an attack by barbarian raiders on a small outpost
(O.Krok. 87).

Numerous documents deal with the interactions between the army and local society or
describe the feeding of the army. From the first until the third century, supplies were ac­
quired mainly by purchase and requisition. On such occasions the military issued receipts
in Greek. Particularly impressive is the long series of such receipts issued by horsemen to
the summus curator for money they received against their hay allowance (P.Hamb. I 39).
Occasionally papyri transmit not only complaints about requisitions by soldiers but also
Page 27 of 39
The Range of Documentary Texts: Types and Categories

complaints by soldiers about the lack of food supplies. In addition to the petitions to sol­
diers or officers mentioned earlier, papyri describe soldiers' civil duties and their rôle as
judges in lawsuits.86

Changes
Late antiquity saw a series of profound social and economic changes that manifest them­
selves importantly in the papyrological documentation. The Roman system of control of
fiscal objects and subjects gradually faded away, together with its documents, starting in
the mid-third century. Many offices changed name or definition. The old land categories
and the katoikic land also disappeared or lost meaning in the fourth century, while bu­
reaucratic control of property changed in the course of Diocletian's reforms. Papyri show
the censitor at work in Egypt, surveying the arable land for the new fiscal system
(P.Cair.Isid. 2). Many of the long-term developments can be traced in papyri, such as the
decline of the council and the bouleutic class, the expansion of the large estates, the
growth of patronage, the ascension of several families into the aristocracy of the empire
through imperial service, and the increasing interweaving of state and private spheres.
The papyrological documentation of the Byzantine period is dominated by large archives
more than in any other era.

As we have seen, from the fifth century on, contracts were drawn up by a tabellio. New
formulas are found in Byzantine contracts of lease, purchase, and loans, even where the
content remains much the same as earlier. Certain clauses mirror the changed historical
conditions; for example, landlords take over the payment of taxes on leased land. Poor
peasants and tenants depend more on the landlord for capital equipment, like spare parts
for the water wheels. With deeds of surety the semipublic Great Houses tried to counter
tendencies to flight. While in (p. 387) earlier centuries deeds of surety were common in
connection with lawsuits, they were now used as a measure against anachorêsis. More
than 180 sureties dating from the fourth to the seventh century have been edited.87 More
than the half of these belong to the sixth century. A substantial number were also under­
taken in order to release an arrested debtor or fugitive. This tradition continued later
with Coptic sureties.88

Another type of document prominent in Byzantine times is the record of private arbitra­
tion. In a compromise, two parties agreed to submit a dispute to one or more arbiters. If
they solved their case by mutual agreement—before filing a lawsuit—they drew up a writ­
ten settlement of the dispute (dialysis).89 So far approximately thirty compromises and
fifty settlements have been published. Most of them explain in detail the history of the
conflict (e.g., P Münch. I 1, 7,14; P.Mich. Aphrod.). The increasing number of these docu­
ments in the sixth century shows that people preferred to solve legal disputes privately
and to avoid litigation. The strange coincidence that about 530 CE bilingual judicial pro­
ceedings from state courts disappear has been interpreted as indicating that formal litiga­
tion was absent and private arbitration paramount for the settlement of civil disputes. Al­
though the conclusion that the Egyptian courts disappeared after Justinian has been re­
futed, a good explanation for the disappearance of proceedings has not yet been found.

Page 28 of 39
The Range of Documentary Texts: Types and Categories

Papyri offer ample information about early Christianity and its institutions (churches,
monasteries, and clergy) starting in the fourth century but little before.90 An archive con­
sisting mainly of letters and contracts gives a colorful impression of the activities in a
fourth-century Melitian monastery in Middle Egypt and its abbot, Nepheros, a “holy man”
respected for his spirituality (P.Neph.). Documents shed light on various events of Christ­
ian lives such as the priest's role as a mediator between individual and state authority, re­
questing mercy for a deserter (P.Abinn. 32), contracts that deal with the sale of a her­
mitage (P.Dub. 32–34), and the economic activities of the church as landowner (e.g., CPR
X 1–16), but there is no repertory of “Christian documents” as such.

Every survey that describes the range of documentary papyri is necessarily subjective
and selective. The diversity of texts written on papyrus is greater than any overview can
demonstrate. As authentic testimonies of societies in which only a tiny elite was able to
read and write, papyri in all their variety witness the authority and ubiquity of the written
word.

Bibliography
Anagnostou-Canas, B. 1998. “ ‘Justice’ oraculaire dans lʼÉgypte hellénistique et romaine.”
Revue historique de droit français et étranger 76: 1–16.

Baccani, D. 1992. Oroscopi greci: Documentazione papirologica (Ricerca Papirologica 1).

Messina: Sicania.

(p. 391) Bagnall, R. S. 1977. “Price in ‘Sales on Delivery.’” GRBS 18: 85–96.

———. 1995. Reading Papyri, Writing Ancient History. London: Routledge.

———, and R. Cribiore. 2006. Women's Letters from Ancient Egypt, 300 BC–AD 800. Ann
Arbor: University of Michigan Press.

Bagnall, R. S., and B. W. Frier. 1994. The Demography of Roman Egypt. Cambridge: Cam­
bridge University Press.

Biezunska-Malowist, I. 1966. “Les affranchis dans les papyrus de lʼépoque ptolémaïque et

romaine.” In PapCongr. XI, ed. O. Montevecchi, 433–443. Milan: Istituto Lombardo di
Scienze e Lettere.

Bowman, A. K. 1971. The Town Councils of Roman Egypt (American Studies in Papyrology
XI). Toronto: Hakkert.

Burkhalter, F. 1990. “Archives locales et archives centrales en Égypte romaine.” Chiron

20: 191–215.

Chalon, G. 1964. Lʼédit de Tiberius Julius Alexander: Étude historique et exégétique. Lau­
sanne: Urs Graf.

Page 29 of 39
The Range of Documentary Texts: Types and Categories

Chapa, J. 1998. Letters of Condolence in Greek Papyri (Pap.Flor. XXIX). Florence: Gonnel­
li.

Clarysse, W. 2003. “Tomoi Synkollesimoi.” In Ancient Archives and Archival Traditions:

Concepts of Record-Keeping in the Ancient World, ed. M. Brosius, 344–359. Oxford: Ox­
ford University Press.

Cockle, W. E. H. 1984. “State Archives in Graeco-Roman Egypt from 30 BC to the Reign of

Septimius Severus.” JEA 70: 106–122.

Coles, R. A. 1966. Reports of Proceedings in Papyri (Pap.Brux. IV). Brussels: Fondation

égyptologique Reine Élisabeth.

Cotton, H. M. 1981. Documentary Letters of Recommendation in Latin from the Roman

Empire. Königstein: Hain.

———, W. E. H. Cockle, and F. G. B. Millar. 1995. “The Papyrology of the Roman Near
East: A Survey.” JRS 85: 214–235.

Drexhage, H.-J. 1982. “Beitrag zum Binnenhandel im römischen Ägypten aufgrund der
Torzollquittungen und Zollhausabrechnungen des Faijum.” MBAH 1: 61–84.

———. 1989. “Zu den Überstellungsbefehlen aus dem römischen Ägypten (1.–3. Jahrhun­
dert n. Chr.).” In Migratio et Commutatio: Studien zur Alten Geschichte und deren Nach­
leben Thomas Pekáry dargebracht, ed. H.-J. Drexhage and J. Sünskes, 102–118. St.
Katharinen: Scripta Mercaturae Verlag.

Erdmann, W. 1941. “Die Ehescheidung im Rechte der graeco-ägyptischen Papyri.” ZSav

61: 44–57.

Fikhman, I. F. 1981. “Les cautionnements pour les coloni adscripticii.” In PapCongr. XVI,
ed. R. S. Bagnall, G. M. Browne, A. E. Hanson, and L. Koenen, 469–477. Chico, Calif.:
Scholars Press.

Fournet, J.-L. 1999. Hellénisme dans lʼÉgypte du VIe siècle: La bibliothèque et lʼœzuvre
de Dioscore dʼAphrodité. Cairo: Institut français dʼarchéologie orientale.

———, and J. Gascou. 2004. “Liste des pétitions sur papyrus des Ve–VIIe siècles.” In La
pétition à Byzance, ed. D. Feissel and J. Gascou, 141–196. Paris: Association des amis du
Centre dʼhistoire et civilisation de Byzance.

Gagos, T., and P. van Minnen. 1994. Settling a Dispute: Toward a Legal Anthropology of
Late Antique Egypt. Ann Arbor: University of Michigan Press.

Ghedini, G. 1923. Lettere cristiane dai papiri greci del III e IV secolo. Milan: Pubbli­
cazioni della Università Cattolica del Sacro Cuore.

Page 30 of 39
The Range of Documentary Texts: Types and Categories

Habermann, W. 1997. “Aspekte des Bewässerungswesens im kaiserzeitlichen

(p. 392)

Ägypten I: Die ‘Erklärungen für nicht überflutetes Land’ (Abrochia-Deklarationen).” In

Miscellanea Oeconomica: Festschrift Harald Winkel (Pharos 9), ed. B. Tenger and K. Ruff­
ing, 223–226. St. Katharinen: Scripta Mercaturae Verlag.

———. 1998. “Zur chronologischen Verteilung der papyrologischen Zeugnisse.” ZPE 122:
144–160.

———. 2001. “Die Deklarationen von Kleinvieh (Schafe und Ziegen) im römischen
Ägypten: Quantitative Aspekte.” In Landwirtschaft im Imperium Romanum, ed. P. Herz
and G. Waldherr, 77–100. St. Katharinen: Scripta Mercaturae Verlag.

Haensch, R. 1992. “Das Statthalterarchiv.” ZSav 109: 209–317.

Hagedorn, D. 1997. “Papyrologie.” In Einleitung in die griechische Philologie, ed. H.-G.

Nesselrath, 59–71. Stuttgart: Teubner.

Hagedorn, U. 1979. “Das Formular der Überstellungsbefehle im römischen Ägypten.”

BASP 16: 61–74.

Hengstl, J. 1972. Private Arbeitsverhältnisse freier Personen in den hellenistischen Papyri

bis Diokletian. Bonn: Habelt.

Herrmann, J. 1958. Studien zur Bodenpacht im Recht der graeco-aegyptischen Papyri

(Münch. Beitr. 41). Munich: Beck.

Husson, G. 1997. “Les questions oraculaires chrétiennes dʼÉgypte: Continuités et change­

ments.” In PapCongr. XXI, vol. 1, ed. B. Kramer, W. Luppe, H. Maehler, and G. Poethke,
482–489. Stuttgart and Leipzig: Teubner.

Jördens, A. 1993. “Kaufpreisstundungen (Sales on Credit).” ZPE 98: 263–282.

Katzoff, R. 1996. “Greek and Jewish Marriage Formulas.” In Classical Studies in Honor of
David Sohlberg, ed. R. Katzoff, 223–234. Ramat Gan: Bar-Ilan University Press.

———. 1980. “Sources of Law in Roman Egypt: The Role of the Prefect.” ANRW 2(13):
807–844.

Kim, C.-H. 1972. Form and Structure of the Familiar Greek Letter of Recommendation.
Missoula, Mont.: Society of Biblical Literature for the Seminar on Paul.

Kreller, H. 1919. Erbrechtliche Untersuchungen auf Grund der graeco-aegyptischen Pa­

pyrusurkunden. Leipzig: Teubner.

Kruse, Th. 2002. Der königliche Schreiber und die Gauverwaltung: Untersuchungen zur
Verwaltungsgeschichte Ägyptens in der Zeit von Augustus bis Philippus Arabs (30 v. Chr.–
245 n. Chr.) (APF Beih. 11). Munich: Saur.

Page 31 of 39
The Range of Documentary Texts: Types and Categories

Kühnert, H. 1965. Zum Kreditgeschäft in den hellenistischen Papyri Ägyptens bis Dioklet­
ian. Diss. Freiburg.

Lewis, N. 1974. Papyrus in Classical Antiquity. Oxford: Clarendon.

———. 1997. The Compulsory Public Services of Roman Egypt, 2d ed. (Pap.Flor. XXVIII).
Florence: Gonnelli.

MacCoull, L. S. B. 1988. Dioscorus of Aphrodito: His Work and His World. Berkeley: Uni­
versity of California Press.

Matthews, J. 2006. The Journey of Theophanes: Travel, Business, and Daily Life in the Ro­
man East. New Haven, Conn.: Yale University Press.

Mazza, R. 1998. “P.Oxy. XVI 1911 e i conti annuali dei pronoetai.” ZPE 122: 161–172.

———. 2001. Lʼarchivio degli Apioni: Terra, lavoro, eproprieta senatoria nellʼEgitto tar­
doantico. Bari: Edipuglia.

Meyer-Termeer, A. J. M. 1978. Die Haftung der Schiffer im griechischen und römischen

Recht (Stud.Amst. XIII). Zutphen, the Netherlands: Terra.

Migliardi Zingale, L. 1997. I testamenti romani nei papiri e nelle tavolette

(p. 393)

dʼEgitto. Silloge di documenti dal I al IVsecolo d.C., 3d ed. Turin: Giappichelli.

Montevecchi, O. 1972. La papirologia. Milan: Società editrice internazionale.

Müller, H. 1985. Untersuchungen zur “misthosis” von Gebäuden im Recht der gräkoägyp­
tischen Papyri. Cologne: Heymann.

Naldini, M. 1998. Il Cristianesimo in Egitto: Lettere private nei papiri dei secoli II–IV, 2d
ed. Fiesole: Nardini.

Nelson, C. A. 1979. Status Declarations in Roman Egypt (American Studies in Papyrology

19). Amsterdam: Hakkert.

Oates, J. F., R. S. Bagnall, S. J. Clackson, A. A. OʼBrien, J. D. Sosin, T. G. Wilfong, and K. A.

Worp. 2001. Checklist of Editions of Greek, Latin, Demotic, and Coptic Papyri, Ostraca,
and Tablets, 5th ed. (BASP Suppl. 9). Oakville, Conn.: American Society of Papyrologists.

Oliver, J. H. 1989. Greek Constitutions of Early Roman Emperors from Inscriptions and
Papyri. Philadelphia: American Philosophical Society.

Pestman P. W., W. Clarysse, M. Korver, M. Muszynski, A. L. Schutgens, W. J. Tait, J. K. Win­

nicki. 1981. A Guide to the Zenon Archive (Pap.Lugd.Bat. 21). Leiden: Brill.

Preissner, H. 1956. “Das verzinsliche und das zinslose Darlehen in den byzantinischen Pa­
pyri.” Diss. Erlangen.

Page 32 of 39
The Range of Documentary Texts: Types and Categories

Rupprecht, H.-A. 1967. Untersuchungen zum Darlehen im Recht der graeco-aegyptischen

Papyri der Ptolemäerzeit (Münch. Beitr. 51). Munich: Beck.

———. 1986. “Zur Frage der Frauentutel im romischen Ägypten.” In Festschrift für
Arnold Kränzlein: Beiträge zur antiken Rechtsgeschichte, ed. G. Wesener, 95–102. Graz:
Leykam.

———. 1992. “Zur Antichrese in den griechischen Papyri bis Diokletian.” In Pap.Congr.
XIX, ed. A. H. S. El-Mosallamy, vol. 2, 271–289. Cairo: Center for Papyrological Studies.

Sijpesteijn, P. J. 1987. Custom Duties in Graeco-Roman Egypt (Stud.Amst. 17). Zutphen,

the Netherlands: Terra.

Simon, D. 1982. “Das frühbyzantinische Emphyteuserecht.” In Symposion 1977: Vorträge

zur griechischen und hellenistischen Rechtsgeschichte (Chantilly, 1.–4. Juni 1977) (Akten
der Gesellschaft für griechische und hellenistische Rechtsgeschichte 3), ed. J. Modrzejew­
ski and D. Liebs, 365–422. Köln/Wien: Böhlau.

Straus, J. A. 2004. Lʼachat et la vente des esclaves dans lʼÉgypte romaine. Munich: Saur.

Thomas, J. D. 1998. “P.Ryl. IV 654: The Latin Heading.” CdÉ 73: 125–134.

Till, W. C. 1958. “Die koptischen Bürgschaftsurkunden.” BSAC 14: 165–226.

Treu, K. 1973. “Christliche Empfehlungs-Schemabriefe auf Papyrus.” In Zetesis: Bijdra­

gen op het gebied van de klassieke Filologie, Filosofie, Byzantinistiek, Patrologie, en The­
ologie, aangeboden aan Prof. Dr. Émile de Strijcker, 629–636. Antwerpen/Utrecht: De
Nederlandsche Boekhandel.

Turner, E. G. 1980. Greek Papyri: An Introduction, 2d ed. Oxford: Clarendon Press.

Valbelle, D., and G. Husson. 1998. “Les questions oraculaires: Histoire de la recherche,
nouveautés et perspectives.” In Egyptian Religion: The Last Thousand Years. Studies Ded­
icated to the Memory of J. Quaegebeur, ed. W. Clarysse, A. Schoors, and H. Willems, vol.
2, 1055–1071. Leuven: Peeters.

Van Minnen, P. 2000. “An Official Act of Cleopatra (with a Subscription in Her Own
Hand).” AncSoc 30: 29–34.

Whitehorne, J. 2004. “Petitions to the Centurion.” BASP 41: 161–169.

Wolff, H.J. 1961. “Der byzantinische Urkundenstil Ägyptens im Lichte der Funde
(p. 394)

von Nessana und Dura.” RIDA 8: 115–154.

———. 1974. “Zur Geschichte der Sechszeugen-Doppelurkunde.” In PapCongr XIII, ed. E.

Kiessling and H.-A. Rupprecht, 469–479. Munich: Beck.

Page 33 of 39
The Range of Documentary Texts: Types and Categories

———. 1978. Das Recht der griechischen Papyri Aegyptens in der Zeit der Ptolemäer und
des Prinzipats, vol. 2: Organisation und Kontrolle des privaten Rechtsverkehrs. Munich:
Beck.

Worp, K. A. 1995. “Letters of Condolence in the Greek Papyri.” Analecta Papyrologica 7:

149–154.

Yiftach, U. 2001. “Was There a “Divorce Procedure” among Greeks in Early Roman
Egypt?” In PapCongr. XXII, ed. I. Andorlini, G. Bastianini, M. Manfredi, and G. Menci, vol.
2, 1331–1339: Florence: Istituto Papirologico “G. Vitelli.”

———. 2002. “Deeds of Last Will in Graeco-Roman Egypt: A Case Study in Regionalism.”
BASP 39: 149–164.

Yiftach-Firanko, U. 2003. Marriage and Marital Arrangements: A History of the Greek

Marriage Document in Egypt, Fourth Century BCE–Fourth Century CE (Münch. Beitr. 93).
Munich: Beck.

Zuckerman, C. 1998. “Two Reforms of the 370s: Recruiting Soldiers and Senators in the
Divided Empire.” REB 56: 79–139.

———. 2004. Du village à lʼEmpire: Autour du registre fiscal dʼAphroditô (525/526). Paris:
Association des Amis du Centre dʼHistoire et Civilisation de Byzance.

Notes:

(1.) Habermann (1998, 144–160).

(2.) For the spread of papyrus in the ancient world, see Lewis (1974, 84–94); cf. chapter
20 for papyri written outside Egypt.

(3.) Montevecchi (1972, 86–89); compare the “Sachübersicht” in SB XXVI pp. V–XI with
ten main categories and many more subcategories.

(4.) Masterly surveys of the papyrological evidence are Turner (1980, 127–153); Bagnall
(1995, 9–31); and Hagedorn (1997, 59–71).

(5.) Haensch (1992, 209–317).

(6.) Cockle (1984, 106–122); much is said on archives by Wolff (1978) and Burkhalter
(1990, 191–215).

(7.) For collections of papyrus letters cf. the references in Checklist (chapter 3)by Oates et
al. (2001); all of the anthologies contain numerous private letters as well.

(8.) Bagnall and Cribiore (2006).

Page 34 of 39
The Range of Documentary Texts: Types and Categories

(9.) For example, the letters of Apion (alias Antonius Maximus) (Sel.Pap. I 112; BGU II
632), Apollinaris (P.Mich. VIII 490, 491), and Claudius Terentianus (P.Mich. VIII 468, 476),
all from the second century CE.

(10.) Famous examples are Hilarion's letter to his wife (P.Oxy. IV 744) and complaints by
the boy Theon (P.Oxy. I 119).

(11.) Cotton (1981); Kim (1972); Treu (1973, 629–636).

(12.) Collected in Chapa (1998). For observations on the genre, see Worp (1995, 149–
154).

(13.) Cf. P.Oxy. LII 3693, introduction, with references and literature.

(14.) Wolff (1961, 115–154); A. Jördens, P.Heid. V, pp. 371–373.

(15.) Straus (2004, 345–349).

(16.) C.Ptol.Sklav. 28–37; Biezuńska-Malowist (1966, 433–443).

(17.) M.Chr. 320 in Greek; ChLA V 290 and IV 269 in Latin.

(18.) For procedure and documents see Rupprecht (1986, 95–102); supplements in
P.Hamb. IV 270, introduction.

(19.) For example, see Jur.Pap. 10; M.Chr. 363; P.Oxy. XVI 1895; P.Köln VII 321.

(20.) Yiftach-Firanko (2003), with the sources on pp. 9–39. For well-preserved marriage
contracts see Sel.Pap. I 2–5.

(21.) Katzoff (1996, 223–234).

(22.) For divorce agreements see Erdmann (1941, 44–57); Yiftach (2001, vol. 2, 1331–
1339).

(23.) Well over 220 examples survive; cf. Yiftach (2002, 149–164). The basic literature is
still Kreller (1919).

(24.) Migliardi Zingale (1997).

(25.) The cavalry officer Dryton drew up at least three testaments—in 164, 150, and 126
BCE: P.Dryton 1, 2, 3, and 4, a copy. On the revocation of a will see Sel.Pap. II 424; P.Oxy.
I 107; SB VIII 9766; and P.Oxy. XXXVI 2759.

(26.) Listed in Jördens, P.Heid. V, pp. 296–301, and Kruit (1992,167–84, supplement). On
the problem of hidden loans see Bagnall (1977, 85–96).

(27.) For example, see M.Chr. 226, SB XXII 15703; cf. Jördens (1993, 263–282).

Page 35 of 39
The Range of Documentary Texts: Types and Categories

(28.) Ptolemaic: Rupprecht (1967); Roman: Kühnert (1965); Byzantine: Preissner (1956).
The HGV lists almost two thousand loans and related documents.

(29.) Explicitly stated in Jur.Pap. 22a, col. 1, 9–12. M.Chr. 334 and 335 maybe “deposits”
of this kind.

(30.) Herrmann (1958, 247–288) compiled an impressive list of 42 Ptolemaic, 228 Roman,
and 198 Byzantine land leases. Since then the numbers have increased significantly.

(31.) Simon (1982, 365–422).

(32.) Müller (1985, 345–361).

(33.) Hengstl (1972); Jördens, P.Heid. V, pp. 125–375.

(34.) Meyer-Termeer (1978, 88–104); supplements in P.Wash.Univ. II 82, introd.

(35.) Sources and discussion in Rupprecht (1992, vol. 2, 271–289).

(36.) A quick impression may be obtained from W.Chr. 65–135, Sel.Pap. I 193–200, and
C.Pap.Hengstl 56–71.

(37.) Representative collections: C.Étiq.Mom.; SB III 7037–7125; SB V 7697–7736.

(38.) Valbelle and Husson (1998, vol. 2, 1055–1071).

(39.) Anagnostou-Canas (1998, 1–16).

(40.) SB XIV 11658; SB XXVI 16703; SB XVlll 13250; Husson (1997, vol. 1, 482–489).

(41.) J. R. Rea, P.Oxy. LVIII 3929 introd., and R. Duttenhöfer, P.Lips. II 152 introd., both
with further references.

(42.) Zuckerman (1998, 79–139).

(43.) Zuckerman (2004, 189–212).

(44.) Collected by Oliver (1989), together with the epigraphical evidence.

(45.) For surveys on edicts see Katzoff (1980, 807–44) and Chalon (1964, 251–256).

(46.) Early examples are P.Petr. II 27 (2); BGU VIII 1767, 1768; late examples are P.Oxy.
LX 4075; CPR XVIIA 18.

(47.) Oxyrhynchus: P.Oxy. XII 1413–1419; XXVII 2475–2477; Hermopolis: SPP V contains
an extensive set of proceedings from the council (mid-third century CE); cf. Bowman
(1971, 32–34).

(48.) Compare the photo of SB I 4639 in Schubart (1911, pl. 35) with that of ChLA XLVII
1421.

Page 36 of 39
The Range of Documentary Texts: Types and Categories

(49.) For a representative selection see M.Chr. 20–33, 50–55, and 79–97.

(50.) Coles (1966, 55–63).

(51.) Catalogued in Thomas (1998, 132–134); supplements in D. and U. Hagedorn,

P.Thomas, p. 217n1.

(52.) P.Oxy. LXIII 4381; P.Acad. 56/1 + 2 + 57/1; and ChLA XLVII 1437.

(53.) P.Mich. X 589–591, introd.; Hagedorn (1979, 61n2); Drexhage (1989, 102–118).

(54.) Clarysse (2003, 344–359). An instructive picture of a tomos synkollêsimos is Pap.

Lugd.Bat. V, pl. 1, in which eighteen declarations from 174 CE are pasted together.

(55.) Whitehorne (2004, 161–169).

(56.) Montevecchi (1972, 189–192) and Fournet and Gascou (2004, 141–196).

(57.) P.Bingen 105, introduction. Well-preseved examples are P.Fay. 28; P.Gen. I2 33;
P.Petaus 1 and 2 contain two copies of a birth notification for an eight-year-old girl; Pap.
Lugd.Bat. VI 33 for an Antinoopolite.

(58.) Nelson (1979, 3–40); Kruse (2002, vol. 1, 252–271; vol. 2, 638–640).

(59.) Haensch (1992, 313–317, app. 4); on the epikrisis papers of Roman citizens see
P.Diog. 6 and 7; on those of illegitimate children see SB I 5217, P.Oxy. XII 1451, BGU IV
1032; on those of slaves see PSI IV 447.

(60.) Approximately twenty applications for circumcision have been edited; cf. Kruse
(2002, vol. 2, 728–733).

(61.) Bagnall and Frier (1994, 179–312) (list of documents).

(62.) Examples are P.Lond. III 904; SB XX 14662; P.Fay. 24.

(63.) See C.Pap.Gr. II; Kruse (2002, vol. 1, 139–168; updated list on page 143n256).

(64.) Normative sources are few; the most famous is perhaps P.Rev., a collection of tax
regulations.

(65.) Hundreds of receipts of sitologoi demonstrate this procedure from the mid-third cen­
tury BCE to the mid-fourth century CE.

(66.) A most impressive example is the extensive roll P.Col. II 1 R 4 + BGU XIII 2270 +
2271 + P.Berl.Frisk 1 + P.Graux III 30 + SB XVI 13060 (verso: P.Col. V 1, verso 4; P.Graux
IV 31), Arsinoite nome, 155 CE.

(67.) Thousands of tax receipts appear in O.Amst., O.Amst.Shelt., O.CairoGPW, O.Wilck.,

O.Bodl. I–II, O.Elkab, O.Heid., and O.Ont.Mus. I–II, among others.

Page 37 of 39
The Range of Documentary Texts: Types and Categories

(68.) Eastern Desert: O.Claud. I–IV, O.Krok.; Western Desert: O.Douch. I–V, O.Kellis. Tax
receipts from outside Egypt are similar to those in Egypt (e.g., P.Petra I 7–10).

(69.) Sijpesteijn (1987, 23–25).

(70.) For lists of such receipts see Drexhage (1982, 61–84); Sijpesteijn (1987, 102–143,
nos. 1–919).

(71.) P.Count I 22–44; P.Tebt. I; P.Mich. IV; BGU IX 1891–1989; P.Col. V 1, verso.

(72.) Tax codices: CPR V 26; P.Sorb. II 69; MPER IX 44–53 + 56 + SB XXII 15711.

(73.) See Zuckerman (2004, 248–271).

(74.) Kruse (2002, vol. 1, 559–609).

(75.) Habermann (1997, 223–226); Kruse (2002, vol. 1, 235–250).

(76.) Kruse ( 2002, vol. 1, 181n361–364).

(77.) Habermann (2001, 77–100, esp. 89 ff).

(78.) The earliest known penthemeros certificate is P.Fay. 286 (41–54 CE); the latest,
P.Fouad I 62 (221 CE). For a list of documents see P.Mich. XVI 141–147.

(79.) For example, see P.Köln VIII 341 verso; P.Tebt. III 1 725; W.Chr. 389.

(80.) Nominations for liturgy are listed in Lewis (1997, 110 f). Cf. also Sel.Pap. II 342–345
(second—fourth centuries CE) and the catalogue of liturgies in P.Petaus 88.

(81.) For example, see P.Wisc. II 81: The prefect reproves a komogrammateus because he
has nominated an unqualified person; CPR XXIII 27: A strategos/exactor refers an inad­
missible nomination back to the boulê.

(82.) For example, see P.Louvre II 114; P.Gen. I2 37.

(83.) See P.Iand. III 33; P.Leit. 12; P.Oxy. XLIII 3097, 3098.

(84.) Just a few examples: W.Chr. 396; W.Chr. 29; SB XIV 11980; and SB XII 10797.

(85.) O.Claud. I–III, O.Krok., O.Florida, and O.Douch I–V.

(86.) M.Chr. II 84; P.Mich. III 159; P.Gen. I2 74; P.Oxy XIV 1637.

(87.) Listed in B. Kramer, P.Heid. IV, pp. 118–125 (fourth cent.); G. Bastianini, Misc.Pap. I
25–27 (fifth–seventh centuries); P.Vind.Sijp. 17–21 (deeds of sureties for liturgists), and
Fikhman (1981, 469–477).

(88.) Till (1958, 165–226); CPR IV 102–112.

Page 38 of 39
The Range of Documentary Texts: Types and Categories

(89.) Compromises: Rupprecht (1997, 267–268); settlements: Gagos and van Minnen
(1994, 121–127).

(90.) Papyrological sources are collected in NewDocs I–IX. For anthologies of Christian
letters see Ghedini (1923); Naldini (1998).

Bernhard Palme

Bernhard Palme is Professor of Ancient History and Papyrology, University of Vienna.

Page 39 of 39
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation

The Multilingual Environment of Persian and Ptolema­

ic Egypt: Egyptian, Aramaic, and Greek Documentation

Dorothy J. Thompson
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0017

Abstract and Keywords

This article considers the extent to which the surviving documentation from the five-hun­
dred-year period from Cambyses to Cleopatra allows us to investigate and reconstruct the
changing contexts of language use and linguistic practices. It looks primarily at the main
languages of the papyrological documentation such as Egyptian (demotic script), Arama­
ic, and Greek. It also considers Phoenician, Carian, Latin, and other languages. It exam­
ines who used which language and in what contexts, how widespread bilingualism may
have been in different periods of non-Egyptian rule, how far Aramaic under the Persians
and later, following Alexander's conquest, Greek took on the role of prestige languages,
and, in contrast, which areas were linguistically unaffected by foreign conquest.

Keywords: Cambyses, Cleopatra, language use, linguistic practices, demotic script, papyrological documentation,
Aramaic language, bilingualism, Alexander's conquest

Introduction
When in 525 BCE Cambyses, the great king of Persia, captured Egypt and initiated a new
dynasty, he was represented in this new province of the Persian empire by his (p. 396)
satrap, and Aramaic became the main language of the administration. Around half a mil­
lennium later, with the death of Cleopatra VII and the capture of Alexandria by the Ro­
man general Octavian in 30 BCE, yet another new dynasty of absentee pharaohs was es­
tablished, and Egypt was incorporated into the Roman empire. This time, however, the
language of the administration remained Greek, as it had been under the intervening
Ptolemaic dynasty. During the Persian period (525–404 and 343–332 BCE), Aramaic, as
the language of rule, was just one of the different languages in use in Egypt. Similarly,
during the period of Ptolemaic rule (332–30 BCE), which followed Alexander of

Page 1 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation
Macedon's conquest of Egypt, though Greek was the main non-Egyptian language em­
ployed, it was not the only one.

This chapter is concerned with the extent to which the surviving documentation from the
five-hundred-year period from Cambyses to Cleopatra allows us to investigate and recon­
struct the changing contexts of language use and linguistic practices. The main lan­
guages of the papyrological documentation treated here are Egyptian (primarily in the
demotic script), Aramaic, and Greek, but Phoenician, Carian, Latin, and other languages
also make an appearance. Who used which language and in what contexts, how wide­
spread bilingualism may have been in different periods of non-Egyptian rule, how far Ara­
maic under the Persians and later, following Alexander's conquest, Greek took on the role
of prestige languages, and, in contrast, which areas were linguistically unaffected by for­
eign conquest are all questions worth consideration. In addition, the limits to the evi­
dence of the papyri and the need to look elsewhere—at epigraphy, for instance, or materi­
al culture—must enter the broader historical picture (Vittmann 2003). And if in the end
some questions remain still unanswered, this simply reflects some of the limitations of pa­
pyrological evidence. For in addition to the hazard of the uneven survival of documenta­
tion, there is also the inherent problem of arguing from literate to oral practice, from the
written, that is, to the spoken word. In any noncontemporary society a gap remains here
that is hard to bridge. The concentration, therefore, in what follows on the social context
of multilingualism is probably unavoidable (Adams and Swain 2002, 11).

The Persian Period

The key areas for investigation in the Persian period, as indeed later under the Ptolemies,
are the central administration and the different local communities in their official and
their private dealings, both in urban centers and in the countryside of the Nile valley, the
Delta, and the western oases. We need to consider both who produced the written record
that survives and how far we can move from this to actual language choice in these vari­
ous areas. In the absence of information on (p. 397) the scale of immigration or the extent
of new settlement in Egypt in this period, we cannot assess the resulting changes in lin­
guistic practices in anything other than the most general manner. Nevertheless, as is
clear from surviving texts, the population of Egypt under Persian control was a mixed one
with many different peoples who spoke (and often also wrote) a range of languages. Some
groups were already well established when the Persians came: the Ionian, Carian, and
Jewish troops employed and settled under the preceding Saite dynasty. Phoenician set­
tlers, too, who were often involved in shipping and commercial ventures, may be traced
back even further; an additional military role is suggested by the name of the Phoenician
quarter at Memphis—the “Tyrian Camp” (Herodotus 2.112.2). With the Persian conquest,
a new influx into Egypt of administrators, garrison troops, and settlers may be document­
ed. Living in the southern fortress settlement at Elephantine, for instance, were Babyloni­
ans, Bactrians, Caspians, Chorasmians, Medes, Persians, Aramaeans, and Jews (Porten et
al. 1996, 18). At the important religious center of Abydos, graffiti were written in Arama­
ic, Phoenician, and Carian in this period, as well as in Greek in the Cypriot syllabary. Now
Page 2 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation
written also in the demotic script, Egyptian, of course, remained the main language of the
country. Demotic was also used in parts of the administration, as, for instance, on ostraca
recording water rights at Manâwir in the Kharga oasis (O.Douch.dem. and O.Man., forth­
coming).

The main non-Egyptian documentation of the Persian period is that in Aramaic, which
served as the administrative language throughout the Persian empire. After Egyptian,
Aramaic was the most widespread language employed in this period, and the main edi­
tions of Aramaic texts (not included in the Checklist) are those of Cowley (1923), Aimé-
Giron (1931), Kraeling (1953), Driver (1957), Bresciani and Kamil (1966), and Segal
(1983), with a selection of translated texts in Grelot (1972) and Porten et al. (1996, for
Elephantine); Porten and Yardeni (1986, 1989, 1993, 1999) have republished these,
adding some new material. To judge from the smaller number of surviving texts, Phoeni­
cian was a minority language (Segal 1983, 5, 9–10, 139–145, from Elephantine and Mem­
phis; Vittmann 2003, 44–83, more generally), along with Carian, Nubian, and other
tongues (Ray 1994, 51).

From Elephantine and Syene (modern Aswan), both garrisons on the southern border,
from Hermopolis and Memphis in Middle Egypt, and to a lesser degree from elsewhere,
texts have survived from licit and illicit excavations (Vittmann 2003, 84–119). Groups of
letters like those from Hermopolis, which originated in Memphis (Bresciani and Kamil
1966; Pap.Eleph.Eng. B1–7; Vittmann 2003, 90–91), family papers, official rulings and cor­
respondence, together with economic documents, all help to illustrate the life of these
communities of foreign settlers. Language plays a complex role in cultural variation, and
other forms of evidence may supplement the papyri. From Edfu, for instance, in the
fourth century BCE, Aramaic tombstones imply a Semitic community in the city (Kornfeld
1973). Edfu may be the origin, too, of Papyrus Amherst 63, a long Aramaic religious text
actually penned in Egyptian script. The demotic writing of this intriguing text implies
both a (p. 398) scribe and a reader more familiar with the script of the land they lived in
than with that of the text in question (Vleeming and Wesselius 1985). As illustrated fur­
ther below, the phenomenon of nonmatching language and script is a feature of multilin­
gual societies.

From Memphis, the papyrological documentation is complemented by Herodotus's fifth-

century-BCE account of this capital city (2.112). Moreover, since Jews were among the
Semitic settlers of Late-Period Egypt, the Old Testament joins the surviving documenta­
tion in illustrating the mixed society of the times (Porten 1968, 3–27). The prophet Jeremi­
ah (44.1), for instance, mentions four Jewish communities already established in Egypt
before the Persian conquest at Migdol (Pelousion), Tahpanhes (Daphnae), Noph (Mem­
phis), and Pathros (the south), presumably centered on Elephantine, where a Jewish tem­
ple was founded before Cambyses' invasion (Pap.Eleph.Eng. B19.13–14, from 407 BCE).
The key border towns of Elephantine and Syene are the source of the most extensive pa­
pyrological documentation for this period (Porten 1968; Porten et al. 1996). From these

Page 3 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation
and other papyri, the balder accounts of the literary sources are filled in, and the picture
takes on life.

Much of the Aramaic documentation from the Persian period is of a private nature, deriv­
ing from a limited number of areas, where groups of family documents were preserved in
pots or other containers. One set of texts, however, is written on skins, described by
Herodotus (5.58.3) as the standard writing material of Ionia, and preserved together with
a leather bag. An Egyptian origin seems likely since these texts, which date from a period
when the satrap Arsames was out of the country (ca. 410 BCE), contain correspondence
about his Egyptian interests, including revenues from his land (Driver 1957; Grelot 1972,
nos. 63–74). Just as later members of the high administration—the dioikêtês Apollonios
(under Ptolemy II), for instance—received gift estates in the Egyptian countryside, so un­
der the Persians it is clear that this satrap, related as he was to the Achaemenid royal
house, had acquired extensive landed interests in the province. And in Arsames' close cir­
cle of associates and senior administrative officials we meet one Neḥtiḥôr, who was in
charge of the satrap's Egyptian estates. His name suggests an Egyptian. We may assume
that, like Udjaḥorresnet from Sais, who some generations before served at the court of
Cambyses and then at that of Darius (Lloyd 1982; Briant 2002, 57–59), Neḥtiḥôr was flu­
ent in the conqueror's language, in this case Aramaic, which he used in correspondence
with the satrap. Reflecting the advantages that come with the acquisition of the rulers'
tongue, such high-level Egyptian cooperation—collaboration, even—is not at all surpris­
ing. It recurs under the Ptolemies and was probably more widespread than a simple con­
sideration of names might suggest since, unlike Neḥtiḥôr, many Egyptians opted for
Greek nomenclature and language as being those of the new ruling class.

Names, therefore, are not entirely reliable as an indicator of an individual's ethnic origin.
Nonetheless, especially on a group basis, they may serve as an introduction to the compo­
sition of a particular society. The Aramaic texts from Memphis, for instance, contain
Babylonian, Aramaean, Sidonian, Jewish, and even Moabite names; they also make refer­
ence to Ionians, Carians, a Cretan slave and his daughter, (p. 399) a Hyrcanian, a (possi­
ble) Lydian, and Arabs as well (Segal 1983, 8). How many different languages were in­
volved is unclear, but Memphis was a cosmopolitan city with a thriving Nile port and
dockyards. Official accounts were kept in Aramaic, and many scribes were apparently in­
volved in the work; one Memphite shipyard journal, for instance, records members of Per­
sian military units and adds a Caspian to the list of non-Egyptians in the city (Porten and
Yardeni 1993, C3.8, from 473 BCE). Texts like this or the ten-month-long customs ac­
count, later reused for a copy of the “Romance of Aḥiqar” (ibid., C3.7, from 475 bce; cf.
C1.1, with Briant 2002, 385), perhaps again from Memphis, are the product of a well-em­
bedded administration that utilized Aramaic speakers and scribes up and down the Nile
and at the major guardhouses. Fragments of household listings (Porten and Yardeni 1993,
C9–10) and records of land registration (ibid., C20–24) imply a thorough control of both
population and land throughout the country. The scale of the bureaucracy involved can
only be imagined, but to judge from Ptolemaic practice it must have been significant.

Page 4 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation
Yet throughout the period of both Persian and Ptolemaic rule, the language of the majori­
ty remained Egyptian. Demotic was the main script employed in legal documents, letters,
accounts, and other texts. With demotic, more than any other of the written languages
and scripts, we face the problem of how far the written text signifies the spoken lan­
guage. Put simply, once Egyptian begins to be written in Coptic, the language appears to
be full of Greek words taken over into Egyptian, up to 20 percent on one calculation (Few­
ster 2002, 228). Of this there is little sign at an earlier date, when few Greek words are
found in demotic (Clarysse 1987), at least in legal, literary, and religious texts. Moreover,
if, as J. D. Ray has argued (1994, 53–54), demotic was originally developed as a stylized
version of hieratic to serve the needs of a central administration in a reunification of
Egypt under Saite rule, then it is likely that this form of the Egyptian language was delib­
erately preserved in a relatively uncontaminated state. The written script will have dif­
fered from the spoken language, and the degree to which Aramaic and, later, Greek vo­
cabulary or figures of speech were incorporated into Late Period Egyptian is unknowable
through demotic.

The Ptolemaic Period

When Alexander of Macedon invaded Egypt in 332 BCE and captured this Persian satrapy,
the Persian administration and Aramaic-based bureaucracy speedily disappeared to be re­
placed by Greeks and Greek. Some of the earlier foreign communities remained, though
the degree of continuation or of new immigration that this involved cannot be deter­
mined. Thus, when an Aramaic tax receipt is found from the Theban area issued by one
Joseph to a certain Simeon in 252 BCE, some eighty (p. 400) years after the conquest, we
cannot tell whether the taxpayer was a descendant of earlier settlers or belonged to a
new wave of immigrants under Ptolemy II (Porten and Yardeni 1999, D8.13, cf. 2004/5). In
some centers, there is good evidence for the continued use of Aramaic in both spoken and
written form (Clarysse 2002, 8, in Thebes, Edfu, and Tebtynis). On the whole, however,
the Ptolemaic administration increasingly functioned in Greek. As the papyri indicate,
Greek was now by far the most common non-Egyptian language of the period.

Along with this change in the language of rule came a change in mortuary practices that
has significant effects on our evidence. An innovation in the manufacture of mummy cas­
ing under Ptolemy II involved the recycling of waste papyrus, mainly from scribal offices,
in a form of papier mâche or “cartonnage” (see chapter 4). This new source of papyrologi­
cal texts is responsible for the bureaucratic slant of Ptolemaic documentation. Numerous
administrative texts, official rulings, and reports now join the private letters, family
archives, and the stray or collected texts that have been found in excavation. The appar­
ent growth in bureaucracy under the Ptolemies may in part be a feature of this new prac­
tice. Furthermore, the increase in administrative documentation makes it possible to ana­
lyze the population in ways not feasible before. Land surveys and tax registers allow the
historian to quantify, at least to some degree, the mixed population of the period and the

Page 5 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation
main economic basis of society. Only when we have a sense of the major groups within
the population can we place the role of the more minor ethnic communities in context.

So what do we know of the makeup of the population of Ptolemaic Egypt? How wide­
spread were the new Greeks, who came as members of the army, as officials in the new
administration, as merchants, freebooters, adventurers, or as craftsmen to settle in
Egypt? What other non-Greek immigrants of long-standing and more recent origin had
their homes there, what roles did they play, and how readily were they accepted or assim­
ilated? Finally, to what extent can the papyri help answer our questions?

As noted already for Persian-period Egypt, one cannot expect to find a consistent pattern.
The newcomers settled in certain areas, especially, of course, the capital. In the earlier
capital of Memphis, the strange group of Greek poets and philosophers and the Dionysiac
sculpture of the main temple avenue to the Serapeum imply a strong Greek presence
(Lauer 1976, plates 1–5, 8–9). Under Ptolemy I, the capital was moved to Alexandria,
which was now developed as a strong center of Greek culture. Here, in the context of the
Museum and the Library, the story of the translation of the Septuagint and the circulation
in Greek of translations from Egyptian (the Dream of Nectanebo, for instance; see UPZ I
81; LDAB 6863) suggest a culture of bilingualism. Unlike the situation elsewhere in the
country, in Alexandria Greeks most probably formed a majority, but the damper climate of
the coastal region means that the only papyri to have survived from this city are those
found upcountry. For the same reason, the situation in the (p. 401) Delta is also virtually
unknown in this period, apart from a few inscriptions and gravestones.

The second main center of Greek settlement that we know of was the Fayyum, a marshy
area drained early in the Ptolemaic period, where the survival of cartonnage texts from a
necklace of surrounding cemeteries allows us to document the local population with its
heavily immigrant element. This area (renamed the Arsinoite nome under Ptolemy II) was
primarily used for the settlement of soldiers, as cleruchs with plots of land sufficient to
provide a living for themselves and their families. Where the military went, others fol­
lowed, and Greeks here are documented as teachers and doctors, wine producers or wine
merchants, as well as actors and artists, working in a whole range of businesses and offi­
cial capacities. Overall in this nome, in the mid-third century BCE, Greeks accounted for
around 18 percent of the civilian population, and when the military element is added, the
figure for Greeks rises to more than 30 percent of a total of approximately eighty-five
thousand (Clarysse and Thompson 2006, 2: 94–95, 154–157). Some of these “Greeks” may
not themselves have been from families that originated in Greek lands, since the term
Hellene under the early Ptolemies came to represent a tax category for a privileged group
among the population, one that might include Egyptians. Nevertheless, such Egyptians
counted as Greek and, by functioning in Greek in various positions within the Greek
sphere, to all intents and purposes they were Greek.

Under the Ptolemies, primarily as a result of cleruchic settlement but also of other oppor­
tunities in this flourishing kingdom, Greeks were to be found throughout the country (see,
for example, Clarysse [1995] on Thebes). The degree, however, of Greek settlement in the

Page 6 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation
Arsinoite nome was certainly exceptional. In the Apollonopolite, or Edfu nome, to the
south, a land survey from 119/118 BCE records just 3 percent of all land as held by
cleruchs (P.Haun. inv. 407, ed. Christensen 2002), an interesting contrast to the 33 per­
cent of cleruchic land recorded for the Arsinoite village of Kerkeosiris in the same year
(Crawford 1971, 44). And since cleruchic plots were significantly larger than average,
Greek landholders in the Apollonopolite must have represented a very small proportion of
the nome's total population. Figures for the country as a whole are lacking. Overall, how­
ever, we may surmise that Greeks are unlikely ever to have represented more than 15
percent of the total and that this was probably a higher figure than that for immigrant
settlers in the preceding Persian period, for these two empires differed in their military
requirements. Both needed garrisons within their country to ensure their hold on their re­
spective countries, but under the Ptolemies a strong national army was additionally re­
quired to protect and extend their kingdom in a world of competing Hellenistic powers.
The Great King faced no such competition.

How Greek did Ptolemaic settler Greeks remain, and to what extent were they assimilat­
ed within the population? Education, the legal system, and the use of Greek (though not
exclusively so) within the administration, the army, and social (p. 402) institutions like the
gymnasium all helped to endorse Greek status (Goudriaan 1988; Thompson 2001). Never­
theless, certain countervailing tendencies (intermarriage, for instance) and local ways,
especially local religion, played a part in the progressive Egyptianization of the Greeks of
Ptolemaic Egypt. Finally, when the Romans came and instituted their rule, those Greeks
who were not citizens of the three Greek cities of Alexandria, Naucratis, and Ptolemais
were now all counted together under the heading of “Egyptians.” These different process­
es are known primarily from the papyrological record. In what follows, and particularly in
the case studies at the end of this chapter, only some aspects of the social and linguistic
context can be examined.

The recognition of different ethnic groups is normally made on the basis of names. Yet, as
already indicated, these are sometimes problematic. The same individual may use differ­
ent names in different contexts, and in Ptolemaic Egypt the use of double names—names
in both Greek and Egyptian—is a complicating factor (Clarysse 1985, 1992). Two exam­
ples will illustrate the problem we face—the impossibility of ascribing identity according
to an individual's name.

In an Arsinoite salt-tax register from 229 BCE, two sons of Nehemsesis are listed living
next door to one another (P.Count 4.113–116). The first, Petoys, is a policeman; for his
brother, no occupation is given, but the lower tax dues recorded under his name indicate
a Hellenic tax status, which is corroborated by his Greek name, Pasikles. We do not know
at what stage Pasikles acquired a Greek name, but he clearly belonged to that bilingual
sector of society which moved with ease between the two worlds, Egyptian and Greek. Po­
licemen, too, like his brother, Petoys, formed part of the Greek side of society; from the
second century BCE on, such men might be included in cleruchic allotments. Living most
probably in the capital of the nome, the two sons of Nehemsesis belonged to a group of

Page 7 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation
Egyptians who played an important role in the success of the Ptolemaic regime. It seems
safe to assume that they were bilingual, though we have no direct evidence of this.

My second example is the dioikêtês Dioskourides, an official from the top echelons of the
Ptolemaic administration in the mid-second century BCE, well known from the papyri (see
UPZ I 14). Philippe Collombert's study of a stone anthropoid sarcophagus with long hiero­
glyphic inscriptions brings surprising new information on this high official (Collombert
2000; cf. Coulon 2002). Identified not only by his name and position (snty is the Egyptian
for Greek dioikêtês), Dioskourides is also given his court title of archisômatophylax, hesi­
tantly transliterated into Egyptian as m3rkysmṭpyrks. His mother's name was Tetosiris,
but his father is not named. The most interesting feature of this sarcophagus, however, is
that a high official in the administration, the son perhaps of a mixed marriage, chose to
be buried according to his mother's ethnicity. This man, who spent his career in Greek of­
ficialdom, where the language of the administration was primarily Greek, would have
moved with equal facility and confidence in the Egyptian circles from which (at least) his
mother came. Furthermore, in the case (p. 403) of Dioskourides, it is material culture—his
sarcophagus—which adds interest to the papyrological record. Not everything is what it
seems to be at first. Both Pasikles and Dioskourides may be added to the ranks of those
bilinguals who regularly moved between two worlds. Like their ethnicity, our evidence is
context specific.

In attempting to quantify the extent of intermarriage in Ptolemaic Egypt, we again rely on

names. A database of 427 adult households, derived from the texts of P.Count from the
mid-third century BCE, yields some interesting information. Wives are recorded in 75 out
of 85 households in which the name of the household head is Greek. Of these wives, the
names of all but six are also Greek. On this evidence, therefore, 92 percent of Greek
males married women who were Greek. Among the far larger number of couples where
the husband has an Egyptian name, not a single one is registered with a Greek-named
wife. According to this third-century evidence, then, intermarriage would seem to have
occurred in only one direction. The data of these households are found to need a little
modification when data from all of the P.Count tax-paying couples from the third and sec­
ond century are considered together. Out of a total of 685 couples in which the name of
both husband and wife is known, 105 Greek-named husbands (74.5 percent) were mar­
ried to wives with Greek names, and 36 (25.5 percent) to women with Egyptian names. Of
those 544 husbands with Egyptian names, 536 (98.5 percent) had wives with Egyptian
names, while the wives of just 8 (1.5 percent) had Greek names; furthermore, most of this
last group were from the second century BCE (Clarysse and Thompson 2006, 2: 297, 327–
328). An important feature of papyrological documentation is that it allows this form of
quantification; once again, more data add complexity to the linguistic picture.

Differences in the Greek and the Egyptian sectors of the population as identified by their
names and documented in their marriage patterns are evident in other areas as well. As
already noted, land surveys bear witness to the Greeks' more extensive land holdings.
They also had larger households. In the database of 427 tax-paying households of P.
Count, no Egyptian household numbers more than eight, whereas one Greek household

Page 8 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation
consisted of twenty-two adults. A significant lack of daughters is also apparent in Greek
households; this is not characteristic of Egyptian families. Various explanations maybe
considered, including the exposure of newborn girls by only the Greek side of the popula­
tion. Slaveholding is also a distinctive feature of the Greek households (ibid., 2: 226–317).
We can only imagine what all of this meant in terms of the languages used in various con­
texts, but it seems likely that at least some used different languages at home and at work.

The Greek immigrants' diverse geographical backgrounds may be illustrated by the range
of ethnic designations in use during the Ptolemaic period. A recent collection of these
contains more than 250 different ethnic terms, the majority from different areas of the
Greek-speaking world (Laʼda 2002). And while such ethnic denominations do not neces­
sarily indicate a specific individual's actual origin (some (p. 404) were also used in an ex­
tended sense with other meanings; the label “Persian of the epigonê,” for instance, came
to function as a status term for those under obligation in a legal context, Pestman [1990,
91]), their geographical range nevertheless provides a measure of the diverse origins of
immigrants to Ptolemaic Egypt.

Not all immigrant groups, of course, were new. We have already met the Ionians and Cari­
ans, who, according to Herodotus (2.152–154), were brought to Egypt as mercenaries by
Psammetichus I, who settled them in the Delta. Moved to Memphis by pharaoh Amasis in
the sixth century, they remained in that city, where they became known respectively as
Hellenomemphites and Caromemphites and enjoyed their own ethnic quarters (the Hel­
lenion and the Karikon) and representatives (Thompson 1988, 82–88). A record of the ma­
jor dike of the city of Memphis (PSIV 488, from 257 BCE) allows us to locate these differ­
ent ethnic neighborhoods. The Hellenion and the Karikon lay to the northwest, and the
Syro-Persikon quarter was situated southwest of the city. According to Strabo (17.1.32),
who wrote in the late first century BCE, the population of Memphis was a mixed one. The
papyri, supported by excavation and inscriptions, show just how mixed it was both lin­
guistically and culturally.

The Carian grave stelae reused in the fourth century BCE in the building of the sacred an­
imal necropolis on the headland of North Saqqara must have come from this community
at an earlier date (Masson 1978), and it is these texts that allowed the decipherment of
the Carian language (Ray 1981, 1998). Because some of the owners of these gravestones
had Egyptian names, the picture is of a mixed Egyptian-Carian community in the Persian
period, when Carian soldiers served the Persian rulers as mercenaries, and some ran
barges on the Nile (Vittmann 2003, 155–179). The dispersal of their gravestones and the
lack of Carian texts from the Ptolemaic period suggest that, linguistically, Carians had be­
come assimilated by the time of Alexander. Nevertheless, their quarter remained distinct,
as were the privileges of their temple. For when, in the mid-third century BCE, the As­
tarte priests of those known as the Phoenico-Egyptians in Memphis wrote (in Greek) to
Zenon, a man of importance in the area, asking for a contribution of oil for their temple,
they requested the same “as is granted to the temples in Memphis of the Carians and Hel­
lenomemphites …. For the temple of Astarte,” they explained, “is similar to those of the

Page 9 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation
Carians and the Hellenomemphites” (PSI V 531). One detects a competitive element
among these different ethnic communities.

The Ionians of Memphis may be traced in the archaeological and the papyrological record
of both Persian and Ptolemaic periods. Like the Ionians at Naucratis, the only Greek city
in Egypt before Alexandria, the Hellenomemphites possessed their own representatives,
known as timouchoi; they are mentioned in a late third-century-BCE shipping account
(UPZ I 149.16). This connection with shipping already in the fifth century is found in the
long Aramaic customs account from the Persian period (Porten and Yardeni 1993, C3.7,
from 475 bce); together with Carians, some Ionians are labeled “rascals” in one Aramaic
text from the city (Segal 1983, no. 26). Their necropolis lay a little to the north of the city
near Abusir; (p. 405) the text of Timotheus's play, The Persians, came from here (van Min­
nen 1997). The “Curse of Artemisia,” one of the earliest Greek papyrus texts from Egypt,
written in a mixture of Ionic and Doric dialects, also derives from this Hellenomemphite
milieu; the spoken Greek of the city appears to have left its mark on this text, the content
of which has a strong Egyptian flavor (UPZ I 1 = Rowlandson 1998, no. 37, from the late
fourth century BCE). Artemisia, daughter of Amasis, calls on Oserapis (the deified Apis
bull, soon to be adopted in its Hellenized form as the Greek god Sarapis) and the gods
who sit in Poserapis (the Egyptian “House of Osiris-Apis,” transliterated here in Greek) to
curse the father of her daughter, who has failed to provide the proper burial rites for their
child. Some of the Hellenomemphites, it is clear from this text, were already well assimi­
lated into Egyptian culture by the time Alexander's army arrived.

One key difference between the Hellenomemphites and the Caromemphites or the
Phoenico-Egyptians of the city is that, whereas the Hellenomemphites kept their own lan­
guage (Ionic Greek), the other immigrant groups gradually abandoned theirs. During the
Ptolemaic period, the only use of Phoenician known in Memphis is preserved not on pa­
pyrus but on a stone dedication, on top of which was placed a stele of Horus standing on
crocodiles (Vittmann 2003, 76, Abb. 36). The face of the stele is covered in hieroglyphs
with magical spells against snakes, scorpions, and other such dangers; the spells were to
be “read” by anyone who drank the water, which, when poured over them, flowed into a
surrounding rivulet and collected in a basin at the front. The Phoenician military visitor
from Thebes who made this dedication, probably in the second century BCE, appears to
be well aware of the problems encountered by those unfamiliar with foreign scripts. Oth­
er dedications and texts penned by Phoenicians under the Ptolemies were written in
Greek. Like the Jews of Ptolemaic Egypt, Phoenician immigrants were by now generally
Hellenized, at least in the language they used.

Other immigrants to Ptolemaic Egypt are known from the papyri, but of their linguistic
practices little is known. The military group of Idumaeans, who settled as a politeuma in
Memphis in the late second century BCE, had their own whitewashed temple dedicated to
Qos, who in Greek is called Apollo (OGIS II 737). A similar army contingent was quar­
tered at Hermopolis, where, along with other Semitic names, Qos/Apollo figures striking­
ly in the nomenclature of the earliest generation recorded. Within three generations,
however, in nomenclature the Idumaeans of Hermopolis had Hellenized almost entirely. It

Page 10 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation
therefore comes as somewhat of a surprise (and a salutary reminder of how dangerous it
is to draw conclusions from what patchy evidence remains) that some three hundred
years later the singing of hymns in a foreign language (xenikê glôssa, presumably
Edomite) and the unusual sacrifice of sheep (?) and goats are recorded as still in use
among this same Hermopolite community (P.Giss. 99.9–13). Yet, as with other liturgies
(Latin, for example, in parts of the Roman Catholic church or Coptic in the Coptic church
today), the continuation of a language in such a context does not necessarily imply
(p. 406) more widespread familiarity or understanding; this “foreign tongue” may well

represent a “fossil language.” In the case of the Jewish politeuma in neighboring Herak­
leopolis, where Jewish laws and practices are recorded in second-century-BCE texts but
where all recorded dealings are in Greek (P.Polit.Jud., with Thompson 2009), it is equally
possible that Hebrew was used in liturgical contexts. This is not, however, recorded in the
papyri.

Arabs formed a further ethnic group in Ptolemaic Egypt, but in this case, to judge from
the contexts where they are found and their nomenclature, assimilation for Arabs was
with the majority population. Under Ptolemy II, Arabs joined Persians and Hellenes with a
tax privilege (P.Count 30.63, from 254–231 BCE), but their names are consistently Egypt­
ian. One Greek appeal, which possibly involves an Arab, highlights the problems we face
in understanding how these different groups within the mixed population of Ptolemaic
Egypt may have perceived themselves and, in turn, have been perceived by those around
them. A camel driver sent to Syria complains to Zenon, as manager of the dioikêtês'
interests, that Zenon's representative has mistreated him: he has not been properly paid
and has been given only poor-quality wine to drink (P.Col. Zen. II 66, from 256/255 BCE).
The reason the writer gives for this treatment is that he is a barbaros who cannot speak
Greek (hellenizein); his suffering is such, he claims, that he is in danger of starvation.
Self-description as a barbarian represents a fairly extreme adoption of the language and
outlook of the ruling power, though we should not forget the probable intervention of the
scribe who translated the complaint—whatever its original language—into what he con­
sidered suitable expression for such an appeal. Nevertheless, the claim as it stands is a
striking one, implying a strong awareness of discrimination that Zenon was expected to
respond to.

Other linguistic groups recorded from the Ptolemaic period include the Nabataeans, who
in 39/38 BCE dedicated a shrine to their local god on the road between Giza and Memphis
(Huss 2001, 750n8), and a Roman or Italian, Gaius Acutius, who left a record in Latin (but
his name written in Greek) recording his visit to the Isis temple at Philae in 116 BCE
(SEG XXVIII 1485). In OGIS I 129, in contrast, a reinscribed Ptolemaic royal order to pro­
tect a synagogue ends with a Latin summary. It is on the more durable medium of stone
rather than papyrus that these particular records survive. Choice of writing material must
join language choice as a variable in the picture.

As earlier under the Persians, however, the main language of Egypt must have remained
Egyptian. For the fact that this is not immediately apparent in the written record, two fea­
tures are responsible: the primacy of Greek in administrative texts from mummy carton­

Page 11 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation
nage and the better record of publication of Greek texts than of the more difficult texts in
demotic. Nevertheless, both in texts from the Egyptian side of the population—from car­
tonnage, excavation, or family collections of legal texts—and in the preponderance of
Egyptian names found in the papyri, we may still approach the majority language through
the written record.

(p. 407) From Text to Speech and Back Again

To move from the written to the spoken word demands a leap of the imagination. How far
was the vocabulary employed by scribes that of everyday speech? How many of those
with literate skills were familiar with more than one language, and to what extent is it
possible to recapture the degree of multilingualism in a society? The papyri present just a
few hints.

The language of the papyri is affected by many factors, sometimes clear and sometimes
obscure. The peculiarity of Egyptian demotic with its apparent distance from the every­
day vocabulary in the Egyptian villages has already been noted. Other influences are re­
flected in the vocabulary of the administration. In the Persian period, some Iranian and
Semitic technical terms—for “document,” “land measurement,” “lawsuit,” “judge,” and so
on—are found in the Aramaic texts; at the same time, some Egyptian terms—for an Egypt­
ian form of “marriage endowment,” for instance, for “castor oil,” “natron,” and “barley”—
were simply transliterated into Aramaic (Segal 1983, 10–11). One key term introduced in
this period—“artaba,” the capacity measure used for cereals—passed into regular usage
and was taken over by the Greeks. Though it is always interesting to compare the con­
texts and different linguistic influences over time, how widely these terms were adopted
into everyday speech is generally unknown.

In Greek, the formal, often standardized, language of the bureaucracy appears to be the
product of scribal training, just as the occasional introduction of literary vocabulary into
documentary texts seems likely to reflect the literary education of Greek schools. Again,
the use of specialized vocabulary sometimes represents the adoption of Egyptian institu­
tions in transliterated form (pheritob, for instance, as the title of a temple official) or the
translation of specialist terms into Greek, like those for land at various stages of irriga­
tion: “waterlogged” (embrochos), “not reached by the water” (abrochos), hypologos
(“unproductive”), and so forth. This is a phenomenon also found in the broader context;
transliteration and translation are alternatives in the process of language contact. The
choice of script is a further variable (as noted earlier). From the mid-Ptolemaic period
(201/200 BCE), one of the rebel kings of Thebes (Porô Yr Gonafor = pharaoh
Haronnophris) is recorded in an Egyptian graffito at Abydos that is actually written in
Greek script (P.Recueil 11; cf. P.Recueil 12).

The language of the texts themselves can sometimes show linguistic features that may be
transferred from text to speech. Dialect is one such aspect. The impact of Doric dialect
forms has been observed, especially in the nomenclature of the elite families of Alexan­

Page 12 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation
dria, as representing Macedonian influence in the highest social circles of the Ptolemaic
court and in the capital (Clarysse 1998).

Bilingualism is a further subject that one may approach through the texts. Both bilingual
texts and bilingual archives survive; their survival, however, is easier to note than to in­
terpret. Bilingual texts have implications for both writer and reader, (p. 408) and in the
case of administrative texts, the bilinguality of the early generations of scribes, now re­
tooled in Greek as the new administrative language, may on occasion be traced through
the technology of writing itself. Whereas Greek was normally written with a sharpened
reed (kalamos), for Egyptian demotic a frayed rush was used (Tait 1988) and different ink.
Greek written with a rush thus implies a biliterate but lazy scribe, who has not bothered
to switch the implement with which he is writing. This occurs quite frequently in the third
century BCE (Clarysse 1993, 186–195). At the same time, the Greek of such texts some­
times shows the influence of Egyptian vocabulary and syntax (Clarysse 1990; 1993, 197–
200).

The same phenomenon suggested by this practice is reinforced by the language of the
texts themselves, though whose language choice was involved remains a question. Some
sets of administrative data switch language within the same operation, at least at the low­
er levels. The salt-tax register of P.Count 2 + 3 (229 BCE), for instance, written on one pa­
pyrus, switches from a village register compiled person by person in demotic to a Greek
tax-district record recorded village by village, and back again to summary tax-area totals,
now registered in demotic. It was only at nome level and above that Greek became the
rule, and many scribes, it seems, moved with ease between the two main languages of the
time (Clarysse and Thompson 2006, 2: 6–7, 70). Within the Ptolemaic bureaucracy, espe­
cially in the third century BCE, bilinguality was an everyday occurrence. Both the lan­
guages of the texts produced and the technology used to write them imply a bilingual
flexibility in speech, as well as in writing.

A similar mix of languages appears in receipts that the tax collectors issued, sometimes in
Greek, sometimes in demotic, and sometimes in both. We may speculate whether the
taxpayer's language played any part in the scribe's choice of language. When the
taxpayer's name appears written in larger and clearer demotic, an awareness of the
receipt's purpose may explain the particular language choice. The verbal interchanges
preceding such transactions can only be imagined. More often language choice may re­
flect the prime identity of the scribes, or the demands of the office they worked in. Thus,
the increased use of Greek for tax receipts in the Thebaid starting in 165 BCE has been
interpreted as resulting from the reestablishment of central control on the south, when
Egyptian officials were downgraded following the great revolt some years before (Vandor­
pe 2000, 177). By the second half of the second century BCE bilingual offices were per­
haps less common than before; the growing use of the rulers' language was a sign of
changing times.

Bilingual archives (archives, that is, in the broader papyrological sense; see chapter 10)
have already been mentioned. Taking many forms, in terms of language use they indicate

Page 13 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation
those areas of society in which bilingualism was present at least to a limited degree. So,
for instance, in the Zenon archive in both official and private dealings, texts in different
languages reflect the language choice of different correspondents. With a handful of de­
motic texts (P.Zen.dem.) and a few bilingual receipts (P.Zen.Pestm. 1–13) among the over­
whelming majority of documents in (p. 409) Greek, the composition of this archive repre­
sents the language environment of Zenon and his circle (Orrieux 1983, 146–150).

In family archives made up of demotic legal contracts, another process is apparent. Over
time Greek-language use began to affect even a predominantly Egyptian practice. In part
this followed a government initiative. From 145 BCE on, in order to have legal validity, de­
motic contracts had to be registered in Greek at a registry office (see chapter 23). From
this date on, Greek dockets are found appended to demotic contracts, which in translated
form might then be used in a Greek court of law. Among the mortuary priests of Hawara
or of Memphis and Thebes, texts written in both Greek and Egyptian are thus found in
later family archives (Thompson 1988, 186–189). Demotic contracts with Greek sum­
maries of their contents represent a limited form of bilingual text. They tell us little of the
owner's bilingualism—perhaps more of the scribe's—but the gradual move to Greek legal
practices seems likely to mirror a more general, ongoing process at least at the institu­
tional level, with concomitant bilingual activities. Thus, by the time the Romans came,
any Egyptian legal forms that survived (for marriage contracts, for instance) were gener­
ally expressed in Greek.

Some Mixed Contexts

One way in which to investigate the multilingual environment of Persian and Ptolemaic
Egypt is to highlight some of the areas where the different peoples who lived together in
Egypt came into contact with one another. There are sufficient similarities between our
two main periods for us to consider them in tandem. The following case studies are very
selective in coverage and made simply exempli gratia, but they involve mixed-language
environments, bilingual archives, and several typical situations of the time.

The guardposts of Persian Egypt at Syene, Elephantine, and elsewhere have already been
introduced. From among the Jewish and Aramaean military settlers of the south, a num­
ber of family archives illustrate a variety of interethnic contacts on an everyday basis.
Particular individuals stand out. Mibtahiah is one (Pap.Eleph.Eng. B23–33, mid-fifth cen­
tury BCE). Daughter of Mahseiah, son of Jedaniah, Mibtahiah was first married to Jezani­
ah, son of Uriah, who served together with her father. The couple was well endowed with
a house and the parcel of land it was built on. Jezaniah may have predeceased his wife
since Mibtahiah was later married to Esḥor, son of Djeḥo, a royal builder by whom she
had two children. From later contracts drawn up after the death of her second husband,
it is known that Esḥor was also known by the Jewish name Nathan (Pap.Eleph.Eng.
B32.8–9, cf. p. 196n6, from 416 BCE; B33.2, from 410 BCE). In this Aramaic-speaking
context, where in the (p. 410) Elephantine fortress families lived close to one another and
where the local governor and troop commander, both Persians, might oversee their con­

Page 14 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation
tracts (B31.4–5), an Egyptian incoming husband had joined the Jewish community and
changed his name. In dealings with another Egyptian builder, Peu, son of Paḥe, Mibtahiah
might meanwhile swear by Sati, the local Egyptian goddess (Pap.Eleph.Eng. B30.4–5,
from 440 BCE). Ethnic (like language) lines were forever being crossed, and identities
changed.

A further case of mixed marriage in the same community is documented in the contracts
of Ananiah, son of Azariah, who married Tamut, the Egyptian slave of his neighbor
Meshullam, son of Zaccur. Meshullam is described as an Aramaean from Syene and a
member of the detachment of (the Persian) Varyazata, whereas the groom, who pays a
bride price (or mohar) to Meshullam, is a servant of Yahweh in the Elephantine fortress.
At the time of their marriage contract (Pap.Eleph.Eng. B36, from 449 BCE), Tamut al­
ready had a son, Palṭî, who came along withher; like his mother, however, he remained
Meshullam's property. Ananiah and Tamut then had a daughter, Jehoishma. About twenty
years later Tamut's owner, Meshullam, drew up a contract offering manumission to her
and her daughter on his death—on the condition that they continue to serve him during
his lifetime and then that of his son, Zaccur (Pap.Eleph.Eng. B39, from 427 BCE); this, in­
deed, they did. Finally, when the daughter, Jehoishma, married, Meshullam's son, Zaccur,
provided her dowry, and in that contract she is called his “sister” (B41, from 420 BCE).

The cultural mix and unusual human situations of these texts are hard to match. The
Egyptian-named and probably Egyptian slave girl Tamut, “she of (the goddess) Mut,”
formed part of a social institution—slavery—that was not native to Egypt but came with
the settlers. Despite her married status, until she acquired her conditional freedom,
Tamut remained the property of her original owner, as did her daughter and son. We do
not know at what age she was enslaved or whether indeed she was born a slave, but in
her family life Egyptian is unlikely to have been her first language. It is striking, if not
surprising, that her children were given Semitic names. On the other hand, the world of
the texts preserves just one side of the experience, and this may sometimes be deceptive.

As with Aramaic in the country's guardposts under the Persians, so later under the
Ptolemies the army provided a context in which Macedonians and Greeks predominated
and the Greek language flourished. Troops, however, need wives, and, as we have seen,
local marriages often followed. The army thus provided a standard context for contact
with the majority Egyptian population.

Not all contact was so peaceable, however. The billeting of troops on a local population is
likely to cause problems. However much the authorities attempted to control the process
—and try they did (e.g., C.Ord.Ptol. 1, 5–10, 24, all under Ptolemy II)—disruption and en­
suing difficulties were probably inevitable (Lewis 1986, 21–24). One bilingual group of
texts (in Greek and demotic) from the second century BCE tells the tale of a certain Sen­
nesis, daughter of Patepnebteus, who was (p. 411) first married to an Egyptian named
Petosiris and later lived with her son, Esoroeris, and his wife. When a Cyrenean soldier,
Neoptolemos, son of Neoptolemos, was billeted on the family, trouble ensued. Sennesis
took up with Neoptolemos and later married him. In subsequent legal proceedings Es­

Page 15 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation
oroeris claimed he had been beaten up and driven from his home by Neoptolemos, who
lived upstairs. A counterclaim was filed, but eventually Sennesis, now known by her
Greek name of Isias, together with her husband, Neoptolemos, came to a settlement with
her son (Pap.Eleph.Eng. C33, from 198 BCE; D8–10, from 137/136 BCE). The language
mix of these texts once again seems likely to reflect that of this sector of society, where
the wife of a Greek army man had something of an advantage.

A less agreeable outcome resulted from another case, at an earlier date, when a Greek
soldier intruded on an Egyptian marriage. Among a collection of texts from the Memphite
Serapeum, there survives a petition presented to the king and the queen by the Egyptian
twin sisters Taous and Thaues, daughters of Nephoris and Argynoutis (?), who played the
parts of the goddesses Isis and Nephthys in the mourning ceremonies for the Apis bull
that died in 164 BCE. Their complaints are directed against their mother, who had left
their father for Philippos, son of Sogenes, a Greek soldier who was serving in the area.
Philippos—they claim—had plotted the death of their father, who had escaped Philippos's
dagger only by plunging into the Nile and swimming to a nearby island, where he took a
boat to Herakleopolis. There he died of grief and, at the time of writing, still lay unburied.
Nephoris and her son, Panchrates, are now accused of stealing the oil allowance that be­
longed to the twins. Taous and Thaues enlisted a friend of their father who was resident
in the Serapeum, Ptolemaios, son of Glaukias, to write for them in Greek (UPZ I 18). Bilin­
gualism runs through the story—Greek soldier, Egyptian wife, her two Egyptian daugh­
ters who impersonate the twin goddesses in a key religious ceremony—but when it came
to writing complaints, then a Greek like Ptolemaios was required. Personal scenarios like
these, with liaisons crossing the ethnic divide, are of the essence of papyrological docu­
mentation.

A third example of the mixed world of the military appears in the archive of the Cretan
cavalry officer Dryton, of his second wife, Apollonia, also known as Senmonthis, and
oftheir daughter, Apollonia, also known as Senmouthis, together with her husband, Kaies
(P.Dryton). As suggested by the women's double names, Dryton, a citizen of Ptolemais in
Upper Egypt, and his family moved in a very mixed world. The archive itself comes from
Pathyris, a garrison town loyal to the crown, which was entirely destroyed during the
great revolt in Upper Egypt in 88 BCE. Almost equal numbers of Greek and demotic texts
survive in this archive (29 Greek and 25 demotic), as well as five bilingual texts. Dryton
came from a Cretan family; the validity of his ethnic designation, which could have been
that of his army unit, is confirmed by the particularly Cretan names of Dryton himself and
of his son (by a previous wife), called Esthladas. Apollonia, his second wife, is labeled a
Cyrenean, but her strongly Egyptianized family would appear to have been in (p. 412)
Egypt for some generations. In her many business dealings, she—or perhaps her scribe—
tended to use Egyptian (Vandorpe 2002), and all of her five daughters (like their mother)
had double names: Apollonia-Senmouthis, Aristo-Senmonthis, Nikarion-Thermouthis,
Apollonia the younger-Senapathis and Aphrodisia-Tachratis. The practice continued into
the next generation, which, to judge from those whose documents survive, increasingly
moved over to the Egyptian side. Apollonia-Senmouthis married an Egyptian soldier,
Kaies, and none of their three daughters has a Greek name recorded. The second half of
Page 16 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation
the archive contains more demotic than Greek texts. Through the study of family
archives, like that of Dryton's descendants or those of other military families (Lewis 1986,
124–152), we may enter the mixed military and civilian environments of the second half of
the Ptolemaic period.

A second area where we may tease out the input of different ethnic groups and their mul­
tilingual activities is along the Nile, the country's main artery, which linked Upper and
Lower Egypt in both geographical and economic terms. Nile barges carried the agricul­
tural wealth of the south to the markets of the capital cities—to Memphis and later
Alexandria. On the whole, as a big business that involved capital investment and large re­
turns, shipping tended to be under immigrant control. Persians and later Greeks, espe­
cially upper-class Alexandrians, and even Ptolemaic queens owned the great Nile barges
that carried the grain, while these were made, mended, and sailed by local Egyptians
(Porten et al. 1996, 15, 77; Thompson 1983). Carians, too, were involved in shipping
(Pap.Eleph.Eng. 11.3, from 411 BCE; Segal 1983, no. 26, together with Ionians), as in­
deed were other immigrants. In mid-first-century-BCE Memphis, for instance, the ship­
ping company of the nauklêroi Hippodromitai appears to have been the preserve of
Phoenicians (Thompson 1988, 60–1). Nowhere is the immigrant presence clearer than in
the long Aramaic customs account of the fifth century BCE, where the ships carrying
wine, oil, empty jars, hardwood (for building), and other products for the king's store­
house are regularly specified as those of Greek (“Ionian”) shipowners (Porten and Yardeni
1993, C3.7). When, under the Ptolemies, patterns of landholding are documented from
different areas of the country, it becomes clear that the Greek settlers (and especially the
soldiers who were settled with plots of cleruchic land by way of a retainer or pension) en­
joyed the larger estates, where they cultivated Egypt's cash crops and reared their flocks
of sheep. Agriculture was by far the most important sphere of production, and, in this po­
tentially profitable area, social standing was matched by economic power. So, too, in ship­
ping and other productive spheres, the immigrants' economic strength is documented in
the papyri. Aramaic, Greek, Carian, and Phoenician all joined Egyptian as languages in
use along the Nile during these five hundred years of foreign rule.

Page 17 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation

Figure 17.1. Memphite Serapeum papyrus (now

in the Louvre) with Greek and Macedonian
months written in Apollonios' hand, plus a de­
motic contract.

Examples of similar mixed contexts could well be multiplied. Religion is another obvious
area (see chapter 24). Here one brief example may suffice. Life in the Memphite Ser­
apeum in the mid-Ptolemaic period is, as already noted, (p. 413) (p. 414) known from the
papyri of the two sons of Glaukias. The peculiar position of Ptolemaios and briefly of his
younger brother Apollonios as detainees (katochoi) in the temple enclosure of Sarapis is
certainly of interest. Of more relevance here, however, the mix of Greek and demotic in
the archive, as in the lives of the brothers and their circle, takes the reader into a com­
plex and fluid environment between two worlds where, in the temple's protected environ­
ment, Greeks and Egyptians lived their lives together and interacted. Ptolemaios copied
down Greek literature and wrote quite passable Greek; they also owned some demotic
wisdom literature (see figure 17.1). Apollonios, who often ran his brother's errands and
led a troubled life, dreamed bilingually in Greek and Egyptian (at least his dreams are
recorded in the two languages). Both brothers, it is clear, were bilingual, as indeed was
the multicultural world in which they lived (Thompson 1988, 212–265).

Up above the city of Memphis on the desert headland of North Saqqara, the Apis bull—
Sarapis now for the Greeks—had a nonexclusive following among the different peoples of
Ptolemaic Egypt. Here, in the Serapeum, Egyptian might be transcribed into Greek, as in
the account of Nektembes' dream (UPZ I 79.4–5, from 159 BCE). Alternatively, Egyptian
sentiments and prayers were translated into Greek, as when Petesis, son of Chonouphis,
mummifier in chief of the Apis and Mnevis bulls, appealed to the king and queen for pro­
tection for his house, which was constantly under attack. He asked for the protection of a
royal order, written (presumably in ink) on a whitened board outside the house in both
Greek and native letters (UPZ 107.30, from 99 BCE). And before he made his request, he
called on the gods whom he served to grant the royal couple “health, victory, power,
strength, and lordship over the lands below heaven” (UPZ I 106.13–14, from 99 BCE). The
wording and the sentiment come straight from the world of Egypt, here translated into
Greek. Eventually his request was granted, but despite the bilingualism of the royal or­
der, Petesis gained no respite from his attackers, and, finally, he resorted to legal means.

Page 18 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation
In this interplay of language, culture, law, and religion, the complex nature of life for yet
another sector of Ptolemaic society comes to life through the papyri.

Bibliography
Adams, J. N., M. Janse, and S. Swain, eds. 2002. Bilingualism in Ancient Society: Lan­
guage Contact and the Written Text. Oxford: Oxford University Press.

Adams, J. N., and S. Swain. 2002. “Introduction.” In Bilingualism in Ancient Society: Lan­
guage Contact and the Written Text, ed. J. N. Adams, M. Janse, and S. Swain, 1–20. Ox­
ford: Oxford University Press.

Aimé-Giron, N. 1931. Textes araméens dʼÉgypte. Cairo: Service des Antiqués de lʼÉgypte.

Bowman, A. K., and G. Woolf, eds. 1994. Literacy and Power in the Ancient World. Cam­
bridge: Cambridge University Press.

(p. 415) Bresciani, E., and M. Kamil. 1966. “Le lettere aramaiche di Hermopoli.” In Atti
della Accademia Nazionale dei Lincei. Memorie della classe di scienze morali, storiche e
filologiche 8 ser. 12.5: 357–428.

Briant, P. 2002. From Cyrus to Alexander: A History of the Persian Empire. Trans. P. T.
Daniels. Winona Lake, Ind.: Eisenbrauns.

Christensen, T. 2002. “The Edfu Nome Surveyed: P.Haun. inv. 407 (119–118 BCE).” Diss.
University of Cambridge.

Clarysse, W. 1985. “Greeks and Egyptians in the Ptolemaic Army and Administration.” Ae­
gyptus 65: 57–66.

———. 1987. “Greek Loan-words in Demotic.” In Aspects of Demotic Lexicography: Acts

of the Second International Conference for Demotic Studies, Leiden, 19–21 September
1984, ed. S. P. Vleeming, 9–33. Studia Demotica 1. Leuven: Peeters.

———. 1990. “An Epistolary Formula.” CdE 65: 103–106.

———. 1992. “Some Greeks in Egypt.” In Life in a Multi-cultural Society: Egypt from Cam­
byses to Constantine and Beyond, ed. J. H. Johnson, 51–56. Studies in Ancient Oriental
Civilization 51. Chicago: Oriental Institute.

———. 1993. “Egyptian Scribes Writing Greek.” CdÉ 68: 186–201.

———. 1995. “Greeks in Ptolemaic Thebes.” In Hundred-gated Thebes: Acts of a Colloqui­

um on Thebes and the Theban Area in the Graeco-Roman Period, ed. S. P. Vleeming, 1–19.
Papyrologica Lugduno-Batava 27. Leiden: Brill.

———. 1998. “Ethnic Diversity and Dialect among the Greeks of Hellenistic Egypt.” In The
Two Faces of Graeco-Roman Egypt: Greek and Demotic and Greek-Demotic Texts and

Page 19 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation
Studies Presented to P. W. Pestman, ed. A. M. F. W. Verhoogt and S. P. Vleeming, 1–13. Pa­
pyrologica Lugduno-Batava 30. Leiden: Brill.

———. 2002. “A Jewish Family in Ptolemaic Thebes.” Journal of Juristic Papyrology 32: 7–
9.

Clarysse W., and D. J. Thompson. 2006. Counting the People in Hellenistic Egypt, 2 vols.
Cambridge: Cambridge University Press.

Collombert, Ph. 2000. “Religion égyptienne et culture grecque: Lʼexemple de

Aιοσκουρίδης” CdÉ 75: 47–63.

Coulon, L. 2002. “Quand Amon parle à Platon (la statuaire Caire JE 38033).” Revue
dʼÉgyptologie 52: 85–112.

Cowley, A. 1923. Aramaic Papyri of the Fifth Century B.C. Oxford: Clarendon Press.

Crawford, D. J. 1971. Kerkeosiris: An Egyptian Village in the Ptolemaic Period. Cambridge

Classical Studies. Cambridge: Cambridge University Press.

Driver, G. R. 1957. Aramaic Documents. Oxford: Clarendon Press.

Fewster, P. 2002. “Bilingualism in Roman Egypt.” In Bilingualism in Ancient Society: Lan­

guage Contact and the Written Text, ed. J. N. Adams, M. Janse, and S. Swain, 220–245.
Oxford: Oxford University Press.

Goudriaan, K. 1988. Ethnicity in Ptolemaic Egypt. Dutch Monographs on Ancient History

and Archaeology 5. Amsterdam: Gieben.

Grelot, P. 1972. Documents araméens dʼÉgypte. Paris: Éditions du Cerf.

Huss, W. 2001. Ägypten in Hellenistischer Zeit 332–30 v. Chr. Munich: Beck.

Johnson, J. H., ed. 1992. Life in a Multi-cultural Society: Egypt from Cambyses to Constan­
tine and Beyond. Studies in Ancient Oriental Civilization 51. Chicago: Oriental Institute.

Kornfeld, W. 1973. “Jüdisch-aramäische Grabinschriften aus Edfu.” Anzeiger der Oester­

reichischen Akademie der Wissenschaften in Wien, Philos.-Hist.Klasse 110: 123–137.

(p. 416) Kraeling, E. G. 1953. The Brooklyn Museum Aramaic Papyri: New Documents of
the Fifth Century B.C. from the Jewish Colony at Elephantine. New Haven, Conn.: Yale
University Press.

Laʼda, C. A. 2002. Foreign Ethnics in Hellenistic Egypt. Prosopographia Ptolemaica 10.

Leuven: Peeters.

Lauer, J.-Ph. 1976. Saqqara the Royal Cemetery of Memphis. Excavations and Discoveries
since 1850. London: Thames and Hudson.

Page 20 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation
Lewis, N. 1986. Greeks in Ptolemaic Egypt: Case Studies in the Social History of the Hel­
lenistic World. Oxford: Clarendon Press.

Lloyd, A. B. 1982. “The Inscription of Udja?orresnet: A Collaborator's Testament.” JEA 68:

166–180.

Masson, O. 1978. Carian Inscriptions from North Saqqâra and Buhen. Texts from Excava­
tions, Memoir 5. London: Egypt Exploration Society.

Matthews, R., and C. Roemer, eds. 2003. Ancient Perspectives on Ancient Egypt. Encoun­
ters with Ancient Egypt. London: UCL Press, Institute of Archaeology.

Melaerts, H., and L. Mooren, eds. 2002. Le rôle et le statut de lafemme en Égypte hel­
lénistique, romaine, et byzantine: Actes du colloque international, Bruxelles–Leuven 27–
29 novembre 1997. Leuven: Peeters.

Orrieux, Cl. 1983. Les papyrus de Zenon. Lʼhorizon dʼun grec en Égypte au Ille siècle
avant J.C. Paris: Macula.

Pestman, P. W. 1990. The New Papyrological Primer. Leiden: Brill.

Porten, B. 1968. Archives from Elephantine: The Life of an Ancient Jewish Military
Colony. Berkeley: University of California Press.

Porten, B., J. J. Farber, C. J. Martin, G. Vittmann, L. S. B. MacCoull, S. Clackson, S. Hop­

kins, and R. KatzoV. 1996. The Elephantine Papyri in English: Three Millennia of Cross-
cultural Continuity and Change. Leiden: Brill.

Porten, B., and A. Yardeni. 1986. Textbook of Aramaic Documents from Ancient Egypt.
Vol. 1, Letters. Winona Lake, Ind.: Eisenbrauns.

———. 1989. Textbook of Aramaic Documents from Ancient Egypt. Vol. 2. Contracts.
Winona Lake, Ind.: Eisenbrauns.

———. 1993. Textbook of Aramaic Documents from Ancient Egypt. Vol. 3, Literature, Ac­
counts, Lists. Winona Lake, Ind.: Eisenbrauns.

———. 1999. Textbook of Aramaic Documents from Ancient Egypt. Vol. 4, Ostraca. Winona
Lake, Ind.: Eisenbrauns.

Porten, B., and A. Yardeni, 2004/5. Two Aramaic Salt-tax Receipts by the Scribe Joseph..
Enchoria: 29.

Ray, J. D. 1981. “An Approach to the Carian Script.” Kadmos 20: 150–162.

———. 1994. “Literacy and Language in Egypt in the Late and Persian Periods.” In Litera­
cy and Power in the Ancient World, ed. A. K. Bowman and G. Woolf, 51–66. Cambridge:
Cambridge University Press.

Page 21 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation
———. 1998. “Aegypto-Carica.” Kadmos 37: 125–136.

Rowlandson, J., ed. 1998. Women and Society in Greek and Roman Egypt: A Sourcebook.
Cambridge: Cambridge University Press.

Rutherford, I. 2003. “Pilgrimage in Greco-Roman Egypt: New Perspectives on Graffiti

from the Memnonion at Abydos.” In Ancient Perspectives on Egypt, ed. R. Matthews and
C. Roemer, 171–189. Encounters with Ancient Egypt. London: UCL Press, Institute of Ar­
chaeology.

Segal, J. B. 1983. With contributions by H. S. Smith. Aramaic Texts from North

(p. 417)

Saqqâra with Some Fragments in Phoenician. Texts from Excavations, Memoir 6. London:
Egypt Exploration Society.

Tait, W. J. 1988. “Rush and Reed: The Pens of Egyptian and Greek Scribes.” Proceedings
of the XVIII International Congress of Papyrology, ed. B. G. Mandilaras, vol. 2, 477–481.
Athens: Greek Papyrological Society.

Thompson, D. J. (Crawford). 1983. “Nile Grain Transport under the Ptolemies.” In Trade
in the Ancient Economy, ed. P. Garnsey, K. Hopkins, and C. R. Whittaker, 64–75. Berkeley:
University of California Press.

Thompson, D. J. 1988. Memphis under the Ptolemies. Princeton: Princeton University

Press.

———. 2001. “Hellenistic Hellenes: The Case of Ptolemaic Egypt.” In Ancient Perceptions
of Greek Ethnicity, ed. I. Malkin, 301–322. Cambridge, Mass.: Harvard University Press.

———. 2009. “Ethnic Minorities in Hellenistic Egypt.” In The Political Culture of the
Greek City after the Classical Age, ed. R. Alston and O. M. van Nijf, Groningen-Royal Hol­
loway Studies on the Greek City after the Classical Age, vol. 2. Leuven: Peeters.

Van Minnen, P. 1997. “The Performance and Readership of the Persai of Timotheus.” APF
43: 246–260.

Vandorpe, K. 2000. “The Ptolemaic Epigraphe or Harvest Tax (shemu).” APF 46: 169–232.

———. 2002. “Apollonia, a Businesswoman in a Multicultural Society (Pathyris, 2nd-1st

Centuries B.C.).” In Le rôle et le statut de lafemme en Égypte hellénistique, romaine, et
byzantine: Actes du colloque international, Bruxelles–Leuven 27–29 novembre 1997, ed.
H. Melaerts and L. Mooren, 325–336. Leuven: Peeters.

Vittmann, G. 2003. Ägypten und die Fremden im ersten vorchristlichen Jahrtausend.

Mainz am Rhein: Verlag Ph. von Zabern.

Vleeming, S. P., ed. 1995. Hundred-gated Thebes: Acts of a Colloquium on Thebes and the
Theban Area in the Graeco-Roman Period. Papyrologica Lugduno-Batava 27. Leiden: Brill.

Page 22 of 23
The Multilingual Environment of Persian and Ptolemaic Egypt: Egyptian,
Aramaic, and Greek Documentation
———, and J. W. Wesselius. 1985. Studies in Papyrus Amherst 63: Essays on the Aramaic
Texts in Aramaic/demotic. Amsterdam: Juda Palache Instituut.

Dorothy J. Thompson

Dorothy J. Thompson is Former Newton Trust Lecturer and affiliated lecturer in an­
cient history, and Life Fellow of Girton College, University of Cambridge.

Page 23 of 23
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation

The Multilingual Environment of Late Antique Egypt:

Greek, Latin, Coptic, and Persian Documentation
Jean-Luc Fournet
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0018

Abstract and Keywords

This article considers three languages—Coptic, Latin, and Pehlevi—all of which were
widely spoken and written in Egypt in the fourth to seventh centuries, analyzing their use
and interaction with Greek, which remained the official language and is by far the most
abundantly documented. Each of these languages poses in a distinctive way the problem
of multilingualism or of multiliteracy and presents a nuanced picture, ranging from a
nearly total and deliberate absence of bilingualism to a deep bilingualism (where the rela­
tionship between the languages tends to reverse itself), passing by way of diglossia.

Keywords: Coptic language, Latin language, Pehlevi language, Egypt, multilingualism, multiliteracy, deep bilin­
gualism

Did the multilingual environment of late antique Egypt change fundamentally compared
to earlier periods? The linguistic situation of the fourth to seventh centuries does not at
first seem much different from that of the past: As in the Ptolemaic period and under Ro­
man rule, we find the same coexistence of an Egyptian substrate and a Greek-speaking
population, plus, since the Roman conquest, a Latin-speaking element with a fairly mar­
ginal place But over the centuries the relationships between these three languages
evolved, and, leaving to (p. 419) one side some isolated cases (Armenian and Syriac), late
antiquity witnessed two important changes: (1) the Sassanid occupation, which produced
documentation in Pehlevi, and (2) the emergence of a new script intended to record the
contemporary vernacular language, Coptic. I limit myself here to these three languages—
Coptic, Latin, and Pehlevi, all of which were widely spoken and written in Egypt in the
fourth to seventh centuries, analyzing their use and interaction with Greek, which re­
mained the official language and is by far the most abundantly documented. Each of
these languages poses in a distinctive way the problem of multilingualism or—for our doc­
umentation is only written—of multiliteracy and presents a nuanced picture, ranging from
a nearly total and deliberate absence of bilingual-ism to a deep bilingualism (where the
relationship between the languages tends to reverse itself), passing by way of diglossia.1

Page 1 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation

Pehlevi: Languages in Contact without Interac­

tion
One language makes a sudden appearance in Egypt's papyrological documentation:
Pehlevi, or Middle Persian. The Pehlevi texts of Egypt are limited to the decade of the
Sassanian conquest (619–629),2 begun by Xusro (Chosroes) II Parwēz. They total nearly
950—a remarkably large number for such a short time-span.They are written on papyrus,
parchment, skin, and linen and come mainly from the Fayyum (figure 18.1). Except for un­
published papyri in Vienna, which vanished after World War II and have only recently
been rediscovered, these have been published or republished by Dieter Weber in two vol­
umes of the Corpus Inscriptionum Iranicarum (CII).

The Pehlevi documentation is actually rather disappointing. The texts, almost all fragmen­
tary, are very hard to read because of the extremely cursive and stylized script. The com­
mentary in the editions is also excessively focused on philological matters, making it diffi­
cult to draw out its full historical value (Huyse 1995).

All the Pehlevi texts are documentary and secular, above all of a military character (or­
ders, lists of provisions for soldiers, itineraries). Some are connected to commerce, but
even there a link with the armyexists. Private documents are very rare: a number of pri­
vate letters (e.g., CII, P. 18, a letter from a Persian to his sister, showing that the Persian
population in Egypt was not entirely male but that some families had followed the sol­
diers) and three sale contracts (Weber 1992a, 501–502).

Figure 18.1. An example of a complete Pehlevi docu­

ment: an order for delivery (CII, P. 69 = P.Heid. Pahl.
9). Photo courtesy of the Institut für Papyrologie,
University of Heidelberg.

It is difficult to draw conclusions from these documents about the impact of the Persian
conquest on the inhabitants of Egypt and the linguistic relationship between conquerors
and conquered. The same is true of the few contemporary (p. 420) Greek and Coptic texts
Page 2 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
(Butler and Fraser 1978, xlvi–xlix; MacCoull 1986; Altheim-Stiehl 1992a; Huyse 1995,
358). The impermeable character of each body of evidence is striking. The Greek and
Coptic texts rarely refer to the Persians and do not characterize the nature of relations
between them and the Greek-Egyptian population. The Pehlevi texts are focused on mili­
tary problems (notably on supply), and the local population is nearly absent from them.
One good indication is that no Greek, Coptic, or even Latin personal name appears in
them, only Persian names, except for Abraam (CII, P. 136) and Samuel, a Jewish mer­
chant, perhaps from Elephantine (Weber 1992b, 341; CII, P. 137). Egyptian place names
are far more visible (CII, P. 55, 148), but for purely military reasons. In short, there are
few connections between the two bodies of documents, with the notable exception of the
case of Šahrālānyōzān, a high official who appears in some Greek papyri in the form
Σαραλανεοζαν (CII, P. 81; Weber 1991; Sänger 2008).

The papyrological sources give no sign of any bilingualism, however limited, in the popu­
lation of Egypt. The total absence of bilingual documents is noteworthy,3 as is that of
Pehlevi texts written by Graeco-Egyptians. Certainly there were oral or (p. 421) written
contacts. For instance, in P.MoscowCopt. 37–38 the reader Menas says, “I asked the Per­
sians to …”; P.Oxy. LI 3637 says that the recipient had “also had written instructions
about this matter from our master the all-praiseworthy Saralaneozan”; CPR IV 48 is a
contract for the delivery of linen between the inhabitants of Pousire (Hermopolite nome)
and “their master,” Perês Kôsrôi (Pērōz-Xusrō), in which the only instance of an oath “by
the safety of the King of Kings” occurs. All these texts show very formal contacts that
were carried out through interpreters on the Persian side or by means of networks or in­
stitutions already in place and used by the Persians. If any effort at linguistic adaptation
was made, it was on the part of the Persians that we must look for it. Thus, the Persians
Rasbanas and Reme wrote letters in Greek to the scholasticus Marinus (P. Oxy. LI 3637,
XVI 1862–1863; cf. Foss 2002, 170–171).

A few rare points of interference between the two languages, which show that these ten
years of conquest, despite everything, did leave various linguistic traces, may nonetheless
be noted: The name of the Persian office sālār (“official”) passed into Greek in the form
σελλάριος and appears in several Greek and Coptic documents of this period, particularly
in the archive of Theopemptos and Zacharias (Foss 2002). In the reverse direction, the
Persians borrowed from Egyptian Greek the words l ṭlʼ (λίτρα, “pound”) and lagānag
(λάγυνος, a liquid measure).4 These points of contact are, however, limited to borrowings
of an institutional nature (title, measures), which can occur even with the most superfi­
cial contact.

Several reasons account for this superficiality: indifference, even rejection of the Persians
on the part of the Graeco-Egyptians, who did not wish to be friendly with the invader, and
most of all the impression that this conquest changed the internal situation in Egypt very
little not only because of its short duration but above all because the Persians limited
themselves to a military presence and relied for the rest on existing institutions. It is true

Page 3 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
that our knowledge of this period is still imperfect and would beneit from a fuller study
bringing together the Greek, Coptic, and Pehlevi documentation.

Latin: Diglossia with Limited or Imperfect

Bilingualism
Latin in Egypt and Imperial Policy

Figure 18.2. Literary papyri in Latin.

The situation of Latin in late antique Egypt is paradoxical: Of around 565 papyrological
documents partly or entirely in Latin, those of the fourth to seventh century amount to a
mere 140.5 To put it another way, the Latin documentation (p. 422) of this period is three
times less abundant than of the principate (first—third centuries). And yet it is a well-
known fact that the reforms of Diocletian and the policies of Constantine brought, along
with Romanization at the institutional level, a Latinization of the administration of the
eastern provinces, to which Egypt belonged (e.g., Rochette 1997, 116–126). The reforms
of Diocletian, by imposing Latin as an official administrative language, undeniably had an
effect: Of the 140 papyri in question, about 90 date precisely to the fourth century. The
Alexandrian Claudian, as comfortable in Greek as in Latin and official poet at the western
court, is one of the best examples of the success of this policy. But the efficacy of this La­
tinization has long been questioned (Turner 1961; Adams 2003, 635–637, 758). From the
end of the fourth century, moreover, imperial policy tended to reaffirm the position of
Greek: In 397, judicial decisions were allowed to be rendered in Greek, as well as in
Latin; in 439, Theodosius II allowed anyone to make a will in Greek; the prefect Cyrus
(439–441) abolished the official use of Latin in the pretorian prefecture of the east; in
450, Latin lost its privileged status at the court of Constantinople; and even Justinian
(529–565), who considered Latin his patria lingua, pragmatically issued his Novels in

Page 4 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
Greek for the eastern provinces and authorized a Greek translation of the legislative col­
lections issued by his government so that the Greek-speaking population might under­
stand them. These (p. 423) decisions help to explain the continued dominance of Greek re­
flected in the Egyptian documents. In any event, although Latin documents are scarce
compared to those of the principate (a characteristic in which they are no different from
the papyrological documentation as a whole), the high proportion of Latin literary texts
for late antiquity is all the more striking (see figure 18.2). This paradox raises questions
about the use of Latin in Egypt.

Conditions of the Use of Latin: Superficial Bilingualism or Diglossia

The Decline of Bilingualism

Latin-Greek bilingualism is less well attested in the late antique papyri than in those of
the Roman period. The reasons for this situation are several: (1) Latin, once obligatory for
status documents (like birth certificates) among the very limited body of Roman citizens,
was no longer closely tied to Roman citizenship, particularly after that had been granted
on all of the subjects of the empire by the Constitutio Antoniniana (212); (2) the growing
recruitment of Graeco-Egyptian auxiliaries for the army led to the disappearance of Latin
private letters, which are a good indicator of true bilingualism practiced in Latin-speak­
ing circles and are found in quantity at military sites from the principate (e.g., O.Claud.,
O. Krok., O.Max., O.Fawâkhir).6 Nor are there for the fourth to the seventh centuries any
archives like those of Claudius Tiberianus and his son, Claudius Terentianus (mainly
P.Mich. VIII 467–481, found at Karanis in the Fayyum, 100–120; cf. Adams 2003, 593–
596), which illustrate the bilingualism of a family of veterans who settled in the Egyptian
chôra.

The only bilingual archive from a military milieu is that of Flavius Abinnaeus (Arsinoite,
341–351).7 Of the eighty-five texts known, two are in Latin. The first is P. Abinn. 1 (340–
342), a petition by Abinnaeus, asking the emperors to confirm his nomination as praefec­
tus alae Quintae Praelectorum, which the dux Aegypti Valacius was refusing him. The sec­
ond, P.Abinn. 2 (344), is a letter in which Valacius removes Abinnaeus from his position.
The use of Latin in these two cases can be explained by the two individuals' knowledge of
the language because of their military status. Considering that the rest of the archive is
entirely in Greek, however, another explanation is more compelling: Valacius uses Latin to
give his letter an official character that might impress Abinnaeus. Similarly, the latter us­
es Latin to address the empire's highest authorities. Here Latin plays the role of “lan­
guage of power.” “On the evidence of this archive, then, Greek was in regular use as an
official, formal language, but Latin was available as a sort of ‘super-high’ language which
could be employed either to make obvious the location of supreme power, or in appeal to
a supreme authority” (Adams 2003, 557).8

(p. 424) Latin as Language and Display of Power

It is striking to observe the disappearance of private letters in Latin. The Latin letters
from fourth-to-seventh-century Egypt, as far as we can identify their authors, are almost

Page 5 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
entirely official letters sent by very high officials: thus, P.Ryl. IV 623 and Ch.L.A. XIX 687
(Hermopolis, 317–324; figure 18.3), letters of recommendation of rationalis (a high fi­
nance official) Vitalis to Delphinius and to Achillius, the governor of Phoenicia, on behalf
of Theophanes (and coming from the only Graeco-Latin archive of the period besides that
of Abinnaeus; cf. Moscadi 1970); P.Abinn. 2 (cf. above); P. Ryl. IV 609 (Hermopolis, 505),
epistula probatoria of the comes rei militaris Thebaici limitis. Latin thus remained the
prerogative of the highest officials of the military and civil administrations, on the model
of texts coming directly from the chancery of the pretorian prefect9 or issued by the em­
peror (constitutions, edicts, rescripts).10

This use of Latin is visible during this period mainly in the bilingual minutes of legal hear­
ings—probably the type of document that most accurately reveals the Greek-Latin “bilin­
gualism” of late antique Egypt, showing the direct influence of imperial policy with re­
gard to Latin, as well as its limits. Despite some earlier examples, it is with Diocletian
that such transcripts of hearings before high magistrates, mixing Latin and Greek, devel­
op (figure 18.4).11 Their particularity lies in the back-and-forth between languages (“code
switching”; cf. Adams 2003, 383–390): The formal structure (date, place, formulas intro­
ducing interventions) is in Latin; the words of the judge are sometimes in Greek (when he
speaks to the defense), sometimes in Latin (when he addresses his officium), while those
of the advocates, the accused, or the witnesses are in Greek; the judge's sentence is in ei­
ther Latin or Greek. “The use of Greek reflects the need for comprehensibility, that of
Latin or mixed-language utterance, the exclusion of some hearers for a moment from the
proceedings, and a desire on the part of the officials to present themselves symbolically
as representatives of Rome and as an exclusive group” (Adams 2003, 386).

One might ask whether the Latin in these texts was not a fictitious representation of what
actually happened. Such doubts would be exaggerated for the fourth century. But a com­
parison between the use of languages in the transcripts of the fourth century and those of
the fifth and sixth (Rochette 1997, 119–20) shows that the colloquy and sentence were on­
ly in Greek from the fifth century on. The increasingly local recruitment of high officials
may explain the gradual loss of adequate competence in Latin. In any case, however, the
formal framework remains in Latin, showing that the language played, even if artificially,
the role of a language of power. In other documents, only the date or a validation formula
(e.g., legi, proponatur) is enough. Rather than bilingualism, we must speak of diglossia,
even if superficial.

This required use of Latin, moreover, increasingly stereotyped and distant from natural
usage, betrays itself in the Latin documents by various symptoms (p. 425) like the faulty
insertion of Greek letters, a failure to respect the rules of Latin spelling, and morphologi­
cal confusions.12

Page 6 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
Latin as a Language of Law

Figure 18.3. Letter of recommendation from the ra­

tionalis Vitalis to the governor of Phoenicia on behalf
of Theophanes of Hermopolis (Ch.L.A. XIX 687, 317–
324 = P.Stras. inv. L1). Photo and collection of the
Bibliotheque Nationale et Universitaire de Stras­
bourg.

Figure 18.4. Report of proceedings before the comes

Aegypti (P.Oxy. LXIII 4381, 375 CE): The beginning
and end are in Latin (date and dialogue between
judge and his staff); the rest is in Greek. Photo cour­
tesy of the Egypt Exploration Society.

Page 7 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation

Figure 18.5. Completion of the lawyer Justus in

Greek and Latin letters (P.Oxy. LXIII 4397, 244–245).
Photo courtesy of the Egypt Exploration Society.

The use of Latin in transcripts of trials is also tied to the fact that Latin was the language
par excellence of law. Roman law, the only valid system at this period, was obviously ex­
pressed in Latin, and after the Antonine Constitution many officials needed to know it.
Diocletian certainly encouraged the study of law, granting privileges to officials who stud­
ied it and creating new professorships of Latin in the east. Alexandria became an impor­
tant center for the study of legal Latin, competing with or complementing the famous law
school of Beirut. This spread (p. 426) (p. 427) of Roman law produced an intense activity of
collecting jurisdictional documents and jurisprudential corpora and of copying legal
books, all of which has left traces in the Egyptian papyri. A nontrivial part of the Latin lit­
erary papyri of Egypt is juristic (fifty-six out of about two hundred texts), reaching a peak
in the fourth to sixth centuries, a period framed by Diocletian, a promoter of Latin lan­
guage and Roman law, and Justinian, whose great activity in legislation and the creation
of legal corpora is well known (figure 18.2).13

The fact that Latin was perceived above all as a language of law had an impact on docu­
mentary practice:

• The notaries (nomikoi, literally “men of law”) might be led to include Latin words in
their documents to refer to legal ideas. This is what Dioscoros of Aphroditê does sever­
al times during a will (P.Cair.Masp II 67152, Antinoopolis, 570), for example in lines
64–66: τὴν ἰδίαν δ ύναμιν καὶ βεβαίωσιν ἔχειν ὡς confirmateumenous ἐν τῇ παρούσᾳ
τελευταίᾳ βουλήσει, “(I wish these codicils) to have their own force and validity as
‘conirmed’ in the present will.” These words are transliterations into Latin letters of
Greek versions of Latin words (here, confirmo becomes confirmateuo, transliterated in­
to Latin with a Greek participial ending).
• From the fifth to the seventh century the notaries of the Arsinoite, Oxyrhynchite, and
Herakleopolite nomes write their signatures at the end of the documents in Latin
transliteration.14 For example, P.Oxy. LXIII 4397 (545): δι ʼ ἐμοῦ Ἰούστου ὑποδιακ(όνου)
συμβολαιογράφ(ου) ἐτελειώθη | di ʼ em(u) Iust(u) upodiacon(u) sumbolai(ografu) etelio­
th ḥ, “completed by me, Justus, subdeacon and notary” (figure 18.5). This artifice,
which sometimes produces its share of blunders (P.Oxy. VII 1042.34: di emu Serηnu
etelioth), gave legal prestige to the document. This is the best example of a very imper­
fect diglossia—I should say digraphia—denuded of all bilingualism.

Page 8 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
(p. 428) Latin Literary Papyri: The Needs of a New Class
The development of Latin as a language of power and law, however marginal or superfi­
cial, led to the emergence of a new class of civil and military officials and of lawyers
(scholastikoi) who had to acquire at least a rudimentary knowledge of Latin. This learning
process is well documented by the Latin literary papyri found in Egypt. The graph of
these papyri is highly significant when compared to that of Greek literary papyri, which
reaches its peak in the second century; it rises abruptly in the fourth century and reaches
a peak in the fifth before falling again (figure 18.2). The Latin papyri attest various phas­
es of learning, ranging from elementary teaching to a higher level (Rochette 1997, 165–
210; Adams 2003, 623–630): (1) Latin alphabets with equivalents of Greek letters, some­
times with the names of Latin letters in Greek; (2) writing exercises; (3) Latin or Graeco-
Latin grammars; (4) Graeco-Latin glossaries (where Latin is often transliterated into
Greek) of several types:15 alphabetic or thematic glossaries or glossaries of Latin authors
(Rochette 1990, 1996; C.Gloss.Biling. II pp. 5–8); (5) conversation manuals; (6) collections
of model letters (just one case, from the third—fourth century, P.Bon. 5, which gives let­
ters with a strongly legal character as examples); (7) Latin authors (Mertens 1987), above
all Cicero and Vergil but also Juvenal, Livy, Lucan, Sallust, Seneca, Terence, anonymous
works like the Montserrat Alcestis (Mertens-Pack3 2998.1, fourth century), or the Psalmus
Responsorius in the same codex (van Haelst 1210) and legal authors (e.g., Gaius, Papini­
anus, Ulpian, Codex Theodosianus).

Various indications, like transliteration into Greek (Kramer 1984, 1378–1379; Rochette
1999) and the use of administrative vocabulary in the glossaries, show that these papyri
were composed mainly for Greek speakers, who were less interested in the glories of Ro­
man culture than in the necessity—more and more pressing from Diocletian on—to learn
Latin quickly for practical reasons.

Latin in Christian Milieus

The Latin-Greek-Coptic conversation manual (Mertens-Pack3 3009 = P.Rain. Unterricht
kopt. 270, C.Gloss.Biling. I 15, sixth century), given its Coptic element and the religious
coloration of its lexicon, seems to have been intended for Christian (perhaps monastic)
circles. It poses the problem of the connections between Latin and Christian milieus
(Cavenaile 1987). A number of Latinized monastic circles existed in Egypt, but these were
very marginal: It was for Latin-speaking monks of the Pachomian congregation that
Jerome translated Pachomius's rule into Latin (404). Apollô directed a community near
Hermopolis where several monks, we are told, were “versed in Greek, Latin, and Egypt­
ian” (Historia Monachorum in Aegypto 8.62). More interesting is the presence of Latin in
the Montserrat codex mentioned earlier, part no doubt of an ancient collection of Christ­
ian books in Greek and Coptic from the region of Nag Hammadi, to which a Greek-Latin
lexicon of the Pauline epistles also belonged (Wouters 1988). This library, the setting of
which is (p. 429) still controversial, may have belonged to a religious community (perhaps
sectarian) of Upper Egypt (Agosti 2002, 80–87), and these texts demonstrate that its
members needed instruction in Latin, as well as Greek and Coptic. Similarly, among the
Manichaeans of Kellis, the need to learn Latin was felt around the same time (P.Kellis VC.

Page 9 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
20.24–26: “The Great Teacher let him travel with him, so that he might learn Latin”), per­
haps for missionary purposes.

Other Latin papyri from a Christian setting are rare and hard to interpret because we
know little about their precise context: P.Lond. V 1792 (fifth—sixth century), a Latin letter
addressed by an epitropos to an ecclesiastic, and the curious cluster of Greek letters in­
troduced by a Latin sentence of biblical inspiration, with Latin docket (P.Oxy. XVIII 2193–
2194 and P.Köln IV 200, fifth—sixth century). Other than those already mentioned, Christ­
ian literary papyri in Latin are the products of schools: PSI XIII 1306 (fourth-fifth centu­
ry), Ephesians in Latin and Greek; P.Rain. Unterricht 184 (sixth—seventh century; with re­
gard to the date, cf. P.Thomas, pp. 19–23), Latin Lord's Prayer with transliteration into
Greek.

The data are scant and difficult to use. Despite everything, Latin remained marginal:
Greek was the official language of the church, Coptic its natural language and that more
customary in monastic milieus.

Consequences of Latino-Greek Bilingualism or Diglossia

It is not always easy to distinguish between the results of Romanization and those of La­
tinization. The most obvious result of the latter is the penetration of numerous Latin
words into the Greek of Egypt, a phenomenon visible already under the principate but
limited essentially to military milieus and certain semantic fields (cf. Ghiretti 1996) and
much more evident from Diocletian on. These words, collected earlier by Daris (1991),
are now the object of a specialized dictionary based on Greek and Coptic sources,
Lex.Lat.Lehnw., which will provide a solid base for a complete study of the impact of
Latin on the Greek of Egypt. For the moment, we may simply recall that Latin words were
obviously introduced in the domain of civil and military institutions and of taxation as a
result of imperial reforms (sometimes going so far as to duplicate Greek words, like
δηφήνσωρ < defensor, next to ἔ κδικος). More revealing are numerous Latin words that
concern artisanry (names of trades and products, especially textiles and foodstuffs). The
impact of Latinization is also visible in the degree of adaptation by the borrowing lan­
guage. We see, indeed, many hybrid words (Daris 1991, introduction): verbs with Greek
suffixes (ἀμβιτεύω < ambio; ἐξπελλεύω < expello), sometimes without any Latin equiva­
lent (ἀδνονμενεύω < ad nomen); nouns created from two Latin words
(δελματικομαφόριον < dalmatica and maforte) or from a Latin plus a Greek word
(σύγκελλος < σύν + cella; παγάρχης < pagus + -άρχης); abstract nouns with no Latin
equivalent (ἐξακτορία < exactor; ῥιπαρία < riparius). Such formations show (p. 430) how
perfectly Latin words were integrated into the borrowing language. The influence of Latin
is also apparent at the level of handwriting, especially in the development of the so-called
Graeco-Latin graphic koinê (see chapter 5). Finally, as far as bibliology is concerned, Ro­
man juristic practice, which privileged the codex, was certainly one of the influences
(along with Christian books) in the eventual triumph of the codex over the bookroll and
its adoption not only for accounting and recordkeeping but also for literary authors (Gas­
cou 2000, 289–291). Greek paideia, so resistant to Latin influence, finally bowed before

Page 10 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
Roman pragmatism, but here we move into the complex domain of Romanization and
leave the realm of language for the broader world of culture.

Coptic: Diglossia with Real Bilingualism

The relationship of Greek and Coptic is richer, better documented, and more complex.
The population of late antique Egypt mostly spoke its vernacular language, Coptic, or
more exactly “Egyptian,” as the papyri say, but in some circumstances it had to use
Greek, the language common to the entire east. The problem, more fundamentally, is one
of the relationship of a population with its administration and elites—relationships in
which the phenomena of multilingualism and diglossia are most clearly observable. There
is much in this area we still do not know, above all because Coptic papyrology has ad­
vanced less rapidly than Greek. Recent discoveries (Kellis) are in the process of shedding
new light, but the moment for a full synthesis has not yet arrived. In short, Coptic papy­
rology is a domain in the midst of change and expansion (Clackson 2004).16

The Birth of Coptic

The first three centuries of Roman domination witnessed the decline of the vernacular
scripts: The hieroglyphs, no longer used except as an artificial script in monumental in­
scriptions, disappear after 393/394 (Philae); hieratic, also tied to temple culture, starts
becoming incomprehensible even to priests starting in the second century, as the Ono­
masticon of Tebtynis shows, with its marginal glosses in demotic and Old Coptic (Osing
1998, papyrus I). Starting in the middle of the first century, even demotic, the only script
in current use to represent the spoken Egyptian language, declines along with the tem­
ples where it was taught, and it vanishes almost completely in the third century, even
though it is attested as late as (p. 431) a graffito at Philae in 452/453 (Bagnall 1993, 237;
Lewis 1993; Fournet 2003a, 429–430). The Egyptians thus found themselves without a
script for communication among themselves and were obliged, if they could, to use Greek
—a point important for the decline in the quality of written Greek, visible in many papyri
of the second to the fourth centuries.

Furthermore, it was to the convenient, alphabetic Greek that they turned to try to create
a new script. Apart from some transliterations of names, it is only in the first—second
century that the first attempts at creating a coherent graphic system are found, one that
used the Greek alphabet augmented by some demotic signs denoting phonemes absent in
Greek: This is what is called Old Coptic (Quagebeur 1982, modified by Bagnall 2005),
born in a pagan milieu intent on preserving traditional learning and allowing the record­
ing of texts tied to cult practices such as magic and astrology. The school ostraca of the
temple of Narmouthis (Fayyum), which date to the second to third century, in which hier­
atic words are transliterated into Old Coptic (O.Narm.Dem. II 34–41), offer a good exam­
ple of the context in which these experiments in Old Coptic were conducted (figure 18.6).

Page 11 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
Among the varied experiments in Old Coptic, it is probably in the third century that one
took hold, was diffused by pathways still obscure to us, and led to the birth of Coptic
properly speaking (Bagnall 2005, 18). This, unlike Old Coptic, was certainly a deliberate
creation that resulted from Christian initiative in well-of, Greek-speaking Egyptian circles
—hence the large quantity of Greek words in Coptic (about 20 percent), not only concrete
ones but also auxiliary words like conjunctions and prepositions.17 Coptic is thus the off­
spring of Greek-Egyptian bilingualism, designed above all for the translation of the scrip­
tures into a vernacular tongue—a need evident already in the third century in the Coptic
glosses to the Greek text of Isaiah in P.Chester Beatty VII or the minor prophets in a
codex in the Freer Collection (van Haelst 293, 284). In the following century, Coptic ex­
panded dramatically not only as a literary medium but also for ordinary use, still in Chris­
tian circles. The Egyptian population, which became mainly Christian in this period, had
from this point on its own means of writing, which allowed it to communicate in its own
language. The expansion of Coptic had consequences for Greek: With those only uneasily
bilingual now able to communicate in Coptic, the quality of the language in Greek letters
recovers something of its higher level seen in the Hellenistic period.

Figure 18.6. A school exercise in hieratic writing

with translation in Old Coptic (O.Narm.Dem. II 38,
second—third century CE), facsimile. Courtesy of
Paolo Gallo.

Despite the rapidity of the spread of Coptic and the homogeneity of its users, from a lin­
guistic standpoint it was highly diverse: It was made up of several dialects identifiable in
their phonological and graphic variation. These are mainly, from north to south, Bohairic
(B), Fayyumic (F), Mesokemic (M), Akhmimic (A), sub-Akhmimic or Lykopolitan (A2 or L),
and Sahidic (S).18 Sahidic (“southern dialect”) was the standard Coptic literary dialect un­
til its displacement by Bohairic (“northern dialect”) in the tenth century, probably be­
cause of the settling of the Coptic patriarchate in Wadi Natrun (west delta). (p. 432)

Page 12 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
Coptic versus Greek

Characterizing the relationship between the use of Coptic and the use of Greek and, on
the basis of that, describing the Egyptian-speaking community are difficult problems that
have been approached through various methods.

A Complex Problem: Inadequate Criteria

(1) Ethnicity
This is difficult to perceive. Legal categories are useless because the terms “Egyptian”
and “Greek,” as the Romans used them, did not correspond to any ethnic realities, only to
legal and fiscal distinctions (thus, the Greeks of the chôra were classified as “Egyptians”);
moreover, these were obsolete by late antiquity. Only with the coming of the Arabs would
a new term arise, “Copt” (a deformation of the Greek word Aigyptios, “Egyptian”), which
designated the Egyptians—all Christians—in opposition to the Muslim Arabs.

(p. 433) Nor are naming practices able to reveal ethnicity from the fourth century on. The
onomastic repertory was Christianized, thereby partly erasing the differences between
Greek and Egyptian names. We see, moreover, particularly in the fourth century, a fashion
for certain Greek names even among the Copts; and, finally, some Graeco-Egyptian names
(and even Latin ones) are of indeterminate character. The names of the authors of fourth-
century Coptic letters found at Kellis (P.Kell. V) are sufficient proof: Egyptian names, to be
sure (Psemnouthes, That, Shamoun), but also Graeco-Egyptian (Horion), Christian
(Matthaios, Makarios), Greek (Lysimachos), and Latin (Valens) names.

In a class by itself was Alexandria, a profoundly Greek city that had no very visible Egypt­
ian-speaking population, as the absence of Coptic inscriptions and graffiti shows.

(2) Religion
Although this is more solid, it is complicated to handle. Pagan (or paganizing) Hellenism
has often been set up against a Coptic Christian culture. This distinction, always too
schematic, becomes unworkable after Christianity becomes the state religion under Con­
stantine. Another schematic opposition, between anti-Chalcedonian Christians who spoke
Coptic and Chalcedonian Christians who spoke Greek,19 depends on a nationalistic vision
that has been strongly criticized of late (Wipszycka 1992, esp. 122–125). This opposition
is all the more futile in that the Chalcedonian/anti-Chalcedonian split is almost entirely
absent from the papyrological documentation and does not seem to have been a criterion
of linguistic diferentiation. The language of the church of Egypt remained Greek as long
as possible. It was the liturgical language and the language of communication between
the patriarch of Alexandria and his bishops. The festal letters, veritable treatises with
doctrinal and disciplinary content sent by the patriarch to the bishops of Egypt each year
to announce the date of Easter, were still written in Greek well after the Arab conquest
(BKT VI 5 = van Haelst 621, from 713 or 719). Greek was the language of Christian pub­
lic space, as, for example, in the monumental inscriptions in Egyptian churches (see the
inscription of the Muʻallaqa, the “Hanging Church” from 735; Fournet 1993).

Page 13 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
This inadequate dichotomy has given way to a more legitimate opposition: Greek was the
language of the church, but Coptic was spoken by most of the monks. Monks and an­
chorites could get by without Greek and use the maternal language of the majority,
whether by deliberate rejection of Greek, by indifference toward the Greek paideia so
ubiquitous in the world, or by cutting themselves off from public life, where Greek re­
mained indispensable. In fact, the majority of Coptic papyri from the fourth to the sev­
enth centuries come from monastic settings or from sectarian religious communities like
the Gnostics or Manichaeans. But Egyptian monasticism was more diverse than is often
thought: Besides the Latin speakers mentioned earlier, it included a Greek or Greek-
speaking component (Wipszycka (p. 434) 1992, 115–116) and even to some extent encour­
aged Greek, which was necessary for managing the monasteries' economic activities.
Moreover, in early Egyptian monasticism, those monks and anchorites, who acted as in­
termediaries between the population and the authorities, needed Greek. We see this from
the bilingual fourth-century archives of the Melitian monastery of Hathor and of Apa Jo­
hannes (mentioned later), which contradict the ideological vision transmitted by hagio­
graphic literature (van Minnen 1994).

At the same time, it was quite possible for important church positions to be occupied by
Copts who did not speak Greek (Wipszycka 1992, 110): Kalosiris, bishop of Arsinoe, need­
ed an interpreter at the Council of Ephesos (ACO II/1/1. p. 185, 90); the bishops Abraham
of Hermonthis and Pisenthios of Coptos (sixth-seventh century) did not know Greek.

This connection between Coptic and the monks (supposedly poor and rejecting the city)
has led scholars to propose a third criterion.

(3) Social and Economic Status

Urban communities, because they were more affluent and cultivated, would on this view
have spoken mainly Greek, while village populations (less rich, uneducated) spoke Coptic.
This schematization is partly valid (especially for the urban elites involved in municipal
duties), but the use of the two languages was actually less simply divided. Centuries of
cohabitation had led Greeks to learn Egyptian, particularly in the course of managing
agricultural estates—a point on which the papyri do not inform us adequately. In return,
some categories of villagers needed Greek in order to manage their businesses and carry
on village duties, as well as in relations with the pagarch and the metropolis. A kind of
natural bilingualism thus developed. In the fourth century, the villagers of Kellis, even if
they have Egyptian names, write both Greek and Coptic. In the sixth century, the archives
of the village of Aphroditê are almost entirely in Greek, and one of its inhabitants,
Dioscoros, not only was capable of drawing up Greek legal documents as a notary in the
capital of his province, the Thebaid, but could also write poems in Greek. In short, this
criterion also gives only an imperfect account of spoken bilingualism and simultaneously
fails to explain the distribution of the written documentation between Greek and Coptic.

The best explanation of the evidence is that the two languages were complementary, de­
pending on the context of usage; in other words, there was a functional link between lan­

Page 14 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
guage and type of written document, which evolved over time. Bilingualism combines
with a clear diglossia, the contours of which gradually change.

The Complementarity of the Two Languages by Genre

Figure 18.7. Quantitative comparison of Greek and

Coptic documents.

The number of Coptic documents follows an upward curve (figure 18.7), almost the re­
verse of that of Greek documents.20 These two curves clearly illustrate that the conditions
of use and complementarity of Greek and Coptic evolved between (p. 435) the fourth and
seventh centuries—to the benefit of Coptic, which would replace Greek in its major do­
main, legal documents, after the Arab conquest, which marked the end of Byzantine rule
in Egypt (Richter 2002, 16–27). Three phases can be distinguished, which we will treat by
looking at the bilingual archives of each period; these demonstrate the relationship be­
tween Greek and Coptic and its evolution, as shown in figure 18.8, which classes the Cop­
tic documents according to three types (letters, accounts/lists, others).21

Page 15 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
(1) Fourth Century to About 570: The Development of Coptic
For this period, we have two Graeco-Coptic archives:

Figure 18.8. The Coptic documents from the begin­

ning up to the Arab conquest.

• The archives of the Melitian monastery of Hathor (Kynopolite nome), made up of that
of its abbot Apa Paieous (334–340)22 and that of one of his successors, Nepheros (350–
360).23 Paieous, with his Egyptian name, knew both Greek and Coptic; he signed in
Greek on a contract (P.Lond. VI 1913.18), giving his name a Hellenized form, Pageus;
he also received numerous letters in Greek and at the same time wrote (?) and re­
ceived letters in Coptic (P.Lond. VI 1920–1922). The same is true of Nepheros. Coptic
is used only for letters in these archives, while other documents (contracts, receipts,
(p. 436) accounts) are in Greek. It is worth noting also that the three Coptic letters in

the dossier of Paieous concern very down-to-earth subjects (making clothes, sending
provisions), in contrast to some of the Greek letters, which deal with important sub­
jects like the relations between the Melitian congregation and the bishop of Alexandria
(P.Lond. VI 1914). Moreover, the editor of P.Lond. VI 1914 regarded the writer as a
Copt (because of the errors he made); as a result, the choice of Greek would have been
linked to the nature and importance of this letter, a kind of official report.
• The archive of the anchorite Apa Johannes (Lykopolis, ca. 375–400).24 This dossier is
made up of letters in Greek and in Coptic, addressed by monks, churchmen, soldiers,
officials, and private individuals to Apa Johannes, who is identified with the famous
John of Lykopolis, a “holy man” known from the literary sources (Zuckerman 1995,
188–194).25 These persons ask John to pray for them or to intercede in their favor with
the authorities. It is difficult to assess the reasons for the choice of language, particu­
larly when the senders do not give their identity. Some of the authors of Coptic letters,
however, are certainly monks (P.Ryl.Copt. 268, 269, 271, 313). Even this, however, is
not so simple: One of supplicants, Psoïs, writes to Apa Johannes in Greek, even though

Page 16 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
the poor quality of his Greek shows that he was a Coptic-speaking Egyptian (P.Herm.
7). It is thus all the more surprising to find Psoïs writing in Greek to a Copt who is said
not to have known Greek. We have in fact a letter by (p. 437) Apa Johannes himself,
written in Greek, but only its signature is in his own hand, and that is in Coptic (P.Amh.
II 145; figure 18.9). To put it another way, Apa Johannes must not have known Greek
(information confirmed by Palladius) and, in order to deal with the numerous requests
he received, used a “secretariat,” which translated those in Greek for him and drew up
at his request letters that he needed to send in Greek. The letter of Psoïs shows that a
Copt could feel himself obliged to write in Greek to another Copt.

To these two archives we may add, besides the papyri from the Nag Hammadi carton­
nage, those from Kellis (Dakhla Oasis), which, although they do not form an archive prop­
erly speaking, are of the greatest interest:

• The texts from Kellis (ca. 350–370) come from a Manichaean circle.26 Apart from a
number of accounts, the Coptic texts are private letters, while the Greek papyri in­
clude, besides private and business letters, petitions, receipts, and a variety of official
and legal documents (e.g., prefectural edict, loans, sales, manumission). Apart from
this clear partition, one is struck by the perfect bilingualism of some of the main fig­
ures. Thus, Tithoes writes a Coptic letter to his son, Samoun (P.Kell. V C. 12), while the
latter writes a letter in Greek to his father (P.Kell. I 12). Moreover, Tithoes and Samoun
receive letters in Greek from Ammonios (P.Kell. I 10 and 11). A woman named That
goes from Coptic to Greek within a single letter without an obvious reason (P.Kell. V C.
43). The bilingual character of this community is well summed up in Makarios's recom­
mendation to his “son,” Matheos: “Study [your] psalms, whether Greek or
Coptic” (P.Kell. V C. 19.13–14).

These three groups very clearly show a division between private documents written in
Coptic and legal documents written exclusively in Greek. Apart from some private ac­
counts, Coptic is reserved for private letters. Nonetheless, these remain strongly marked
by Greek in that they borrow structure and formulary from Greek letters; in some cases,
the address (prescript), the final salutation, and the endorsement are written in Greek.
All of this presents a picture of a population more bilingual than one might have expect­
ed, one that, except the particular case of legal documents, moved from one language to
the other according to criteria that are not always evident to us and are not based on be­
longing to a monastic community or on not knowing Greek. The case of Psoïs plainly
shows that, in the fourth century, a permeability between the two languages continued to
exist, made possible by widespread bilingualism even if sometimes imperfect. The nation­
alist conception of the use of languages is clearly untenable.

Page 17 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
(2) Circa 570–642: A Period of Transition

Figure 18.9. A Greek letter of anchorite Apa Jo­

hannes, signed by him in Coptic (P.Amh. II 145, end
of the fourth century CE).

This was a hinge period, during which Coptic began to be used, rather timidly, for legal
documents (included in “Others” in figure 18.8). Four archives date to this period and il­
lustrate this phenomenon well: (p. 438)

• The archive of Dioscoros of Aphroditê (506–585)27 is the largest group of

(p. 439)

Byzantine papyri, the Coptic part of which is still poorly known and partly unpub­
lished. Its protagonist, Dioscoros, a notable of the village whose business affairs he
managed, a small landowner, a notary, and curator of the monastery founded by his fa­
ther, knew both Coptic (his native language) and Greek (which he mastered sufficiently
to write poetry; see P.Aphrod.Lit. IV). In contrast to the Greek texts (petitions, con­
tracts and other notarial texts, administrative letters, tax receipts, private and official
accounts), the majority of the Coptic texts are letters, either private or concerned with
his father's monastery (MacCoull 1991, 1992, 1993). But we also find here, for the first
time in Coptic, texts of another sort: two arbitrations drawn up by Dioscoros (569 and
ca. 570),28 the first between monks (figure 18.10), the second between laity.
• The archive of Pathermouthis, son of Menas (Syene, 574–613),29 contains primarily
Greek documents concerning the properties of Patermouthis and his wife, Kako (sales
of houses and boats, cessions) and settlements of the disputes that these provoked
among members of the family. The few Coptic documents have a characteristic profile:
We find at least one letter (SB Kopt. III 1293), a legal settlement (P.Lond.Copt. I 446),
and a debt settlement (SB Kopt. III 1395). Moreover, the only documents of the archive
in which Kako acts without her husband are in Coptic (SB Kopt. III 1394, 1395), which
may link the use of Coptic to an inability, more widespread among women, to express
oneself in Greek (Clackson 2004, 29).

Page 18 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation

• The archive of Abraham, bishop of Hermonthis (Thebes, ca. 600),30 is an important

group written almost entirely in Coptic. It contains a number of letters concerning
Abraham's pastoral activity. More original is, for the first time, a coherent body of texts
of a juridical nature. These are mostly sureties, but we find also orders, work con­
tracts, and loans. They are sometimes described using the Greek technical terms as­
phaleia (“guarantee”) or homologia (“acknowledgement”). They close with the signa­
ture of the contracting party and sometimes with a witness (using the Greek verb
stoikhei, “I agree”). Still, we must underline the fact that these apparently legal texts
come from an ecclesiastical context. On the other hand, while he is abbot of the
monastery of Phoibammon (Thebes), Abraham has his will written (as was legally
obligatory) in Greek (P.Lond. I 77—the only Greek document of his archive). Some thir­
ty years later (634), Abraham's successor as head of the same monastery, Apa Victor,
made his will in Coptic (P.KRU 77). This development shows the advance of Coptic.

Figure 18.10. The first dated Coptic text: a settle­

ment drafted by Dioscoros of Aphrodite (P.Alex. in v.
689, 569 CE). Photo by Alain Lecler (IFAO), courtesy
Graeco-Roman Museum, Alexandria.

• The small archive of Pachymios, purple-seller (Panopolis, 592–616)31 contains various

types of contracts in Greek (house sales, property division, loans, work contract) and
offers the first dated example of a contract in Coptic, a betrothal agreement (CPR IV
23, 608), drawn up at This by the (p. 440) bilingual notary Paul, to whom we owe other
contracts in Greek (SB I 4503–4505; cf. MacCoull 1995, 347–350).

These archives illustrate the same polarization between the private sphere, in which Coptic is
ever more extensively used (letters), and the public sphere, where Greek is needed (legal texts).
To be sure, the dossier of Bishop Abraham, who did not know Greek, shows how the church of
Upper Egypt, by pragmatism and necessity, sought to (p. 441) develop legal texts in Coptic in im­
itation of public usage, but these concern only clergy and monks, who were mainly Coptic-speak­
ing.

Page 19 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
The first Coptic texts of a legal nature, drawn up by bilingual notaries, begin appearing in
569/570: arbitrations (Fournet forthcoming). But these texts display above all the parties'
desire not to involve themselves in legal process (petition, trials in which Greek was re­
quired) and to resolve their controversies amiably and in their own language. They open
up a kind of parallel track, which anticipates the development of Coptic legal texts in
open competition with Greek. The list of those more or less precisely dated before the
Arab conquest is short: CPR IV 90 (Hermopolite, 596), sale on delivery (the first Coptic
text with notarial subscription); CPR IV 23 (Panopolite, 608), betrothal contract; CPR IV
48 (Hermopolite, 625), sale on delivery; O.CrumST 436 (Panopolite, 619 or 634), acknowl­
edgement of debt; P.Vat.Copt. 2 and 3 (Antaiopolite, 624/625 or 654/655), division of
house property and undetermined contract; P.Vat.Copt. 1 and 5 (Antaiopolite, 625/626 or
655/656), sales; P.CrumST 48 (Thebes, 625 or 640), receipt for a deed; O.CrumVC 5 (Ed­
fou, 627), joint declaration concerning an inheritance; P.KRU 77 (Thebes, 634), will.32
How are we to explain the appearance of these texts, which pose the thorny problem of
the legal status of the Coptic language? It is true that we cannot judge a priori the legal
validity of these documents even if they likely were valid. They come only from the The­
baid; does this reflect regional distinctiveness among the notaries? Or the weaker influ­
ence of Byzantine power and of Romanization? Or perhaps a deterioration of the judicial
system? Should we connect it with the development of jurisdictional power in the church,
some of whose bishops in the Thebaid did not speak Greek because they came from
monastic circles to which they remained closely tied? In any event, it is noticeable that
the acts with the most significant legal implications (sales or exchanges of property) date
either to the Persian period (619–629) or to the Arab period (after 642), that is, to times
of the weakness or disappearance of Byzantine power.

Page 20 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
(3) After 642: The Rise of Coptic in All Types of Writing

Figure 18.11. The beginning of a Coptic legal text (a

release) with a Graeco-Arabic protocol, a Greek invo­
cation, and a dating formula (P.CLT I, 698 CE; Metro­
politan Museum of Art, Accession Numbers 24.2.3a-
c). Image copyrighted by the Metropolitan Museum
of Art.

With the end of Byzantine domination as a result of the Arab conquest, Greek lost its ex­
clusive status for legal documents, the number of which in Coptic rapidly increases at the
expense of Greek, until Arabization in turn drives Coptic into disuse. Greek nonetheless
continued in use until the eighth century (Worp 1984). It must, however, be stressed that,
until that point, Coptic legal documents, which in any case owed much of their phraseolo­
gy to their Greek counterparts, often continued to be preceded (after a Graeco-Arabic
protocol) by an invocation and a dating formula in Greek, or at least concluded with a
Greek notarial subscription (figure 18.11). This “fossilized” Greek plays, mutatis mutan­
dis, the same role that Latin occupies in a number of documents of an earlier period, giv­
ing them an official tone. (p. 442)

Some Aspects of Coptic Literacy in Greek

The historical circumstances of late antique Egypt made Greek indispensable in the ad­
ministrative sphere, the business world, and the liturgy. The Copts were thus pushed to­
ward bilingualism, that is, acquiring some Greek literacy.

Bilingual Education
“Coptic, which was born in bilingual milieus, in its beginning stages was probably taught
in conjunction with Greek” (Cribiore 1996, 4). The papyrological documentation has pro­
duced numerous witnesses to the combined teaching of Greek and Coptic or to the teach­
ing of Greek in Coptic milieus, at least down to the seventh century (Cribiore 1999).

Page 21 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation

Figure 18.12. An example of the difference in graph­

ic style between Coptic and Greek in the same docu­
ment (O.BawitIFAO 49, transport order, VII/VIII CE).
The framed area is in Greek (cursive with ligatured
letters); the rest of the text is written by the same
hand in Coptic (capitals without ligatures). Repro­
duced by permission of the Institut francais
d'archeologie orientale, Cairo.

There is one Graeco-Coptic school text as old as from the end of the third century
(Cribiore 1996, no. 388): a codex of tablets containing, in Greek, among other (p. 443)
items, a paraphrase of the Iliad, and, in Coptic, Psalm 46:3–10, as well as writing exercis­
es. Whether the library constituted by the codices now in the Bodmer, Chester Beatty, and
Montserrat (formerly Barcelona) collections and found not far from Nag Hammadi be­
longed to a schoolteacher or to a religious body (Fournet 1992, 253; Schubert 2002;
Agosti 2002, 80–87), it attests the teaching of Greek and Coptic, along with Latin, in the
fourth century. In the same period, in Kellis, the young Matheos was learning the Psalms
in both Greek and Coptic (P.Kell. V C. 19.13–14). In the following centuries, papyrological
evidence becomes more plentiful. Thus, we find the Sententiae of Menander, written in
Greek and in Coptic (P.Rain. Unterricht. kopt. 268, seventh century, and 269, fifth-seventh
centuries). The (p. 444) monasteries, which attracted people of all ages, varying back­
grounds, and unequal literacy, have provided many examples of this double education;
thus, in the Theban region, that of Phoibammon (Cribiore 1996, no. 19) or that of Epipha­
nius (P.Mon.Epiph. II 611–620).

To these we must add the lexica, glossaries, and Graeco-Coptic conversation manuals, no­
tably P.Rain.Unterricht. kopt. 270 (=C.Gloss.Biling. I 15), a Latin-Greek-Coptic manual of
the sixth century; the Greek-Coptic lexicon of Dioscoros of Aphrodite (P.Rain.Unterricht.
kopt. 256, sixth century); and an Old Testament glossary (P.Rain.Unterricht. kopt. 257a,
third-fourth centuries).

Page 22 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
Fluent Literacy
Besides the bilingual letter writers of Kellis, I have already mentioned the bilingual no­
taries (e.g., Dioscoros of Aphroditê; Paul of This). These notaries were proficient in not
merely two languages but two scripts as well: They were in effect capable of switching
from one language to the other, fluently employing the writing style appropriate to each
(from the fifth-sixth century on, Greek was more cursive than Coptic) (figure 18.12). In
addition, when they drew up a contract in Greek they were aware of the problems of
bilingualism and its possible limits in their clients. To avoid misunderstanding, they did
not hesitate to translate certain Greek technical terms, glossing them with the appropri­
ate Coptic word, introduced by ἤτοι “that is” (e.g., P.Münch. I 11.27: τον̂ ύποπ ϵσ<σ>ίου
ἤτοι χρηρ ϵ, “the cubbyhole under the staircase, or khrere”; cf. Sijpesteijn 1992, 242;
Torallas Tovar 2004, 170).

Deficient Literacy
Coptophone writers frequently betray themselves by their handwriting (e.g., in unliga­
tured uncials), their Copticization of personal or Egyptian geographical names, introduc­
ing Coptic letters, or replacing Greek terminations (-ος by -ϵ, for example), their untimely
insertions of Coptic words, and their phonetic mistakes such as the confusion of conso­
nants (e.g., mixing voiceless and voiced consonants, replacing ρ with λ in the Fayyum) or
of vowels (e.g., ignorance of Greek quantities, variation in the use of vowels because of
the tendency of the Coptic system accent to change vowel qualities).33 Sometimes defi­
cient syntax is the clue. Cases of Copticisms are more doubtful or dificult to distinguish
(see, e.g., P.Ross.Georg. IV, appendix, p. 99, a letter full of Copticisms, or, in literature,
Torallas Tovar 2004, 172).

Illiteracy
As in the Roman period, those illiterate in Greek could avail themselves of various solu­
tions: asking a literate person to write a letter for them or, in the case of legal documents,
to subscribe on their behalf with the usual formula: “I, so and so, wrote on his behalf be­
cause he does not know how to write.” (p. 445) Translation must have played a more im­
portant role than our documentation allows us to see. In contracts where the main party
did not know Greek, the text was translated for him by the notary. We know only two ex­
amples in which such a phenomenon is expressly signaled:

• P.MünchA 13.71 (Syene, 594), a sale contract entered into by two women with Egypt­
ian names (Tsônê and Tseure), who declare at the end that they agree with the text “af­
ter it had been read to us and translated into Egyptian.”
• P.Lond. I 77.69 (p. 231) (Hermonthis, ca. 600), the will of Bishop Abraham, men­
tioned earlier, which ends with the statement that “everything was translated into
Egyptian for me by the notary named below.”

It is obvious that in most cases this formula was omitted even though translation had oc­
curred. That went without saying.

Page 23 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
Public offices also needed translation in order to publicize certain texts. Thus, an order by
the duke of the Thebaid on payments to officials had to be “translated into the vernacular
language and posted” (P.Cair.Masp. I 67031.16, sixth century). Here we see the adminis­
tration taking account of the linguistic limitations of its population, something that con­
tradicts the popular image of a haughty imperial power that despised the vernacular lan­
guages.

One methodological conclusion is unavoidable: Egypt, because of the quantity of its pre­
served written documentation, is an ideal setting in which to observe the phenomenon of
multilingualism. This cannot be understood without taking into account the totality of the
written sources in the various languages, each of which illumines a different and comple­
mentary facet of its history. The era of a Hellenocentric papyrology has passed, and the
papyrologist or historian of Egypt today must dismantle the barriers between the various
bodies of material in order to let them converse with one another.

Bibliography
CII = D. Weber, Ostraca, Papyri, und Pergamente: Textband. Corpus Inscriptionum Irani­
carum. Part 3. Pahlavi Inscriptions. Volume 4. Ostraca and Papyri. Texts l. London 1992;

D. Weber, Berliner Papyri, Pergamente und Leinenfragmente in mittelpersischer Sprache.

Corpus Inscriptionum Iranicarum. Part 3. Pahlavi Inscriptions. Volume 4, Ostraca, and
vol. 5, Papyri. London 2003

(photos in
J. de Menasce, Ostraca and Papyri. Corpus Inscriptionum Iranicarum. Part 3. Pahlavi In­
scriptions. Volume 4–5. Ostraca and Papyri. Portfolio l. London 1957).

Mertens-Pack3 + number = Catalogue des papyrus littéraires grecs et latins. http://

promethee.philo.ulg.ac.be/cedopal/index.htm.

Van Haelst + number = Van Haelst, J. 1976. Catalogue des papyrus littéraires juifs
(p. 448)

et chrétiens. Paris: Publications de la Sorbonne.

Adams, J. N. 2003. Bilingualism and the Latin Language. Cambridge: Cambridge Universi­
ty Press.

Agosti, G. 2002. “Il poemetti del Codice Bodmer e il loro ruolo nella storia della poesia
tardoantica.” In Le Codex des Visions, ed. A. Hurst and J. Rudhardt, 73–114. Geneva: Li­
brairie Droz.

Altheim-Stiehl, R. 1991. “Wurde Alexandreia im Juni 619 n. Chr. durch die Perser er­
obert? Bemerkungen zur zeitlichen Bestimmung der sasanidischen Besetzung Ägyptens
unter Chosrau II. Parwez.” Tyche 6: 3–16.

———. 1992a. The Sasanians in Egypt: Some Evidence of Historical Interest.. BSAC: 31.

Page 24 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
———. 1992b. “Zur zeitlichen Bestimmung der sasanidischen Eroberung Ägyptens: Ein
neuer terminus ante quem für Oxyrhynchos ist nachzutragen.” In MOYΣIKOΣ AN HP:
Festschrift für Max Wegner zum 90. Geburtstag, ed. O. Brehm and S. Klie, 5–8. Antiqui­
tas, Reihe 3. Bonn: Habelt.

Bagnall, R. S. 1993. Egypt in Late Antiquity. Princeton: Princeton University Press.

———. 2005. “Linguistic Change and Religious Change: Thinking about the Temples of
the Fayoum in the Roman Period.” In Christianity and Monasticism in the Fayoum Oasis,
ed. G. Gabra, 11–19. Cairo: American University in Cairo Press.

———, and K. A. Worp. 2004. “Dating the Coptic Legal Document from Aphrodite.” ZPE
148: 247–252.

Butler, A. J., and P. M. Fraser. 1978. The Arab Conquest of Egypt and the Last Thirty Years
of the Roman Dominion, 2d ed. Oxford: Clarendon.

Cavenaile, R. 1987. “Le latin dans les milieux chrétiens dʼÉgypte.” Miscel-lània Papirològ­
ica Ramon Roca-Puig, 103–110. Barcelona.

Clackson, S. J. 2004. “Papyrology and the Utilization of Coptic Sources.” In Papyrology

and the History of Early Islamic Egypt, ed. P. M. Sijpesteijn and L. Sundelin, 21–44. Lei­
den: Brill.

Cribiore, R. 1996. Writing, Teachers, and Students in Graeco-Roman Egypt. American

Studies in Papyrology 36. Atlanta: Scholars Press.

———. 1999. “Greek and Coptic Education in Late Antique Egypt.” In Ägypten und Nubi­
en in spätantiker und christlicher Zeit, ed. S. Emmel, M. Krause, S. G. Richter, and S.
Schaten. Vol. 2, 279–286. Wiesbaden: Reichert.

Crum, W. E. 1927. “Some Further Melitian Documents.” JEA 13: 21–25.

———. 1939. A Coptic Dictionary. Oxford: Clarendon.

Daris, S. 1991. Il lessico latino nel greco dʼEgitto, 2d ed. Estudis de Papirologia i Filologia
bíblica 2. Barcelona: Institut de Teologia Fonamental.

Diethart, J. M., and K. A. Worp. 1986. Notarsunterschriften im byzantinischen Ägypten,

MPER N.S. 16. Vienna: Hollinek.

Förster, H. 2002. Wörterbuch der griechischen Wörter in den koptischen dokumen­

tarischen Texten, TU 148. Berlin: De Gruyter.

Förster, H., and F. Mitthof. 2004. “Ein koptischer Kaufvertrag über Anteile an einem Wa­
gen. Edition von P.Vat.Copt. Doresse 1,” Aegyptus 84: 217–242.

Foss, C. 2002. “The Sellarioi and the Other OYcers of Persian Egypt.” ZPE 138: 169–172.

Page 25 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
Fournet, J.-L. 1992. “Une éthopée de Caïn dans le Codex des Visions de la Fondation Bod­
mer.” ZPE 92: 253–256.

———. 1993. “Lʼinscription grecque de lʼéglise Al-Mu?allaqa.” Bulletin de lʼInstitut

(p. 449)

français dʼArchéologie Orientale 93: 237–244.

———. 2003a. “Langues, écritures, et culture dans les praesidia.” In La Route de Myos
Hormos: Lʼarmée dans le désert Oriental dʼÉgypte, ed. H. Cuvigny, 427–500. Fouilles de
LʼIFAO 48. Cairo: Institut français dʼarchéologie orientale.

———. 2003b. “Une lettre copte dʼAphrodité? (Revision de SB Kopt. I 290).” In Études
coptes VIII, ed. C. Cannuyer, 163–175. Lille-Paris: Association francophone de coptologie.

———. 2008. “Liste des papyrus édités de lʼAphrodité byzantine.” In Les archives de
Dioscore dʼAphrodité cent ans après leur découverte. Histoire et culture dans lʼÉgypte
byzantine, ed. J.-L. Fournet, 307–343.Études dʼarchéologie et dʼhistoire ancienne. Paris:
de Boccard.

———. forthcoming. “Sur les premiers documents juridiques coptes.” In Études coptes XI,
Treizième journée dʼétudes (Marseille, 7–9 juin 2007), ed. A. Boudʼhors and C. Louis.
Cahiers de la Bibliothèque copte 17. Paris: de Boccard.

Gascou, J. 2000. “Fragment dʼun codex juridique du Bas-Empire (P.Strasb. L 9).” In La

codification des lois dans lʼAntiquité, ed. E. Lévy, 285–291. Travaux du CERPEOGA 16.
Paris: de Boccard.

Ghiretti, E. 1996. “Note sul bilinguismo greco-latino dellʼEgitto romano.” Aevum 9: 275–
298.

Gignac, F. T. 1976. A Grammar of the Greek Papyri of the Roman and Byzantine Periods.
Vol. 1, Phonology. Milan: Istituto editoriale Cisalpino-La Goliardica.

Grohmann, A. 1952. From the World of Arabic Papyri. Cairo: Al-Maaref Press.

Habermann, W. 1998. “Zur chronologischen Verteilung der papyrologischen Zeugnisse.”

ZPE 122: 144–160.

Huyse, Ph. 1995. “Die mittelpersische Papyrologie: Fortschritte und Ziel einer jungen
Wissenschaft.” Indo-Iranian Journal 38: 357–367.

Kramer, J. 1984. “Testi greci scritti nellʼalfabeto latino e testi latini scritti nellʼalfabeto
greco: un caso di bilinguismo imperfetto.” In Atti del XVII Congresso Internazionale di Pa­
pirologia, vol. 3, 1377–1384. Naples: Centro internazionale per lo studio dei papyri er­
colanesi.

Krause, M. 1956. “Apa Abraham von Hermonthis: Ein oberägyptischer Bischof um 600.”
Ph.D. Diss., University of Berlin.

Page 26 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
———. 1980. “Koptische Literatur.” In Lexikon der Ägyptologie, vol. 3, 694–728. Wies­
baden: Harrassowitz.

Lewis, N. 1993. “The Demise of the Demotic Document: When and Why.” JEA 79: 276–
281.

MacCoull, L. S. B. 1985. “A Coptic Cession of Land by Dioscorus of Aphrodito: Alexandria

meets Cairo.” In Acts of the Second International Congress of Coptic Study, ed. T. Orlandi
and F. Wisse, vol. 7, 159–166. Rome: CIM

(
=L. S. B. MacCoull, Coptic Perspectives on Late Antiquity [Aldershot: Variorum, 1993],
no. VII).

———. 1986. “Coptic Egypt during the Persian Occupation: The Papyrological Evidence.”
Studi Classici e Orientali 36: 307–313

(
=L. S. B. MacCoull, Coptic Perspectives on Late Antiquity [Aldershot: Variorum, 1993],
no. XII).

———. 1991. “A Coptic Monastic Letter to Dioscorus of Aphrodito.” Enchoria 18: 23–25.

———. 1992. “More Missing Pieces of the Dioscorus Archive.” In Actes du IVe Congrès
copte, ed. M. Rassart-Debergh and J. Ries, 104–112. Publications de lʼInstitut orientaliste
de Louvain 41. Louvain-la-Neuve: Université catholique de Louvain, Institut orientaliste.

———. 1993. “The Apa Apollos Monastery of Pharoou (Aphrodito) and Its Papyrus
Archive.” Le Muséon 106: 21–64.

———. 1995. “Further Notes on Interrelated Greek and Coptic Documents of the
(p. 450)

Sixth and Seventh Centuries.” CdʼÉ 70: 341–353.

———. 1997. “Dated and Datable Coptic Documentary Hands before A.D. 700.” Le
Muséon 110: 349–366.

Mertens, P. 1987. “Les papyrus littéraires latins dʼauteurs classiques durant les deux
dernières décennies.” Miscel-lània Papirològica Ramon Roca-Puig, 189–204. Barcelona.

Moscadi, A. 1970. “Le lettere dellʼarchivio di Teofane.” Aegyptus 50: 88–154.

Orlandi, T. 1998. “Koptische Literatur.” In Ägypten in spätantik-christlicher Zeit: Ein­

führung in die koptische Kultur, ed. M. Krause, 117–147. Wiesbaden: Reichert.

Osing, J. 1998. Hieratische Papyri aus Tebtynis, vol. 1. Copenhagen: Carsten Niebuhr In­
stitute.

Quagebeur, J. 1982. “De la préhistoire de lʼécriture copte.” Orientalia Lovaniensia Periodi­

ca 13: 125–136.

Page 27 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
Richter, T. S. 2002. Rechtsemantik und forensische Rhetorik: Untersuchung zu
Wortschatz, Stil, und Grammatik der Sprache koptischer Rechtsurkunden. Kanobos 3.
Leipzig: Wodtke und Stegbauer. Reprinted with a new introduction, Wiesbaden: Harras­
sowitz, 2008.

Rochette, B. 1990. “Les traductions grecques de lʼÉnéide sur papyrus: Une contribution à
lʼétude du bilinguisme gréco-romain au Bas-Empire.” Études Classiques 58: 333–346.

———. 1996. “Papyrologica bilinguia Graeco-latina.” Aegyptus 76: 57–79.

———. 1997. Le latin dans le monde grec: Recherches sur la diffusion de la langue et des
letters latines dans les provinces hellénophones de lʼEmpire romain. Collection Latomus
233. Brussels: Latomus.

———. 1999. “Écrire en deux langues: Remarques sur le mixage des écritures grecque et
latine dʼaprès les papyrus littéraires bilingues dʼauteurs classiques.” Scriptorium 53: 325–
334.

Sänger, P. 2008. “Saralaneozan und die Verwaltung Ägyptens unter den Sassaniden.” ZPE
164: 191–208.

Schubert, P. 2002. “Contribution à une mise en contexte du Codex des Visions.” In Le

Codex des Visions, ed. A. Hurst and J. Rudhardt, 19–25. Geneva: Librairie Droz.

Sijpesteijn, P. J. 1992. “The Meanings of ἤτοι in the Papyri.” ZPE 90: 241–250.

Thomas, J. D. 1998. “P.Ryl. IV 654: the Latin Heading”, CdʼÉ 73: 125–134.

Torallas Tovar, S. 2004. “Egyptian Lexical Interference in the Greek of Byzantine and Ear­
ly Islamic Egypt.” In Papyrology and the History of Early Islamic Egypt, ed. P. M. Si­
jpesteijn and L. Sundelin, 163–198. Leiden: Brill.

Turner, E. G. 1961. “Latin versus Greek as a Universal Language: The Attitude of Dioclet­
ian.” In Language and Society: Essays Presented to Arthur M. Jensen on His Seventieth
Birthday, 165–168. Copenhagen: Berlingske.

Van Minnen, P. 1994. “The Roots of Egyptian Christianity.” APF 40: 71–85.

Weber, D. 1991. “Ein bisher unbekannter Titel aus spätsassanidischer Zeit?” In Corolla
Iranica: Papers in Honour of Prof. Dr. David Neil MacKenzie on the Occasion of His 65th
Birthday, ed. R. E. Emmerick and D. Weber, 228–236. New York: P. Lang.

———. 1992a. “Pahlavi Papyri Revisited.” In Proceedings of the XIXth International Con­
gress of Papyrology, vol. 2, 493–508. Cairo: Ain Shams University.

———. 1992b and 2003. cf. CII.

Wipszycka, E. 1992. “Le nationalisme a-t-il existé dans lʼÉgypte byzantine?” Journal of Ju­
ristic Papyrology 22: 83–128

Page 28 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
(
=E. Wipszycka, Études sur le christianisme dans lʼÉgypte de lʼAntiquité tardive. Studia
Ephemeridis Augustinianum 52. Rome: Studia Ephe-meridis Augustinianum, 1996, I.1).

Worp, K. A. 1984. “Studien zu spatgriechischen, koptischen und arabischen Pa­

(p. 451)

pyri.” Bulletin de la Société dʼArchéologie Copte 26: 99–107.

Wouters, A. 1988. The Chester Beatty Codex Ac. 1499: A Graeco-Latin Lexicon on the
Pauline Epistles and a Greek Grammar. Leuven-Paris: Peeters.

Zuckerman, C. 1995. “The Hapless Recruit Psois and the Mighty Anchorite, Apa John.”
BASP 32: 183–194.

Notes:

(1.) Diglossia means here the coexistence in a single society of two languages distin­
guished by their function—a phenomenon not identical to bilingualism. For the applica­
tion and development of this concept see Adams (2003, 537–541).

(2.) For the dates see Altheim-Stiehl (1991, 1992a, 1992b).

(3.) The few papyri with both Greek and Pehlevi are actually reuses of Greek documents
by the Persians.

(4.) For the first, cf. CII, P. 149; for the second, CII, P. 140 (with commentary on pp. 23–
24).

(5.) The Latin papyri have been collected in the following corpora: C.Pap.Lat. gathered
documents and literary texts; C.Epist.Lat. is devoted to letters; finally and most impor­
tant, Ch.L.A. assembles texts that contain Latin, whether published or not, by the country
of collection. The numbers that I give here are based on the Heidelberger Gesamtverze­
ichnis der griechischen Papyrusurkunden Ägyptens (http://www.rzuser.uni-heidelberg.de/
~gvo/gvz.html): They exclude about 430 texts that are undated or of non-Egyptian prove­
nance (like the papyri of Dura on the Middle Euphrates or the ostraca of Bu Njem in
Libya).

(6.) See Fournet (2003a, 431–446).

(7.) P.Abinn., SB XIV 11380, and perhaps P. Gen. I 80, SB VI 9605, X 10755, and XX 14954.

(8.) Here Adams modifies the opposition developed by C. A. Ferguson in his definition of
diglossia between low language and high language (Adams 2003, 538). Adams rightly
prefers to see Latin as a “super-high” language and Greek as a high language.

(9.) SB XX 14726 (Arsinoite, 399), letter of the pretorian prefect to the governor of Arca­
dia (accompanying an order in Greek).

Page 29 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
(10.) For example, PSI I 1112 (316), P.Lips. I 44 (324–337), Ch.L.A. XVII 657 (436–450), SB
XX 14606 (425–450), with the end of a rescript that concludes with the autograph signa­
ture of Theodosius II.

(11.) A list of these texts is given in P.Oxy. LI, pp. 47–48 and Thomas (1998, 132–134). See
also Rochette (1997, 119–120).

(12.) For example, in Ch.L.A. XVIII 660, a list of soldiers of the ala III Assuriorum (326 or
329), we read in II 2: prigceps I turmηs in place of princeps I turmae (cf. Adams 2003,
618–621).

(13.) The graph is based on the Leuven Database of Ancient Books, http://ldab.arts.
kuleuven.be/ldab_text.php.

(14.) Byzantine notarial signatures are collected by Diethart and Worp (1986). Kramer
(1984, 1382) studies those in Latin letters briefly.

(15.) Collected in C.Gloss.Biling.

(16.) The volumes of Coptic documentary papyri are included in the Checklist (http://
scriptorium.lib.duke.edu/papyrus/texts/clist.html). There is also now the Banque de don­
nées des textes coptes documentaires by Alain Delattre (http://dev.ulb.ac.be/philo/bad/
copte/base.php?page=accueil.php). The Leuven Database of Ancient Books now includes
Coptic literary papyri. For Coptic literature see Krause (1980) and Orlandi (1998). An
electronic corpus of Coptic manuscripts is being developed under the direction of T. Or­
landi (http://rmcisadu.let.uniroma1.it/~cmcl/).

(17.) Greek words in Coptic documentary texts have been collected by Förster (2002; 914
pages).

(18.) In parentheses I give the abbreviations used in the dictionary of Crum (1939).

(19.) The ecumenical council of Chalcedon (451), by condemning the theological positions
of Dioscoros, the bishop of Alexandria, created a schism between partisans of the Chal­
cedo-nian creed (supported by the emperor) and those of Dioscoros (anti-Chalcedonians
or monophysites).

(20.) The graphs are based, for the Greek documents, on Habermann (1998, 147) and, for
the Coptic, on the Banque de données des textes coptes documentaires of Alain Delattre
(see note 16).

(21.) The data in this graph should be considered with some reservation: Many papyri
have only approximate dates. Thus, I have interpreted “III/IV” as “third or fourth” rather
than “end of the third, beginning of the fourth.” Moreover, many documents were pub­
lished with no title or only a vague one.

(22.) Greek texts: P.Lond. VI 1913–1919; Coptic: P.Lond. VI 1920–1922, and Crum (1927).

Page 30 of 31
The Multilingual Environment of Late Antique Egypt: Greek, Latin, Coptic,
and Persian Documentation
(23.) Greek texts: P.Neph. 1–14, 17–42; Coptic: P.Neph. 15, 16.

(24.) Greek texts: P.Herm. 7–10, 17, P.Amh. II 145 (with signature in Coptic), P.Lond. III
981 (p. 241); SB XVIII 13612; Coptic texts: P.Ryl.Copt. 268–274, 276, and perhaps 275,
292, 301, 310–314, 396, P.Lond.Copt. I 1123. On this archive see Zuckerman (1995) and
Clackson (2004, 24–25).

(25.) Palladius, Historia Lausiaca, 35, and Historia monachorum in Aegypto, 1.

(26.) Greek texts: P.Kell. I (documents and subliterary texts), III (lsocrates), IV (account
book), O.Kellis (documents), and some literary texts in vol. II (G. 91–94); Coptic texts: P.
Kell. II and VI (literary), V (documents). It should be noted that some of these texts, al­
though found at Kellis, were sent from the Nile valley.

(27.) Greek texts: principally P.Cair.Masp. I–III, P.Flor. I 93, III 279–298, 342, P.Lond. V
1660–1718; Coptic texts: see a provisional list in Fournet 2003b, 170–175. For a complete
list of the texts composing this archive and the other archives related to it, see Fournet
(2008, 307–343). On the Coptic archive of Dioscoros, see Clackson (2004, 25–28).

(28.) P.Cair.Masp. III 67353 recto, copy of the document edited by MacCoull (1985), the
first dated Coptic text (569), and P.Lond. V 1709. About these texts (and another still un­
published) see Fournet (forthcoming). On the arbitrations, cf. P.Mich.Aphrod.

(29.) This archive also includes an older group (493–557/558), which is indirectly connect­
ed to it. Greek texts: P.Lond. V 1719–1737, 1846–1850, 1852–1859, 1861, P.Münch. I 1–16;
Coptic texts: O.CrumST 439, SB Kopt. III 1293, 1394, 1395, 1666, and perhaps
P.Lond.Copt. I 446 and O.CrumST 116, 423. On the Coptic side of this archive, see Clack­
son (2004, 28–29).

(30.) Coptic texts: Krause (1956); Greek text: P.Lond. I 77 (p. 231).

(31.) Greek texts: P.Par. 20; 21; 21bis; 21ter, SB l 4503–4505; 5285–5287; Coptic texts:
CPR lV 23; perhaps 83.

(32.) For the dates of P.Vat.Copt., still unpublished except the first one (Förster and Mit­
thof 2004), see Bagnall and Worp (2004). On some of these texts, see MacCoull (1997).
P.CrumST 48 and O.CrumVC 5 can be dated owing to their link with SB I 5112.

(33.) See Gignac (1976, 332–333).

Jean-Luc Fournet

Jean-Luc Fournet is Directeur dʼÉtudes en papyrologie grecque, École Pratique des

Hautes Études, Section des sciences historiques et philologiques, Paris.

Page 31 of 31
Arabic Papyri and Islamic Egypt

Arabic Papyri and Islamic Egypt

Petra M. Sijpesteijn
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0019

Abstract and Keywords

The Arabs enjoyed a long-standing acquaintance with papyrus and its benefits. Papyrus
and other traditional media, such as ostraca, leather, parchment, textiles, stone, and
bone, were already fully in use on the Arabian Peninsula in pre-Islamic times. After pro­
viding a brief history of Arabic papyrology, this article discusses the reading and publish­
ing Arabic texts; Arabic literary and subliterary papyri; the Islamic narrative tradition;
and Arabic documentary papyri.

Keywords: Arabic documentary papyri, Islamic narrative, Arabian Peninsula, Arabic papyrology

Introduction: A History of Arabic Papyrology

Although they borrowed their words for it from the Greek—bardī, abardī, fāfīr, and chartis
(transformed into qirṭās)—the Arabs enjoyed a long-standing acquaintance with papyrus
and its singular benefits. As references in the Qurʼān(Q 6:7, 91)and in pre-Islamic and
early Arab poetry show, papyrus (presumably imported from Egypt) and other traditional
media, such as ostraca, leather, parchment, textiles, stone, and bone, were already fully
in use on the Arabian Peninsula in pre-Islamic times. Among the sheets and other writing
material used to record God's words as revealed to Muhammad, later to become the
Qurʼān, were certainly also papyri, 1 but none of this material has survived.

Papyrus cultivation and use were unaffected by the Muslim conquest, and Egypt's new
rulers even continued to export it to the Byzantine territories and to Italy. Indeed, produc­
tion was expanded in the ninth century, when plantations were developed in Sicily and
Iraq, the latter presumably to feed the ever-expanding bureaucratic appetite of the Ab­
basid chancellery in the caliphal capital, Baghdad.2 And just as the Muslims adopted pre-
Islamic administrative habits, so they also took on pre-Islamic languages: Not only did pa­
pyrus itself continue to be used as before, but Greek was maintained in (p. 453) both the

Page 1 of 22
Arabic Papyri and Islamic Egypt

official and the private spheres at least until the end of the eighth century and Coptic un­
til the eleventh (CPR XXII, introduction).3

The first two known Arabic papyri date to a mere twenty-two years after Muḥammad's hi­
jra from Mecca to Medina in 622 CE; written during the conquest of Egypt, they record
deliveries of matériel to the conquering Arab armies.4 From that moment on, Arabic pa­
pyri appear in steadily larger numbers and span the same wide topical range that we
know from Greek and Coptic documents.

Not until the introduction of paper making, which was allegedly brought about by the
capture of Chinese paper makers at the battle of Talāz in 751 (Bloom 2001, 42–45), did
the ascendancy of papyrus face a significant challenge. By the early tenth century, al­
though papyrus continued to be used in Iraq, paper began to supersede it, as paper mills
began springing up in Damascus, Tiberias, and Tripoli; by the twelfth century, they had
spread to Morocco and Spain as well (Grohmann 1952, 53). The first paper documents
found in Egypt date between 796 and 816, 5 but here the papyrus industry continued into
the second half of the eleventh century, when the last dated papyri were written.6 The last
papyri preserved in Italy are from the same period.7

Although Arabic papyri are well represented in papyrus collections around the world, the
study of Arabic papyrology is younger and less well developed than that of its Graeco-Ro­
man counterparts, having been launched only in 1824, when a small sealed jar containing
three Arabic papyrus texts was found near the Saqqāra pyramid by a team of French ex­
cavators. With the publication of two of these papyri a year later by Silvestre de Sacy
(1758–1838), the discipline of Arabic papyrology was born(1825, 1831).

Even though no archaeological missions were dispatched solely to search for Arabic pa­
pyri, like those that resulted in the great discoveries of Greek papyri at the end of the
nineteenth and the beginning of the twentieth century, Arabic papyri nonetheless found
their way into most of the world's major papyrus collections, and today there are some ex­
clusively Arabic papyrus collections (Sijpesteijn 2003, 2006; Grohmann 1952). Altogether
an estimated 150,000 are extant today (including parchment and paper documents that in
some cases date into the modern period), and more are being unearthed every year (Si­
jpesteijn 2006; Rāghib 1996, 2). The major caches, however, have come from only a hand­
ful of sites in Egypt (e.g., the Fayyum, Ihnās (Herakleopolis), Ashmūnayn (Hermopolis),
Akhmīm (Panopolis), Edfu (Apollonopolis Magna), and Ishqaw (Aphrodito)), while some
Arabic papyri and ostraca have also come to light in Fusṭāṭ. Outside Egypt, significant
finds have been made in the Negev desert (P.Ness. III) and in Khirbet Mird near the Dead
Sea (P.Mird), as well as outside Damascus and in Samarrāʼ in Iraq (Grohmann 1952,11).

After the 1824 texts, Silvestre de Sacy published two more Arabic papyri (1833), but
thereafter no substantial work was conducted for some fifty years, until the arrival of two
of Arabic papyrology's greatest practitioners: Joseph Karabacek (1845–1918) and Adolf
Grohmann (1887–1967). Karabacek's name is connected in particular with the papyrus
collection of Archduke Rainer at Vienna, of which he served as director. Grohmann edited
Arabic papyri from many collections in Europe and Egypt and also (p. 454) wrote several
Page 2 of 22
Arabic Papyri and Islamic Egypt

Arabic papyrological handbooks (e.g. P.Cair.Arab.1–VI, 1952, 1966). Other names of note
from this “golden age” include C. H. Becker (1876–1933), who published some important
early papyri from the correspondence of the governor Qurra b. Sharīk (in office 709–715),
and several historical studies based on these texts (P.Heid.Arab.I, 1902–1903, 1906,
1911). In England, another important name from this period is that of D. S. Margoliouth
(1858–1940), who worked mostly with manuscripts and published an edition of the Arabic
papyri preserved at the ManchesterUniversity library (P. Ryl. Arab.). These “founding fa­
thers,” however, left few successors, and the discipline had to wait until the 1950s before
their legacy bore fruit in the work of, most notably, Claude Cahen (1909–1991), Albrecht
Dietrich (1912–), G. Levi Della Vida (1886–1967), and Nabia Abbott (1897–1981).

Again around thirty years passed before a third generation of Arabic papyrologists
emerged to pick up the torch. Although continuing to focus on texts, this third wave has
been characterized by two new developments: the systematic use of Arabic papyri for lin­
guistic studies and the gathering of texts according to themes, which had been done to
some extent before, especially to reaggregate archives, but on a much less ambitious
scale. With the work of Gladys Frantz-Murphy (1944–) (CPR XXI), Werner Diem (1944–)
(P.Vindob.Arab. I–V; CPR XVI), Geoffrey Khan (1958–) (1990, 1994), Raif Georges Khoury
(1936–), and Yūsuf Rāghib (1941–) (P.Marchands I, II, III, V/1; P.Vente), Arabic papyrology
in the early twenty-first century seems finally to have achieved sustainable momentum.

The issue of training for Arabic papyrologists is also receiving welcome attention. With
the founding of the International Society for Arabic Papyrology (ISAP) in 2001, the con­
ferences and workshops it organizes, and the initiatives it supports, progress has been
made in setting Arabic papyrology on the road to becoming an established and recog­
nized discipline (http://www.ori.unizh.ch/isap.html). The ISAP has organized Arabic papy­
rology workshops in Europe, and online Arabic papyrological training is offered at the
University of Zurich's Arabic papyrology school (http://orientx.unizh.ch:8080/aps_test_2/
home/index.jsp). The multilingual diversity of the early Islamic world is also being ad­
dressed through cross-disciplinary initiatives and studies as Arabic papyrology builds
bridges to other fields and scholars become increasingly alert to the advantages of study­
ing contemporary documents together, even if written in different languages.

The incorporation of documents written on paper, parchment, leather, and other materi­
als makes the chronological boundaries of Arabic papyrology difficult to draw. While ac­
knowledging the specific skills involved and the knowledge necessary to read papyri (as
opposed, for example, to Mamlūk paper documents), all documents—from Andalusia to
Khurasan—written before the arrival of the Ottomans in Egypt in 1517 are generally con­
sidered to belong to the field of Arabic papyrology.8 The Checklist of Arabic Documents
(http://www.ori.unizh.ch/isap/isapchecklist.html) accordingly reflects the wider chronolog­
ical and geographical (p. 455) limits of the Arabic documents. For practical purposes, how­
ever, I have concentrated on papyrus documents from Egypt.

Page 3 of 22
Arabic Papyri and Islamic Egypt

Reading and Publishing Arabic Texts

Besides the problems well known to papyrologists of all times and places—the damaged,
fragmentary state of the papyri and the lack of archaeological and historical context for
most documents—Arabic papyrology has labored under additional challenges.

First are philological peculiarities that demand special technical attention. Arabic papyri
and paper documents do not use diacritical dots in the same systematic manner as do
printed texts, thereby making decipherment often extremely difficult. Vowel signs are al­
so absent, and the letter hamza is generally not written, coinciding with the manuscript
tradition. Most orthographic and linguistic peculiarities of the Arabic papyri have been
described by Hopkins in his grammar (1984). On top of this, Arabic papyrology still lacks
most of the basic hermeneutic tools—such as dictionaries, lexicographical works, and
onomastica—that papyrologists rely on in other languages. A crucially important tool for
the editing of documents is the expanding Arabic papyrology database, a searchable on­
line database of edited Arabic papyri (http://orientx.unizh.ch:9080/apd/project.jsp), which
will be accompanied by a dictionary of terms that appear in the papyri.

The publication of Arabic papyri editions generally follows the “Leiden method” in the use
of brackets. Uncertainly read letters cannot be indicated by dots underneath them as is
done in Greek papyrology, and even though it has been suggested that horizontal super­
strokes be used for this purpose, no generally accepted alternative has been developed as
of yet.

Arabic papyrus texts, following late Byzantine practice, are generally written first on the
side with vertical fibers, and then on the side with horizontal ones. The text on early Ara­
bic papyri is generally very spaciously arrayed, with large, elongated letters and large
spaces between the letters, which sometimes results in words being broken off at the end
of the line and continued on the next. From the ninth century on, the script becomes
smaller and more cursive and, especially in private letters, can spill over into the upper
and side margins, with lines being written perpendicular or even upside down relative to
the main text. In later paper documents the lines often curve upward at the end of the
page as a result of changed aesthetic conventions. The implications of this evolving lay­
out and the other physical features of the Arabic documents still await full scholarly treat­
ment.

The convention of folding and sealing Arabic papyrus letters continues pre-Islamic prac­
tice: The names of the sender and addressee appear on the back side of the written
(p. 456) text (or in the top margin of the main text), on the right and left side of the pa­

pyrus respectively, with a space in between to allow a string and clay seal to be attached.
The layout of the address is continued in later paper letters, when the practice of sealing
has been abandoned but when letters still seem to be folded in the same manner. The or­
der of the sender and addressee can be reversed when the addressee is of a higher rank
than the sender. The addresses are often accompanied by instructions for delivery of the
letter and may mention mosques, churches, markets, and streets by name, as well as

Page 4 of 22
Arabic Papyri and Islamic Egypt

houses and shops located in them, thereby providing unique information on the urban
structure of towns and villages.

Arabic Literary and Subliterary Papyri

Largely because most Arabic papyrologists have come to the field as historians or lin­
guists, Arabic literary papyri have tended not to receive the same level of attention as
documentary texts. Following the pioneering work of Abbott, who edited a wide variety of
literary fragments from the Oriental Institute collection at the University of Chicago,
David-Weill (1939) and Khoury (1972, 1986) have published some important manuscripts
in their entirety. Nevertheless, the number of literary papyri that remain unedited and un­
studied is significant, including many that antedate or are completely unknown from the
manuscript tradition.9

For the Qurʼān, the earliest surviving papyrus fragments come from the eighth-ninth cen­
turies and are datable on the basis of their palaeography to more than a century after the
third caliph, ʻUthmān(r. 644–656), compiled the standard-version Qurʼān, according to the
Islamic narrative tradition.10 The absence of any securely datable maṣḥaf (copy of the
Qurʼān) or fragment thereof from this early period has fuelled discussion about both the
date and place of the Qurʼān's formation (Hawting and Shareef 1993), as well as the de­
gree of linguistic and orthographic variation among its early redactions. Qurʼānic verses
also appear on amulets and in prayers for private use and are quoted in letters on pa­
pyrus and paper, but these are generally difficult to date.

The papyri also preserve other well-known texts, such as the earliest version of a frag­
ment from the maghāzī rasūl (the prophet's campaigns) and the ḥadīth Dāwūd (actions
and words of David), dating to the ninth century.11 of the numerous prophetic ḥadīth
fragments that survive, few have been edited, and many even remain unidentified. Two
large early ḥadīth collections—those of the Egyptian ḥadīth scholars ʻAbd Allāh b. Laḥīʼa
(d. 790) (Khoury 1986) and Ibn Wahb (d. 812) (David-Weill 1939)—have been preserved
on a ninth-century papyrus roll and (p. 457) codex, respectively. In these early versions,
one can trace the development of isnāds (chains of transmitters) and the structure of the
stories told in them, especially where it is possible to compare these with later sources.
The processes of ordering and categorizing traditions, the formation of the canonical
ḥadīth collections, and especially the process by which certain ḥadīths were excluded can
all be traced through the papyri and suggest promising lines of inquiry.

Other important literary survivals are the earliest fragment of the Thousand and One
Nights on paper (Abbott 1949) and the earliest preserved Arabic astronomical text on a
ninth-century papyrus.12 Poetry is represented by a large unedited panegyric for the
prophet's son-in-law and nephew, the fourth caliph, ʻAlī (r. 656–61), and verses from an
ode by Dhū al-Rummah (d. 735–736), now housed in the University of Michigan and Ori­
ental Institute (Abbott 1972, doc. 7). Significant fragments of historical, grammatical, and
legal works have also been preserved on papyrus (Abbott 1957, 1967, 1972).

Page 5 of 22
Arabic Papyri and Islamic Egypt

Of the many scribal exercises preserved in papyrus collections, only a few have so far
been published (e.g., CPR XVI 35), despite the rich potential of these documents to pro­
vide important insights into the school systems of the time, the levels of literacy preva­
lent, and the kinds of texts deemed culturally valuable or educationally suitable. Of a re­
lated nature, we also have what seem to be aides-mémoire for public presentations or
notes written down at lectures, offering an interesting insight into how the Islamic em­
phasis on mnemonic retention and oral transmission was adapted for practicability.

In addition, literary fragments, along with autograph manuscripts, through their dele­
tions, repetitions, and corrections, often open a valuable window onto the processes of
writerly practice, as well as the time and circumstances of composition. Scribal mistakes,
such as the accidental loss of identical adjacent syllables (haplology), erroneous repeti­
tion of a word, phrase, or combination of letters (dittography), and writing once what
should be written twice (haplography), all shed light on the process of copying manu­
scripts, while marginal notes and corrections offer hints about collocation with earlier
manuscripts or verification by an authoritative teacher or author.

Also well represented in papyrus collections, amulets and private prayers are another
genre of texts that is still in need of proper scholarly attention.13 A detailed study of
amulets could tell us much about chronological changes in the popularity of certain reli­
gious customs, as well as the interaction between so-called popular cults and official rites
and between the three monotheistic faiths. Medical recipes and requests for medicines
similarly offer insights into medical practice (e.g., CPR XVI 24; Dietrich 1954).

When studied together, literary texts have much to offer on textual production and con­
sumption, as well as issues of literacy and the circulation of literary works. And even
small fragments can be useful here by offering clues, mapping their finding places to re­
gional preferences, and supplying evidence on the relationship between codices and rolls
as media for literary production—topics about which hardly anything is known from the
Islamic period (Grohmann 1966, 65–67).

(p. 458) The Islamic Narrative Tradition

Medieval Islam is blessed with an exceptionally deep and well-developed store of narra­
tive sources—the countless Arabic chronicles, legal, religious, and literary works that
have long shaped our view of this period. It is the interplay and exchange with this paral­
lel body of material, as well as the potential it creates for the Arabic papyri to provide a
check or control on its evidence, that gives Arabic papyrology much of its vitality. But to
understand how papyri can complement and enrich the evidence of the Islamic narrative
tradition, we need first to consider some of the specific historiographical problems inher­
ent in that tradition.

The canonical Islamic historical works, while based on earlier oral and written traditions,
make their appearance only at the end of the eighth and the beginning of the ninth centu­
ry—that is, two hundred years after the rise and formation of Islam. This extended time
Page 6 of 22
Arabic Papyri and Islamic Egypt

lapse, combined with a cultural emphasis on the oral transmission of historical informa­
tion, necessarily left its mark on how ideas and events were conceived of and presented.
Even more important, by the time these texts were written, notions about the origins and
formation of Islam had become formalized and standardized, resulting in a historical or­
thodoxy that makes alternative narratives very difficult to retrieve.

This process of standardization was further compounded by the homogeneity of

historiography's practitioners. Almost exclusively urban, bourgeois, and male, most of
them also lived and worked under Islam's second dynasty, the Abbasids. Not only did they
absorb and propagate the Abbasid anti-Umayyad prejudice (the Abbasids had seized pow­
er from the Umayyads in 750 and moved the caliphal capital from Damascus in Syria to
Baghdad in Iraq), but their focus tends to be on the eastern part of the Islamic empire,
where most of them were active.

How then can the papyri correct or adjust the view these sources offer?

Arabic Documentary Papyri

Historical Events

The clearest cases occur where an historical event described ambiguously in the narra­
tive sources can be checked or corrected against the papyri. The battle of Badr, to take
but one example, to which the narrative sources have ascribed contradictory dates, may
now be securely placed in Jumāda II of the second year of the hijra (November 623),
thanks to a seventh-century papyrus from Palestine (P.Mird 71; Cook and Crone 1977,
160n56). Additionally, the papyri can also help us to reassess (p. 459) impressions or per­
ceptions and even wholesale interpretations. For example, papyrological evidence of the
diligence and conscientiousness of his administration allows us to revisit the (mostly Ab­
basid) narrative tradition of the Umayyad governor Qurra b. Sharīk as an unjust tyrant
(Sundelin 2004, 8, where earlier references can be found).

Language and Script

Despite the attempts of early papyrologists to link them, the writing of the papyri does
not fit the canon of seven well-balanced scripts famously ascribed to Ibn Muqla (d. 990);
while parallels between the script of the papyri and that used for chancellery documents
and bookhands can be determined in individual cases, the palaeography of the documents
in general follows its own path, and a detailed and well-chronicled study of Arabic
palaeography in the documents is still lacking.14 Nevertheless, thanks to the probings in­
to Arabic orthography of Simon Hopkins (1950–) (1984) and Werner Diem (1979–1981),
among others, we now have a fairly detailed picture of the papyri's “Middle Arabic” and,
through its mistakes and hypercorrections, some idea of the spoken language of the time
(Hopkins 1984).15

Page 7 of 22
Arabic Papyri and Islamic Egypt

Language in a more general sense and in particular multilingualism also help us to under­
stand the process of Arabization. The continued use of Greek and Coptic in the adminis­
tration (discussed later) contradicts the assertion in the Arabic narrative sources that the
Caliph ‘Abd al-Malik (r. 685–705) imposed the exclusive use of Arabic on the chancellery
in the year 700 (Balādhurī, Futūḥ al-buldān, 193), an order said to have been implement­
ed in Egypt by his governor, ʻAbd Allāh (in office 705–709) in 706 (Kindī, Kitāb al-Wulāt,
58–59). The first known Arabic documents are all related to the administration and ap­
pear in bilingual papyri together with Greek translations. They also continue the Byzan­
tine practice of having each papyrus roll start with a thicker sheet that gives the names
ofthe reigning emperor, the governor, and sometimes the financial director under whom
the roll was produced, as well as the date of manufacture and name of the papyrus mill.
The earliest bilingual Arabic-Greek document of this type was found in Nessana and men­
tions (in place of the emperor) Caliph Muʻāwiya (r. 661–680); the first entirely Arabic pro­
tocols do not appear until 732.16 Again, the continued use of Greek in protocols runs
counter to the tradition preserved in Arabic narrative sources that ʻAbd al-Malik, angered
over the “Coptic” practice (not, incidentally, corroborated by the papyri) of putting a
cross and the name of Christ at the top of protocols, replaced these with Arabic religious
phrases (Balādhurī, Futūḥ al-buldān, 240). Other bilingual administrative documents in­
clude a receipt written in 643 for sixty-five sheep delivered to an Arab army unit; the late
seventh-century demand notes for oil and wheat found in Nessana; and demand notes for
workers, food, and other products for the Muslim rulers from the governorship of Qurra
b. Sharīk (in office 709–715).17

Entirely Arabic documents increase in number from the first quarter of the eighth
(p. 460)

century on. Even though Greek and Coptic documents continued to be produced in large
quantities, as discussed earlier, the tenth-century apocalypse of Samuel of Qalamūn (Zi­
adeh 1915–1917) provides evidence that an increased use of Arabic was considered a
threat to the native Egyptian language. Just what influence Arabic writings had on the
use of Greek and Coptic is, however, not clear. In some tenth-century legal documents
from Tebtynis in the Fayyum, one of the parties to the transaction has the contents “ex­
plained in the foreign language,” presumably Coptic.18 The last datable Coptic documents
are a series of legal documents, letters, and lists of the eleventh century.19 While Coptic
continued to be used as a liturgical language, Christian literary production, including the
Bible, switched to Arabic from the tenth century on.

Arabization can also be traced through the phenomenon of linguistic interference. The
use of Arabic loanwords and expressions in Coptic texts is attested in documents from the
seventh century on but was insignificant until the ninth century and remained mostly lim­
ited to technical administrative and medicinal terms (Richter 2004, 107–112). Arabic, by
contrast, shows a steady development from the importation of Greek words (often via
Coptic immediately following the conquest) to the introduction of Persian words in the
Abbasid period, when Persian officials arrived in Egypt.20 In the Islamic period certain
pre-Islamic Greek honorifics continued to be used for Christian and even Muslim officials,
though whether this indicates Muslim appropriation or assimilation or perhaps an unin­
terrupted Greek scribal tradition is unclear. As for place names, at least in the early
Page 8 of 22
Arabic Papyri and Islamic Egypt

stages, Greek forms persisted in Arabic texts, though from the eighth century on they be­
gan to be replaced by their Coptic variants, which frequently referred to an ancient
Egyptian toponym (Grohmann 1959). What these changes from Greek to Coptic/Egyptian
and/or Arabic place names signify deserves to be examined in more depth.

Significant interference is also apparent in the borrowing of Greek terms for Arabic tech­
nical, administrative, and managerial concepts, a practice that forces us to ask what this
meant for the Egyptian perception of these newly introduced concepts and functions and
what this tells us about Arabic rule.21 The introduction into Arabic of Greek and Latin
terms that had been out of use for several centuries in Egypt or had never been part of
the Egyptian Byzantine vocabulary raises further questions about the provenance of Is­
lamic administrative practice.22

An Arabic Legal Tradition

The early evolution of Islamic law is notoriously difficult to reconstruct from the Arabic
literary sources, where it remains overshadowed by the “classical system” presumed by
later jurists to have existed from Islam's beginning. The development of Islamic law and
its relation to its predecessors, however, is not totally lost; it can be traced to some extent
through the development of Arabic legal terminology. (p. 461) Ignored by the narrative
sources, this subject touches upon topics of integration, conversion, Arabization, immi­
gration, and authority.

The earliest Arabic papyri reflect a legal system different from indigenous Egyptian prac­
tice (Khan 1994, 2005). A similar distinction between the local and newly introduced Arab
legal tradition can be observed in Umayyad and early Abbasid documents found in the
eastern Islamic empire in Sogdia and Bactria (Khan 1994, 2005). In bilingual Greek-Ara­
bic documents, the Arabic is not a translation of the Greek portion, but has instead its
own features and traditions, and a different scribe is named at the end of the Greek and
the Arabic texts. In the ninth century, however, Arabic papyri begin to exhibit influences
from the local, Greek/Coptic legal practice, with expressions and words originating in the
eastern Islamic Empire, and witnessed there in earlier documents, starting to find their
way into the documents (Khan 2005).

Behind these issues lies the larger question of the relationship between legal theory and
practice, for which the documents again offer uniquely useful evidence. In Islamic law, for
example, documents do not constitute evidence (cf. Q 2: 282), but as their ubiquity shows,
they were obviously an essential force in day-to-day business and transactions of all sorts.
This raises questions about how documents acquired contractual force in an environment
of minimal enforceability, how written records functioned in a largely oral society, and
how those with low levels of literacy interacted with the judicial and bureaucratic system.
Other documents reveal how certain legal prohibitions, such as the ban on usury, were
circumvented by permissible legal transactions, such as future sale (CPR XXVI 17, intro­
duction).

Page 9 of 22
Arabic Papyri and Islamic Egypt

For a later period, Christian and Jewish legal documents written in (Judeo-) Arabic use
the same formulas as Islamic ones, even though they are sometimes drawn up according
to different legal principles (cf. Abbott 1941). This process of reciprocal influence on the
formulas of the papyrus documents needs to be examined in more detail, as it does for ex­
tralegal documents, such as official and private letters, and tied in with what we know
about historical events and circumstances.23

Administrative History

One of the most vexing questions about early Islamic administration is the degree of bu­
reaucratic and administrative sophistication and experience that the Arabs brought to
their empire and, by extension, the degree of continuity and change in the conquered
provinces. Straightforward numerical insufficiency on the conquerors' part ruled out a
top-to-bottom restaffing of the Byzantine administration and meant that only the highest
central offices were at least initially taken over by Muslims. The continuity in administra­
tive personnel at the lower levels of the administration, necessary to ensure the uninter­
rupted collection of tax revenues, also accounts for the overwhelming sense of continuity
in the daily life of the indigenous population that comes through in the documents. Never­
theless, this (p. 462) continuity should not be allowed to conceal certain important
changes that also took place.

Not only were the administrative districts reorganized (Sijpesteijn 2007b), but within a
generation of their takeover, the Arab conquerors imposed the collection of a poll tax, an
adaptation of the Qurʼānic jizya (Q 9: 29) but framed according to the new Greek term an­
drismos (CPR XXII 1). We should probably view this melding of selective change and un­
derlying continuity not as an Arab failure to offer an alternative to the Byzantine system
but rather as an intentional and focused adaptation of the system to the conquerors' own
needs and traditions (Sijpesteijn 2007a, 2007b, forthcoming-a). Bureaucracy, indeed,
seems to have been an area of special interest to the Arabs, as the net increase in admin­
istrative documents that resulted from the conquest amply testifies.

The documentary legacy of the Arab army is thinner than for prior armies that came
through Egypt, though perhaps much has been lost beneath Fusṭāṭ, the Muslim-founded
garrison town where it was initially stationed (south of the center of modern Cairo).
Greek and bilingual Arabic-Greek papyri list the special deliveries of fodder, food, clothes,
horses, and other supplies to units moving up the Nile or stationed in the Egyptian coun­
tryside (CPR VIII; SB VII 9748–9760). Other Arabic documents indicate how the system of
soldiers' payments was organized (Morimoto 1994). Seniority in Islam and experience in
battle determined the ranking on the dīwān (register of state pensions), which specified
the payments to soldiers and their families, a system seemingly new to the Byzantine and
Sassanian regions. Records of the soldiers and their families who were eligible for a
stipend were kept by officials sent to the garrisons to keep track of births, deaths, and the
arrivals of newcomers (Maqrīzī, Khiṭāṭ, 1, 252; Kindī, Kitab al-Wulāt, 86). One seventh-to-
eighth-century list of houses in Fusṭāṭ and the individuals living in them might be related
to the dīwān recordkeeping (Sijpesteijn 2008b). From Qurra b. Sharīk's letters to the pa­

Page 10 of 22
Arabic Papyri and Islamic Egypt

garch of Ishqaw/Aphrodito, which date to the early eighth century, we know much more
about how Egyptian communities contributed supplies, labor, and technical skills to the
Muslim army and fleet, as well as the other public activities of the caliph in Damascus
and the governor in Fusṭāṭ (P.Lond.IV; P.Heid.Arab. I).

These letters originate, however, fifty years after the conquest, when economic and ideo­
logical factors had effected profound changes in the Islamic social and administrative
structure in Egypt. From the end of the seventh and the beginning of the eighth century,
Christian officials started being replaced by Muslims at the level of the pagarchy, where­
as at the village level indigenous Christians retained their authority. At the same time, the
Muslim authorities in Egypt initiated several large-scale land surveys and censuses aimed
at improving their knowledge of the country's landholdings and increasing their tax take
as recorded in the papyri (Abbott 1965; CPR XXI 55). These developments were part of an
empirewide program to increase the Muslim presence in the administration and to Islami­
cize the (p. 463) administration (Sijpesteijn 2007b; forthcoming-a). We have some indica­
tion of the reaction this caused from a papyrus that bears witness to the first Christian
uprising (Bell 1945); another late seventh-century letter encourages the inhabitants of
Nessana to organize a tax protest against the new Muslim rulers in the area (P.Nessana
III 75).

Later documents are equally important for examining the day-to-day functioning of Mus­
lim rule in Egypt and the development of the administrative structures that underlay it.
Petitions show the channels open to citizens for redress against corrupt officials and
rulers (Khan 1990; P.Khalili I 16), hence the many threats of punishment that the
Umayyad governor Qurra b. Sharīk addresses to lower officials who were abusing taxpay­
ers or allowing others to do so (P.Qurra; P.Heid. Arab. II).

Conversion and Religious Practice

The relationship between Arabization and conversion is another issue for which the pa­
pyri can be very helpful. Onomastics has been used to trace the process of acculturation
in pre-Islamic Egypt and for Islamization in medieval Iraq and could also be used for simi­
lar research on Islamic Egypt.24 The differences between Arabic and Islamic names, be­
tween Christian and Muslim Arabic variants ofthe same name (e.g., Yaḥyā and Yuḥannis),
and the presence of Muslims with non-Muslim parental names can all be drawn upon in
this regard. Some caution, however, is required, as it is not clear how consciously names
were chosen and whether an Arabic or even a Muslim name signifies a sense of Arab or
Muslim identity or whether it represents merely an attempt to join the new ruling class.
These problems are well illustrated by a Coptic letter written between two correspon­
dents with Muslim names and contemporary Arabic letters written or received by men
bearing Christian names.25 A related problem involves the privileges that were associated
with conversion in this early period. The literary sources are not conclusive about the de­
gree to which converts in practice enjoyed the rewards that the law prescribed for con­
version: release from the poll tax, a lower agricultural tax rate, and inclusion on the

Page 11 of 22
Arabic Papyri and Islamic Egypt

dīwān. The papyri add to the complexity, but unlike the narrative sources, they have the
potential, through their anchoring in time and place, to give more detailed information.

Islamic religious practice (as well as the practice of other religions), the celebration of
festivals, and the expression of religious duties are all generously illuminated by the pa­
pyri. From the earliest Arabic dated document, in which the sheep received from the
town of Ihnās are described as “fifty for slaughtering and fifteen other sheep” (SB VI
9576) to the ninth-century letter in which the sender asks the addressee to buy a sacrifi­
cial animal for the Great Offering Feast after Ramaḍān (Rāghib 1980, no. 19), we learn
about the practicalities of daily religious life. In a late seventh-century letter, the ad­
dressee is invited to join the sender on the caliph's pilgrimage caravan and urged to bring
camels for the trip.26 Other (p. 464) documents written during the pilgrimage give back­
ground information and contemporary color to the travel accounts preserved in narrative
sources.

We can also see from the papyri how Islam's religious and legal institutions interacted
with local practices and historical conditions from the everyday experience of the Chris­
tians, Muslims, and Jews living under its rule. Unlike the narrative sources, moreover, the
papyri have the special advantage of being relatively free of moralizing judgments in their
treatment of minority behaviors. In a ninth—tenth century complaint to the governor
about a prayer leader who has introduced innovations in the prayer and has even brought
a bottle of date wine into the mosque, the writer asks the governor to take his responsi­
bility as supervisor of religious morals seriously and replace the wrongdoer (Rāghib 1978,
no. 5).

Information about how the execution of the Muslim religious duty of giving alms could
range from a state-collected and -distributed tax to a personal expression of piety is an
especially useful illustration of Islam's evolution and adaptability (Sijpesteijn 2007c). Peti­
tions from prisoners beseeching the sender's intercession or from a “poor man who has
nothing, not little nor much,” as well as lists of widows, orphans, and poor people receiv­
ing alms, all help to show us how the duty to distribute alms functioned in practice (e.g.,
Jahn 1937, no. 7; P.Khalili I1).

The papyri also tell us about the process of Muslim appropriation of local religious prac­
tices, cultic places, and religious ideas—in short, about the Islamization of the Egyptian
landscape. A late seventh-century letter records the visit of a number of Muslims at Saint
Catherine's monastery in the Sinai (P.Ness. III 72, 73, dated 683 or 684?).

Commerce and Production

While the papyri generally have a rather narrow geographical scope, citing mostly Egypt­
ian place names, the earliest Arabic commercial letter was written from somewhere in
North Africa to a town in Upper Egypt (Rāghib 1991). Other documents add to our under­
standing of how Egypt engaged the Mediterranean economic system and how commercial
contacts with East Asia via the larger Islamic empire increased after the Arab conquest
(Sijpesteijn 2004a). Thus, while Alexandria's status as the primary city in Egypt is gener­
Page 12 of 22
Arabic Papyri and Islamic Egypt

ally believed to have declined dramatically after the Muslim conquest, an early eighth-
century letter indicates that it was still an important focus of activity for Fayyum mer­
chants, presumably because it was connected to overseas markets, as was also the case
in the medieval period (Sijpesteijn 2004b; Udovitch 1996).

While a detailed study of prices for transport, goods, and services has been written on the
basis of the later medieval Geniza documents (Ashtor 1969), nothing comparable exists
for the earlier period, although various documents contain a wealth of information on this
topic. The well-developed commercial structures that greased the wheels of economic
life, such as checks and bills of credit (suftaja, (p. 465) ṣaḥifa), legal institutions such as
partnerships, and transport networks are well known from narrative sources. But it is the
documents that tell us how these worked “on the ground”—their safety, effectiveness, and
availability—and portray the importance of trading on credit versus a money economy
(Udovitch 1970).

Two mercantile archives demonstrate the potential of this material. The first consists of
about one hundred letters and legal documents related to a ninth-century family of textile
merchants situated in the Fayyum and their partners in Fusṭāṭ (P.Marchands I–III).27 The
letters show us how cloth was ordered piecemeal from different weavers in the Fayyum,
how clothes and textiles were traded and sold in Fusṭāṭ, and how new orders were filled.
Other letters give insights into the private lives of these merchant families when they
mention requests for financial support from distant family members, the illness or death
of relatives, or exchanges between a mother and her son.

The second archive, which comprises at least thirty letters, remains largely unpublished
(P.Berl.Arab.II 38–43).28 Also from the ninth century, they deal with trade in wool, soap,
textiles, and other household goods. The large amounts of money involved and the exten­
sive commercial network within which these traders operated point to considerable so­
phistication in terms of both volume and value.

It is from documents such as these that we are also able to see two social groups that
played an essential role in medieval economic life but which remain largely invisible in
the narrative record. Women, sometimes writing their own documents and sometimes us­
ing (female) scribes to do so (P.Khalili I 17), appear in a wide range of economic activities
from traditional female occupations, such as the domestic textile industry (Rapoport
2005), to commercial partnerships (P.Khalili I 21). Their relative freedom, depending on
their social and marital status (unmarried, married, widowed), and the way in which their
legal and social position changed over time and differed across religious communities all
come through in the documents.

The second group is slaves. Slave sale and manumission contracts and inheritance docu­
ments show how human property was transferred and traded, by whom, and for what
purposes (P.Vente; P.Cair.Arab.I 37). Contracts that record partnerships in which slaves
play a role show their active involvement in the commercial sphere, and the frequent ap­
pearance of domestic slaves in documents of all sorts can be contrasted with the over­

Page 13 of 22
Arabic Papyri and Islamic Egypt

whelming impression given in the literary sources that all slaves were either musicians or
dancing girls.

Rural History

Since most of the Arabic papyri have been found in the Egyptian countryside as opposed
to permanently occupied towns, one would expect them to give us an especially good
vista onto the organization of the countryside and the nature of rural life.

The reliance on the yearly inundation cycle of the Nile did not diminish under the
(p. 466)

Muslims, and the papyri contain much information on the organization and maintenance
of irrigation systems, such as who was responsible for them (state officials or landown­
ers) and how they were financed (from taxes, private monies, or corvée labor). These lead
us directly into one of the central questions of this period: the distribution of power in the
countryside—that is, whether the large landowners continued to play the important role
they had played in late antique Egypt, how the appearance of Muslim landowners was re­
lated to conversion and Arab settlement, and how (or whether) Muslim landholders dif­
fered from their Christian counterparts (cf. Banaji 2001, 141–170). A related issue is the
determination of who represented and carried out the wishes of the Muslim authorities in
the countryside in their attempts to maximize agricultural fiscal income through land sur­
veys, the development of “dead lands,” and the supervision of peasants' movements. The
fluctuating effectiveness of revenue-raising measures and the different methods applied
by the authorities to collect agricultural taxes have been studied by Frantz-Murphy in her
collection of agricultural leases and receipts (CPR XXI).

Historical events and processes, which might be difficult to trace in the narrative sources,
colored as they often are by later discussions, have left a fertile residue in the papyri. The
later legal debates about the fiscal status of conquered lands, for example, which suppos­
edly depended on the circumstances of their capture, are difficult to disentangle from the
economic and political concerns by which they were obviously influenced (Noth 1973). A
change in Egypt's fiscal status effected in the early eighth century is, however, reflected
in the agricultural leases and receipts (CPR XXI), thereby allowing us to understand the
narrative sources better.

The agricultural revolution presumed to have followed in the wake of the Arab con­
querors has to be studied in more detail with the aid of the many documents that show us
the kinds of produce that grew in Egyptian fields and how this changed in the course of
time. The degree to which agricultural production was specialized and what infrastruc­
ture and investment was required for such specialization are questions to which the pa­
pyri hold the answers. The abundance of material related to the medieval flax and textile
trade, for example, would make an exceptionally good case study.

The history of the early Islamic Egyptian countryside is but one topic for which papyri can
offer us invaluable new insights. While the importance of papyrological evidence and the
light it shines on almost every aspect of life under medieval Islam is now widely acknowl­
edged, the actual use of papyri as a historical source has been slower to take hold. With
Page 14 of 22
Arabic Papyri and Islamic Egypt

the development of new tools and infrastructure, however, many of the barriers that hin­
der students from entering this field are being broken down, and the road is opening on a
new level of Islamic historical research.

(p. 467) Acknowledgment

For this chapter I have made extensive use of the following overviews on the study of Ara­
bic papyrology: Sundelin (2004), Grohmann (1952, 1966), and Rāghib (1996).

Bibliography
Abbott, N. 1941. “Arabic Marriage Contracts among Copts.” Zeitschrift der Deutschen
morgenländischen Gesellschaft 95: 59–81.

———. 1949. “A Ninth-century Fragment of the Thousand Nights: New Light on the Early
History of the Arabian Nights.” Journal of Near Eastern Studies 8: 129–164.

———. 1957. Studies in Arabic Literary Papyri. Vol. 1, Historical Texts. Chicago: Universi­
ty of Chicago Press.

———. 1965. “A New Papyrus and a Review of the Administration of ʻUbaid Allāh b. al-
Ḥabḥ āb.” In Arabic and Islamic Studies in Honor of Hamilton A. R. Gibb, ed. George
Makdisi, 21–35. Leiden: Brill.

———. 1967. Studies in Arabic Literary Papyri. Vol. 2, Qurʼānic Commentary and Tradi­
tions. Chicago: University of Chicago Press.

———. 1972. Studies in Arabic Literary Papyri. Vol. 3, Language and Literature. Chicago:
University of Chicago Press.

Al-Akwāʻ, I. ʻA. 1985. “Jāmiʻ Ṣanʻaʻ abraz maʻālim al-ḥaḍāra al-islāmiyya fī al-Yaman.” In
Maṣāḥif Ṣanʻāʻ, ed. Ḥuṣṣa al-Ṣabāḥ. Kuwait: Dār al-āthār al-Islāmiyya.

Ashtor, E. 1969. Histoire des prix et des salaires dans lʼOrient médiéval. Paris: SEVPEN.

Bagnall, R. 1982. “Religious Conversion and Onomastic Change in Early Byzan­

(p. 469)

tine Egypt.” BASP 19: 105–124.

Balādhurī. (d. 892). Futūḥ al-buldān, ed. M. J. de Goeje. 1866

;
repr., 1992, Leiden

. In
The Origins of the Islamic State, trans. P. K. Hitti and F. Murgotten. 1916

;
repr., Piscataway, N.J.: Gorgias Press, 2002.

Page 15 of 22
Arabic Papyri and Islamic Egypt

Banaji, J. 2001. Agrarian Change in Late Antiquity: Gold, Labour, and Aristocratic Domi­
nance. Oxford: Oxford University Press.

Becker, C. H. 1902–1903. Beiträge zur Geschichte ägyptens unter dem Islam. Strassburg:
K. J. Trübner.

———. 1906. “Arabische Papyri des Aphroditofundes.” Zeitschrift für Assyriologie und ver­
wandte Gebiete 20: 68–104.

———. 1911. “Neue arabische Papyri des Aphroditofundes.” Der Islam 2: 245–268.

Bell, H. I. 1945. “An Official Circular Letter of the Arab Period.” JEA 31: 75–84.

Bilabel, F., and A. Grohmann. 1934. Griechische, koptische und arabische Texte zur Reli­
gion und religiösen Literatur in Ägyptens Spätzeit. Veröffentlichungen aus den badischen
Papyrussammlungen 5. 2 vols. Heidelberg: Verlag der Universitätsbibliothek.

Bloom, J. M. 2001. Paper before Print: The History and Impact of Paper in the Islamic
World. New Haven, Conn.: Yale University Press.

Bukhārī. 1862–1908. (d. 870). Ṣaḥīḥ, ed. L. Krehl. Leiden: Brill.

Bulliet, R. W. 1994. Islam: The View from the Edge. New York: Columbia University Press.

Cook, M., and P. Crone. 1977. Hagarism: The Making of the Islamic World. Cambridge:
Cambridge University Press.

David-Weill, J. 1939. Le Djamīʻ dʼ Ibn Wahb. Textes arabes 3. Cairo: Imprimerie de

lʼInstitut français dʼarchéologie orientale.

Déroche, F. 1983. Les manuscrits du Coran: Aux origines de la calligraphie coranique.

Bibliothèque nationale, département des manuscrits, catalogue des manuscrits arabes 2,
1. Paris.

———. 1992. The Abbasid Tradition: Qurʼāns of the 8th to 10th Centuries A.D. Nasser D.
Khalili Collection of Islamic Art 1. Oxford: Oxford University Press.

Diem, W. 1979–1981. “Untersuchungen zur früahen Geschichte der arabischen Orthogra­

phie.” Orientalia 48: 207–257; 49: 67–106; 50: 332–383; 52: 257–404.

Dietrich, A. 1954. Zum Drogenhandel im islamischen Ägypten: eine Studie über die ara­
bische Handschrift nr. 912 der Heidelberger Papyrus-Sammlung. Heidelberg: Carl Winter.

Frantz-Murphy, G. 1994. “Papyrus Agricultural Contracts in the Oriental Institute Muse­

um from Third/Ninth Century Egypt.” Itinéraires dʼOrient: Hommages à Claude Cahen,
Res Orientalis 6: 119–131.

Page 16 of 22
Arabic Papyri and Islamic Egypt

———. 2007. “The Economics of State Formation in Early Islamic Egypt.” In From al-An­
dalus to Khurasan: Documents from the Medieval Islamic World, ed. P. M. Sijpesteijn, L.
Sundelin, S. Torallas Tovar, and A. Zomeño, 101–114. Leiden: Brill.

George, A. 2006. “The Geometry of Early Islamic Calligraphy.” Diss. Oxford University.

Grohmann, A. 1932. “Griechische und lateinische Verwaltungstermini im arabischen Ae­

gypten.” CdÉ 7: 275–284.

———. 1952. From the World of Arabic Papyri. Cairo: Al-Maaref Press.

———. 1959. Studien zur historischen Geographie und Verwaltung des frühmittelalter­
lichen Ägypten. Österreichische Akademie der Wissenschaften, Philosophisch-historische
Klasse. Denkschriften, 77, 2. Vienna: In Kommission bei R. M. Rohrer.

———. 1960. “Zum Papyrusprotokoll in früharabischer Zeit.” Jahrbuch Österre­

(p. 470)

ichische Byzantinische Gesellschaft 9: 1–19.

———. 1966. Handbuch der Orientalistik. Vol. 1, Arabische Chronologie. Vol. 2, Arabische
Papyruskunde. Leiden: Brill.

———. 1967–1971. Arabische Paläographie. Österreichische Akademie der Wis­

senschaften. Philosophisch-historische Klasse. Denkschriften 94.1. 2 vols. Vienna: Verlag
der Österreichischen Akademie der Wissenschaften.

Gruendler, B. 1993. The Development of the Arabic Scripts: From the Nabatean Era to the
First Islamic Century according to Dated Texts. Harvard Semitic Studies 43. Atlanta:
Scholars Press.

Hanafi, A. 2004. “Two Unpublished Paper Documents and a Papyrus.” In Papyrology and
the History of Early Islamic Egypt, ed. P. M. Sijpesteijn and L. Sundelin, 45–62. Leiden:
Brill.

Hawting, G., and A.-K. A. Shareef, eds. 1993. Approaches to the Qurʼān. London: Rout­
ledge.

Hopkins, S. 1984. Studies in the Grammar of Early Arabic, Based upon Papyri Datable to
before 300 A.H./912 A.D. London Oriental Series 37. Oxford: Oxford University Press.

Ibn Ḥawqal. (d. after 973). Kitāb ṣūrat al-arḍ, ed. J. H. Kramers. Bibliotheca Geographo­
rum Arabicorum II. Leiden: Brill, 1967.

Jahn, K. 1937. “Vom frühislamischen Briefwesen.” Archiv Orientální 9: 153–200.

Khan, G. 1990. “The Historical Development of the Structure of Medieval Arabic Peti­
tions.” Bulletin of the School of Oriental and African Studies 53: 8–30.

———. 1994. “The Pre-Islamic Background of Muslim Legal Formularies.” ARAM 6:193–
224.

Page 17 of 22
Arabic Papyri and Islamic Egypt

———. 2005. Arabic Documents from Early Islamic Khurasan. London: Nour Foundation
in association with Azimuth Editions.

———. 2007. “Newly Discovered Arabic Documents from Early Abbasid Khurasan.” In
From al-Andalus to Khurasan: Documents from the Medieval Islamic World, ed. P. M. Si­
jpesteijn, L. Sundelin, S. Torallas Tovar, and A. Zomeño, 201–215. Leiden: Brill.

Khoury, R. G. 1972. Wahb b. Munabbih. Wiesbaden: Harrassowitz.

———. 1986. ʻAbd Allāh ibn Laḥīya (97–174/715–790), juge et grand maître de lʼEcole
égyptienne: Avec édition critique de lʼunique rouleau de papyrus arabe conservé à Heidel­
berg. Wiesbaden: Harrassowitz.

Kindī. (d. 961). Kitāb al-wulāt wa-kitāb al-quḍāt, ed. R. Guest. In The Governors and
Judges of Egypt. Leiden: Brill, 1912.

Maqrīzī. (d. 1442). Al-Mawāʻiẓ wa-ʼl-iʻtibār fī dhikr al-khiṭaṭ wa-ʼl-āthār, ed. A. F. Sayyid.
London: Al-Furqān Islamic Heritage Foundation, 2002–2003.

Morimoto, K. 1994. “The Dīwāns as Registers of the Arab Stipendiaries in Early Islamic
Egypt.” Res Orientalis 6: 353–366.

Moritz, B. 1905. Arabic Palaeography: A Collection of Arabic Texts from the First Century
of the Hidjra till the Year 1000. Cairo: Publications of the Khedival Library 16.

Noth, A. 1973. “Zum Verhältnis von kalifaler Zentralgewalt und Provinzen in umayyadis­
cher Zeit: Die Ṣulḥ-ʻAnwa-Traditionen für Ägypten und den Iraq.” Die Welt des Islams 14:
150–162.

Petrosyan, Y. A., and M. L. Swietochowski. 1995. Pages of Perfection: Islamic Paintings

and Calligraphy from the Russian Academy of Sciences, St. Petersburg. Milan: Electa.

Rāghib, Y. 1978. “Lettres arabes (I).” Annales Islamologiques 14: 15–35.

———. 1980. “Lettres arabes (II).” Annales Islamologiques 16: 1–29.

———. 1991. “La plus ancienne lettre arabe de marchand.” In Documents de

(p. 471)

lʼislam médiéval: Nouvelles perspectives de recherche, ed. Y. Rāghib, 1–9. Cairo: Im­
primerie de lʼInstitut français dʼarchéologie orientale.

———. 1996. “Les plus anciens papyrus arabes.” Annales Islamologiques 30: 1–19.

Rapoport, Y. 2005. Marriage, Money, and Divorce in Medieval Islamic Society. Cam­
bridge: Cambridge University Press.

Richter, T. S. 2000. “Spätkoptische Rechtsurkunden neu bearbeitet (II): Die Rechtsurkun­

den des Teschlot-Archivs.” Journal of Juristic Papyrology 30: 95–148.

Page 18 of 22
Arabic Papyri and Islamic Egypt

———. 2004. “O.Crum Ad. 15 and the Emergence of Arabic Words in Coptic Legal Docu­
ments.” In Papyrology and the History of Early Islamic Egypt, ed. P. M. Sijpesteijn and L.
Sundelin, 97–114. Leiden: Brill.

Schaefer, K. 2002. “Arabic Printing before Gutenberg: Block Printed Arabic Amulets.” In
Middle Eastern Languages and the Print Revolution: A Cross-cultural Encounter. A Cata­
logue and Companion to the Exhibition, ed. E. Hanebutt-Benz, D. Glass, and G. Roper,
123–128. Westhofen, Germany: WVA-Verlag Skulima.

Sijpesteijn, P. M. 2003. “North American Papyrus Collections Revisited.” Al-Bardiyyat 1:

5–18.

———. 2004a. “A Request to Buy Silk from Early Islamic Egypt.” In Gedenkschrift Ulrike
Horak, ed. H. Harrauer and R. Pintaudi, 255–272. Papyrologica Florentina 34. Florence:
Gonnelli.

———. 2004b. “Travel and Trade on the River.” In Papyrology and the History of Early Is­
lamic Egypt, ed. P. M. Sijpesteijn and L. Sundelin, 115–152. Leiden: Brill.

———. 2006. “Arabic Papyri from Current Excavations in Egypt.” Al-Bardiyyat 2: 10–23.

———. 2007a. “The Muslim Conquest and the First Fifty Years of Muslim Rule in Egypt.”
In Egypt in the Byzantine World, 300–700, ed. R. S. Bagnall, 437–459. Cambridge: Cam­
bridge University Press.

———. 2007b. “New Rule over Old Structures: Egypt after the Muslim Conquest.” In
Regime Change in the Ancient Near East and Egypt: From Sargon of Agade to Saddam
Hussein, ed. H. Crawford, 183–202. Oxford: Oxford University Press for the British Acad­
emy.

———. 2007c. “Creating a Muslim State: The Collection and Meaning of Sadaqa.” In Ak­
ten des 23. Internationalen Papyrologenkongresses, Wien, 22.–28. Juli 2001, ed. B. Palme,
661–673. Vienna: Verlag der Österreichischen Akademie der Wissenschaften.

———. 2008a. “Arabic Palaeography.” In Encyclopedia of Arabic Language and

Linguistics, ed. C. Versteegh. Vol. 3, 513–524. Leiden: Brill.

———. 2008b. “A Seventh-Eighth-Century List” In Sixty-Five Papyrological Texts Present­

ed to Klaas A. Worp on the Occasion of his 65th Birthday, eds. F.A.J. Hoogendijk and B.
Muhs. Papyrologica Lugduno-Batava 33, Leiden: Brill, 2008, 369–377.

———. Forthcoming-a. Shaping A Muslim State: The World of a Mid-Eighth-Century

Egyptian Official. Oxford: Oxford University Press.

———. Forthcoming-b. “An Invitation to Join the Caliph's Pilgrim's Caravan.” Journal of
Near Eastern Studies.

Page 19 of 22
Arabic Papyri and Islamic Egypt

Silvestre de Sacy, A. I. 1825. “Mémoire sur papyrus écrits en arabe et récemment décou­
verts en Égypte.” Journal des Savants: 462–473.

———. 1831. “Mémoire sur papyrus écrits en arabe et récemment découverts en Égypte.”
Mémoires de lʼInstitut Royal de France, Académie des Inscriptions et Belles-Lettres 9: 66–
85.

———. 1833. “Mémoire sur deux papyrus, écrits en langue arabe, appartenant à la collec­
tion du Roi.” Mémoires de lʼInstitut Royal de France, Académie des Inscriptions et Belles-
Lettres 10: 65–88.

Sundelin, L. 2004. “Introduction: Papyrology and the Study of Early Islamic

(p. 472)

Egypt.” In Papyrology and the History of Early Islamic Egypt, ed. P. M. Sijpesteijn and L.
Sundelin, 1–19. Leiden: Brill.

Udovitch, A. L. 1970. Partnership and Profit in Medieval Islam. Princeton: Princeton Uni­
versity Press.

———. 1996. “Medieval Alexandria: Some Evidence from the Cairo Genizah Documents.”
In Alexandria and Alexandrianism: Papers Delivered at a Symposium Organized by the J.
Paul Getty Museum and the Getty Center for the History of Art and the Humanities and
Held at the Museum, April 22–25, 1993, 273–283. Malibu: J. Paul Getty Museum.

Yaʻqūbī. (d. 284/897). Kitāb al-buldān, ed. M. J. de Goeje. Bibliotheca Geographorum Ara­
bicorum 7. Leiden: Brill, 1892.

———. Taʼrīkh, ed. M. T. Houtsma. Leiden: Brill, 1883.

Ziadeh, J. 1915–1917. “lʼapocalypse de Samuel, supérieur de Deir-el-Qalamoun: Texte

arabe édité et traduit en français.” Revue de lʼorient chrétien 20: 374–407.

Zomeño, A. 2003. “Del escritorio al tribunal: Estudio de los documentos notariales en la

Granada nazarí.” In Grapheîon: Códices, manuscritos, e imágines: Estudios filológicos e
históricos, ed. J. P. Monferrer Sala and M. Marcos Aldón, 75–98. Cordoba: Universidad de
Granada.

Notes:

(1.) The tradition mentions sheets (riqāʻ), shoulder blades (aktāf), and palm branches
(ʻusub) (Bukhārī, Ṣaḥīḥ III 257). The conqueror and first governor of Egypt, ʻAmr b. al-ʻĀṣ
(d. 664), allegedly sent papyrus and wheat from Egypt to Medina (Yaʻqūbī, Taʼrīkh, vol.
2,177).

(2.) The caliph al-Muʻtaṣim (r. 833–842) built a papyrus mill in Samarrā’in 836 (Yaʻqūbī,
Tarʼīkh, II, 577; Kitāb al-buldān, 39). Pliny reported that papyrus grew at the Euphrates
near Babylon (NH 13.11.73). For papyrus growth in Sicily, see Ibn Ḥawqal, Kitāb ṣūrat al-
arḍ. Cf. Grohmann 1952, 19–21.

Page 20 of 22
Arabic Papyri and Islamic Egypt

(3.) After the conquest, the use of Coptic as an administrative language actually in­
creased in Upper Egypt, where Greek had never made much of an impact.

(4.) The first one, SB VI 9576, is mentioned later in the chapter; the second is a demand
note for taxes paid in coin, of which only a photograph has so far been published (P.Berol.
inv. 15002, depicted in Grohmann (1966, Tafel II).

(5.) A Inv. Chart. Ar. 7161.

(6.) The last known Arabic papyrus is dated A.H. 480 (1087 CE) and is kept in the John
Rylands collection in Manchester (P.Ryl.Arab. X no. 10).

(7.) The latest is a papal bull dated 1057, which is preserved in the Vatican.

(8.) For the distribution of medieval Arabic leather and paper documents and parchments
throughout the Islamic empire, see Khan 2005; Petrosyan et al. 1995, 122–123 (Central
Asia); al-Akwaʻ 1985 (Yemen); and Zomeño 2003 (Andalusia).

(9.) See also the list of early ninth-century literary works preserved in manuscript form on
parchment (Rāghib 1996, 3–4).

(10.) Like Qurʼānic paper and parchment codices (parchment seems to have been the pre­
ferred material), these Qurʼāns were also written on pages wider than they are high, a re­
verse of the later custom (Déroche 1983, 19; 1992).

(11.) The first two, published by Khoury (1972), are ascribed to Wahb b. Munabbih (d.
728) and are currently kept in the Heidelberg papyrus collection.

(12.) This ninth-century astronomical papyrus, part of a Dutch private collection, remains
unpublished.

(13.) For published amulets on papyrus see Bilabel and Grohmann (1934). For handwrit­
ten paper amulets see, for example, Hanafi (2004), and for block-printed ones, see Schae­
fer (2002).

(14.) Abbott, Karabacek, Silvestre de Sacy, and Becker are reported as having tried to do
so by Khan, who emphasizes the problems of such a method and instead gives a detailed
description of the features of the papyrus scripts (P.Khalili I, pp. 27–46, esp. 44–46 and
note 71). But see the recent work by Alain George, who has identified a number of papyri
that follow chancellery characteristics (2006). For a description of the palaeography of
documents, see also Sijpesteijn (2008a).

(15.) For palaeographical albums see, for example, Moritz (1905); Gruendler (1993);
Grohmann (196, 1971).

(16.) Grohmann (1960); CPR III 1/2, nos. 109–112.

(17.) SB VI 9576; P.Nessana III; P.Heid.Arab. I; P.Lond. IV.

Page 21 of 22
Arabic Papyri and Islamic Egypt

(18) (.) Frantz-Murphy (1994).

(19.) For this so-called Teschlot archive, see Richter (2000).

(20.) CPR XXI; Frantz-Murphy (2007); Rāghib (1991, 8).

(21.) For example, the terms symboulos and protosymboulos for governor and caliph, re­
spectively. P.Lond. IV 1335.2, commentary.

(22.) For example, veredarius (“fast messenger”) reappears in the Islamic period after an
absence from the papyrological record for about four centuries (CPR XIV 33.2, commen­
tary). See also, for example, the words cursus and andrismos, which have no Egyptian
precedents; cf. Grohmann (1932).

(23.) See Khan's study of the historical development of petition formulae (1990).

(24.) For pre-Islamic Egypt, see Bagnall (1982); for Iraq, Bulliet (1994).

(25.) CPR II 228; e.g., P.Khalili I 14; Jahn (1937, no. 12).

(26.) Sijpesteijn (forthcoming-b).

(27.) P.Marchands IV contains twenty-two letters from three ninth-century textile agents.

(28.) The unpublished material can be found in the University of Michigan and University
of Chicago collections. It is being prepared for publication by Yūsuf Rāghib and Petra Si­
jpesteijn.

Petra M. Sijpesteijn

Petra M. Sijpesteijn is Professor of Arabic Language and Culture, Leiden University.

Page 22 of 22
The Papyrology of the Near East

The Papyrology of the Near East

Jean Gascou
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0020

Abstract and Keywords

The first section of this article discusses the Near Eastern texts from Egypt and discover­
ies in the Near East. The second section notes that the Near Eastern papyri form five
principal bodies: Dura, the Middle Euphrates, Nahal Hever, Petra, and Nessana. The third
section discusses paleography, which requires a mastery of many variables and a good
knowledge of the Egyptian texts. The fourth section considers formats and diplomatics.
The fifth and sixth sections discuss language and law. The seventh section compares Near
Eastern papyrology and Egyptian papyrology.

Keywords: Near Eastern papyri, Nahal Hever, Petra, Middle Euphrates, Nessana, Egyptian texts, paleography

THE written documentation of the Hellenized Orient has already been inventoried and
discussed.1 Most of the writing materials, both flexible and rigid, attested in Egypt are al­
so represented in the Near East: tablets, ostraca,2 and, to be sure, papyrus, but writing in
oriental languages is part of papyrology only by assimilation. When the Greeks and Mace­
donians destroyed the Achaemenid monarchy and opened the East and central Asia to
Greek, they encountered in Asia Minor and on the Mediterranean coast peoples who al­
ready used papyrus. Farther to the East, in contrast, they found the custom of writing lit­
erary and religious works and public and private legal documents on skin (diphthera or
membrana) well entrenched, whether this was finished as leather or as parchment.3
Although the Seleucids introduced papyrus into countries like Babylonia, the use of skin
remained dominant in north Syria, Mesopotamia, and central Asia under the Greek and
non-Greek monarchies that emerged from the dissolution of Alexander's empire and un­
der the Romans. We have direct testimony of this practice in the Parthian “parchments”
from Avroman in Iranian Kurdistan,4 a Graeco-Bactrian receipt for a tax on a transaction,5
a substantial late antique dossier in the Bactrian language transcribed into the Greek al­
phabet,6 and the texts of Dura and Beth Phouraia/Appadana on the Middle Euphrates.7
Indirect testimony comes from the impressions left by writing materials on the Hellenistic
and Roman sealings found in Mesopotamia, Palmyra, and other cities.8 Numerous skins
also appear in the finds from the Dead Sea, mainly for Jewish theological and liturgical

Page 1 of 23
The Papyrology of the Near East

works but also for legal documents. Only with the Roman conquest of inland areas does
papyrus begin to (p. 474) compete with or supplant skins for some types of acts, often in
military contexts, like the archives of units, or administrative settings, as with petitions.

A similar situation prevails with languages, for even if the Greeks and Macedonians pro­
moted Greek as far east as central Asia and the borders of India, the Semitic languages of
the Near East continued to be written. From a linguistic point of view the Roman con­
quest had contrasting and even paradoxical effects. Under the Romans, Greek gains
ground in writing and is introduced even into countries where it was not customary. The
most striking case is that of the Arabo-Nabataean zone, where Nabataean writing rapidly
disappeared in favor of Greek after the creation of the province of Arabia in 106. Similar­
ly, the final disappearance of the kingdom of Oshroene under Gordian III was accompa­
nied by the replacement of Syriac by Greek at Marcopolis.9 Edessa, however, the former
capital of the Abgarids, forms an exception of great historical significance because, like
Palmyra until its capture by Aurelian, this city kept its national language alongside
Greek.10 Hebrew experienced a nonliturgical resurgence, particularly between 132 and
135, during the Bar Kokhba revolt.11

The Discoveries
Near Eastern Texts from Egypt

Some papyri written in the Near East by local persons or by people from Egypt living in
or visiting the Near East, especially soldiers, have been found in Egypt. These include let­
ters12 title deeds for slaves13 and public14 and private15 documents. Some of these papyri
belong to travel files. The archive of Zenon contains several accounts, letters, and docu­
ments from the years 261–252, collected by Zenon himself while on business trips to
northern Palestine, southern Syria (the Hauran), the Ammanitis, and Idumaea (ca. 261–
259).16 Several centuries later, around 318 CE, the Hermopolite scholasttkos Theophanes,
a high official in the prefecture of Egypt, went to Antioch and brought back from this five-
month trip his expense accounts, itineraries, and other papers. 17 Other texts come from
Zenon's correspondents and those of his employer, the dioiketes Apollonios. They testify
in their own way, during a period of conflict with the Seleucids, to the Ptolemaic domina­
tion over these areas. Several of them inform us about Toubias, a Transjordanian aristo­
crat in Ptolemaic service and a member of a family famous in Jewish history.

For good measure, we may add that the finds from the Near East include a number of
“traveling” texts written at Edessa18 Marcopolis,19 Carrhai,20 Antioch,21 Gaza,22 and other
localities where so far no papyri have been found.

(p. 475) Discoveries in the Near East

So far only about 600 Greek and Latin papyri from the Near East (including those found
in Egypt) have come to light. That is far more than just fifty years ago, when the texts of

Page 2 of 23
The Papyrology of the Near East

Dura and Nessana had not yet been made fully available, but this is hardly anything com­
pared to the roughly 50,000 Greek and Latin papyri of Egyptian provenance published so
far. In some parts of the Near East, as in the area of Palmyra, the middle and lower Eu­
phrates valley, and the Arabo-Nabataean zone, the climate is almost as arid as that of
Egypt. But far more than Egypt, these countries experienced rebellions, wars, invasions,
natural catastrophes, and other forms of devastation in antiquity, all unfavorable to the
survival of texts. Vast stores of archives must have vanished in the destructions of Seleu­
cia on the Tigris by the Parthians and then by Trajan, as well as in the sacking of Palmyra
by Aurelian.23 In other regions, excessive rainfall or the continuity of human occupation
from antiquity to the present have destroyed or made the written materials inaccessible.

Indirect Discoveries
Some of our information is indirect in nature. Earlier I mentioned the sealings of Syro-
Mesopotamian cities. Careful study of these, like that carried out by H. Seyrig for several
Syrian cities, can inform us not only about the writing materials used but also about the
format of the vanished documents and thus, to some degree, their content.

More eloquent, one might say, are the impressions of literary texts left by their ink on the
mud bricks of one room at the Hellenistic site of Aï Khanoum in Afghanistan.24

Direct Discoveries
Fortunately for scholarship, most of the papyri of the Near East have been discovered
during regular excavations and in known archaeological contexts. This is the case with
Dura, Nessana, Petra, Khirbet Mird, and Aï Khanoum. As in Egypt, however, some of the
finds have been clandestine. Before reaching the hands of editors, they have been the ob­
ject of transactions, sometimes accompanied by false or doubtful information and some­
times dismembered into homogeneous batches. We are therefore still ignorant of the ori­
gin and environment of a dossier of such prime importance for the history of Roman Syria
as that of the Middle Euphrates. The same obscurity afflicts two papyri said to come from
Bostra.25 In the case of the Dead Sea caves, the situation is mixed: Irregular discoveries
have been intermingled with texts brought to light by archaeological missions. As a re­
sult, their editors and interpreters spent more time than they would wish in checking in­
formation and reassembling dossiers.26 In consequence, the publications are rather disor­
derly and present difficulties to those who use these catalogues. Such is the case with the
Jericho papyri (P.Jud.Des.Misc.), where a (p. 476) number of provenances other than Jeri­
cho and its environs are mixed in (e.g., Nahal Hever).

In general it is notable that, even taking into account texts written in the vernacular lan­
guages, most of our finds come from the Byzantine period (Nessana, Petra) and above all
the Roman period (Dead Sea, Dura, Middle Euphrates); they tell us little about earlier
states, particularly about the Seleucid monarchy. Moreover, they are distributed in two
extended and discontinuous zones, namely the eastern edges of the Roman province of
Syria (Dura, Beth Phouraïa/Appadana) and the Judaean and Arabo-Nabataean region
(Dead Sea, Nessana, Petra, Bostra).

Page 3 of 23
The Papyrology of the Near East

Within these zones, urban sites, the ancient “cities,” are poorly represented except for
Dura and Petra. (It is by no means certain that the papyri said to belong to Bostra actual­
ly come from Bostra itself.) The papyri discovered at Palmyra are insignificant and consist
of some sixty unpublished fragments of Greek and Palmyrene papyri from the tomb of Ki­
tot (40 CE), as well as one book of school tablets.27 The twenty-one pieces from the Mid­
dle Euphrates may come from Appadana, a site in the Lower Khabour Valley that had just
been promoted to the rank of city and given the name of Neapolis, but they were proba­
bly brought there by villagers from Beth Phouraia. Nessana and its neighbor, Sobata28
were only villages, and Khirbat Mird was just a small monastic establishment.29 As to the
discoveries from the Dead Sea, these were mainly items from various provenances and
were hidden or discarded in caves.

The case of the grottoes and ravines of the Dead Sea leads to the positing of another
characteristic useful to keep in mind in considering these dossiers. There are internal
reasons (notably the narrowness of the chronological spread), sometimes reinforced by
archaeological observations, to consider some of the finds deliberate deposits connected
to specific crises of urgency or distress.

Such an interpretation seems nearly certain with the roughly forty Aramaic papyri from a
cave in the Wadi Daliyeh, fourteen kilometers northwest of Jericho, which, considering
the period covered (375–335 BCE) and the large quantity of human remains connected
with them, must record in their own way the last episode of the Samaritan revolt under
Alexander, in 331.30 An analogous case appears to be attested at Jericho, this time in con­
nection with the capture of the city by Ptolemy I in 312 BCE.31 Similarly, the last secure
dates found at Qumran place those finds in the period of the Jewish war of 66–73 CE. The
contemporary pieces of the famous Simeon bar Kokhba, the “prince of Israel” who direct­
ed the Jewish revolt of 132–135 CE, provide a link among the finds of Murabbaʼat, Wadi
Sdeir, and Nahal Hever. It is quite possible that the Judaeo-Nabataean women Babatha
(daughter of Simon) and Salome Komaïse, the centers of many of the papyri from Nahal
Hever, wanted to shelter their papers and their persons in the caves of this valley. At
Masada it is also possible to see how, after the Roman capture of the site, the papers of
the defeated Jews were discarded in various places.

The Middle Euphrates dossier must have been put together under analogous cir­
(p. 477)

cumstances. A number of villagers from Beth Phouraia, worried about the Sassanid inva­
sion of the years 253–256 (which actually led to the destruction of Dura), sought to pre­
serve the documents connected with their legal proceedings and their property titles,
hoping to recover them when things returned to normal.

The Five Main Dossiers

The Near Eastern papyri form five principal bodies: Dura, the Middle Euphrates, Nahal
Hever, Petra, and Nessana. Some description of each follows.

Page 4 of 23
The Papyrology of the Near East

Dura

Dura-Europos is an ancient stronghold on the right bank of the Middle Euphrates, exten­
sively altered by the Seleucids. From the beginning, they provided it with urban struc­
tures, including an archives (chreophylakeion) that functioned for centuries, particularly
for the registration of legal documents (see P.Dura 15, 17, 28). Even in the third century
of our era, Hellenistic law remained in use at Dura (P.Dura 12). Two texts are datable to
the Seleucid period (P.Dura 15 and 34). Dura then came under Parthian rule (113 BCE)
and eventually Roman control ( 166 CE) and even became a Roman colony in 211. The
city, which was at that time part of the Roman province of Syria Coele, was occupied,
then depopulated and destroyed between 253 and 256 by the Sassanids. This succession
of governments is visible, among other ways, in the variety of languages represented at
Dura: Greek and Latin, but also Palmyrene and Iranian. The published textual corpus
amounts to 154 documents (45 on skin and 109 on papyrus), found in various parts of the
site, along with some ostraca, mainly of the Parthian or Roman period.32 The pre-Roman
documents are written on skin, and those of the Roman period are to a large extent on pa­
pyrus, particularly in the military context. Most of the Dura texts that record private
transactions are drawn up as double documents.33

The papers of the third-century Roman garrison deserve particular notice. They are large­
ly in Latin and tell us in detail about the organization of the unit (the cohors XX
Palmyrenorum), its festivals (in the famous P.Dura 60, the feriale duranum), its relation­
ships, and its litigation.34

The written documentation of Dura can and must be studied in its archaeological context,
which is well preserved and extensively (although not yet completely) (p. 478) published.35
Fruitful excavations have resumed after a long hiatus under the direction of P. Leriche.
This context includes the site's monumental inscriptions and above all its numerous and
rich graffiti, which, in matters of the economy and accounting, display types of text and
handwriting that one would have thought limited to papyrological texts. This epigraphy,
however, is still incompletely published and scattered in the preliminary reports.36

The Middle Euphrates

This homogeneous and well-preserved ensemble of nineteen Greek and two Syriac items
arrived in Europe in 1988 through the antiquities trade and is associated with several
small objects of daily life that have not yet been published.37 The place of discovery is still
not entirely certain; the editors judge that it was fairly close to Dura, perhaps the Syrian
village of Beth Phouraia on the Euphrates (where many of the people mentioned in the
dossier lived) or, perhaps more likely, Appadana/ Neapolis, the administrative center on
which Beth Phouraia depended. Appadana, already known from the documents of Dura, is
not well localized, but its name seems to survive in the present-day toponym, Tell Fudayn,
on the Lower Khabour, the ancient Aboras, an eastern tributary of the Euphrates.

Page 5 of 23
The Papyrology of the Near East

The nineteen publishable texts, datable to the years 232 to 252 and consisting of peti­
tions, private legal documents, and letters, have all been published.38 Even before their
publication, Fergus Millar pointed out their unusual historical interest.39 They do indeed
help us to better understand the government of Roman Mesopotamia, the frontiers of Syr­
ia (which must now be pushed beyond the Euphrates along the Khabour), law, local au­
thorities,40 the army, and the municipal and religious institutions of cities like Nisibis,
Marcopolis, and Carrhai. They illumine the end of the Abgarid monarchy at Edessa
(P.Euphr.19, 20). The petition P.Euphr. 1 is addressed to a historical figure, the prefect of
Mesopotamia and consular governor of Syria, Julius Priscus, brother of the emperor
Philip. The contribution of these texts to onomastic studies and to Syro-Mesopotamian
historical geography is also noteworthy and received particular attention from the edi­
tors. So far, however, the Euphrates texts have not evoked quite the same degree of inter­
est as those from Nahal Hever (discussed later), as one can see by comparing the modest
scholarly bibliography the former have generated to the mass of writing on the latter.41

Nahal Hever

Hellenistic and Roman Judaea and Samaria have furnished numerous texts, most often
found in the caves of the wadis emptying toward the Dead Sea. It is difficult to describe
here all of these discoveries, some of which, like those of Qumran, are (p. 479) not directly
connected to papyrology.42 Some of these sites have already been mentioned,43 including
Wadi Daliyeh and other places in the vicinity of Jericho,44 Khirbet Mird, Murabbaʼat,45
and Masada.46 Still-unpublished texts from Wadis Nar and Ghweir are known, and some
Greek texts from the caves of Engadi have recently been published.47 The texts from Na­
hal Hever, south of Engadi, form a discrete body both in their number and in their
archival coherence. They come both from chance finds and from regular excavations car­
ried out in 1960 and 1961 by an Israeli team in the cave called the “cave of horror” be­
cause of the human remains and especially in the “cave of the letters,” by far the richer of
the two. These have yielded biblical fragments48 and documents in Greek, Hebrew, Ara­
maic, and Aramaic-Nabataean, belonging to three dossiers, the “archive” of Babatha and
that of Salome Komaise, as well as a packet of fifteen letters addressed by Bar Kokhba be­
tween 132 and 135 to his lieutenants at Engadi.49 This dossier of “Simon Choseba” in­
cludes at least two letters written in Greek,50 with the balance in Hebrew (a linguistic re­
vival corresponding to the messianic spirit of the revolt) and in Aramaic. There is also one
tablet in Aramaic.51 The whole witnesses to the prince's ability to create a political struc­
ture and to rally around him non-Jewish elements of the population. The two other groups
(sale contracts, a donation, petitions, property declarations, receipts, mortgage loans,
marriage contracts) concern two Jewish women of Arab-Nabataean origin (from the vil­
lage of Mahoza, “the port,” on the south of the Dead Sea, in the territory of Petra) but
who had interests in Judaea in the region of Engadi. These dossiers are linked at several
points and break off in 131/132, that is, at the moment of the Bar Kokhba revolt, during
which these women, after having hidden their papers, lost their lives or were otherwise
unable to recover their documents.

Page 6 of 23
The Papyrology of the Near East

Salome Komaïse (daughter of Levi and of Salome alias Groptê) (date range, January 125
to August 131) was married twice and kept papers that belonged to her first husband
(P.Hever 12, 60–65), including two census declarations subscribed at Rabbath-Moab in
Arabia (P.Hever 61 and 62) and the interested party's dowry receipt (P.Hever 65).

The much richer dossier of Babatha, daughter of Simon, covers the years 94 to August
132 and straddles a major historical change, the disappearance of the Nabataean king­
dom and the creation in 106 of the Roman province of Arabia. Of the 35 pieces, 17 are in
Greek, 9 bilingual, and the remainder in Aramaic-Nabataean and Aramaic. They are con­
nected with Babatha's economic struggles with her second husband's first wife. One piece
that is particularly interesting from the point of view of public procedure is a census dec­
laration subscribed at Petra (P.Yadin I 16). Another, no. 10, Babatha's Aramaic marriage
contract, is an excellent piece for jurists studying the position of the Jews.

The documents from Nahal Hever have aroused sustained interest, considering the criti­
cal period in the history of ancient Judaism that they span and the light they throw on the
situation of Jews in a Hellenized and Romanized non-Jewish (p. 480) environment.52 From
them we now have a better understanding of the institutions and political structures of
Judaea and of the province of Arabia. The literature concerning these documents has be­
come extensive, particularly the numerous works of Hannah Cotton.53

Petra

“Mute” until 1993, the Arabo-Nabataean site of Petra produced a large cache of car­
bonized fragments discovered in the ruins of a serving room of the principal church. Their
publication is still in progress, slowed by the extreme difficulties of consolidating, plac­
ing, and mounting the fragments, but already a considerable amount of material is avail­
able in two volumes.54 About fifty pieces ranging in date from 537 to 593 have been iden­
tified; thus, they are contemporary with the Byzantine portion of the Nessana dossier,
with which the Petra papyri have similarities from the point of view of diplomatics and
onomastics; for these two sites, both of which were part of the Byzantine province of
Palestine III during this period, shared the same Arabo-Nabataean heritage. Even the sit­
uations of the finds are analogous. The Petra dossier is in large part centered on the fami­
ly of Theodoros, son of Obodianos (a distinctive Nabataean patronymic), a member of the
local clergy (another point of similarity with the Nessana papyri). The papyri reveal not
the commercial milieu of a caravan city but a settled, civic world of landowners and cler­
gy living on the revenues from their landholdings. The documents have close parallels in
the diplomatics of contemporary Egyptian material (settlements, or dialuseis,55 as well as
requests for corrections in land registers, or epistalmata tou somatismou).Tax receipts
conform to Justinian's laws in specifying the number of assessment units (juga) and the
status of parcels of land. A settlement document comes from Gaza in Palestine I (P.Petra I
2). As at Nessana, the local place names are largely of Arabic derivation, and one of the
editors has shown that several place names in P.Petra have survived to the present.56
They may even go back to the Nabataeans, who are considered an Arabophone people

Page 7 of 23
The Papyrology of the Near East

who were writing in Aramaic. Their role and place in the Arab conquest of the years
635/642 deserve a fresh examination.

Nessana

If Byzantinists have been slow to take into account the Egyptian papyri of late antiquity,
they have been quicker to take an interest in those of Nessana.57 This Byzantine and
Umayyad corpus (spanning the years from 500 to 700) was discovered in 1935 at el-Auja,
on the frontier between the Sinai and the Negev, in the ruins of the village of Nessana,
which, like Petra, was part of the Byzantine province of Palestine III, a product of the
breakup of the old province of Arabia. The Nabataean (p. 481) cultural and linguistic base
is still visible, notably in personal names. These papyri were probably discarded, but, in­
stead of being destroyed, they were kept in rooms attached to religious establishments.
These texts are essentially Greek, but some are bilingual Greek-Arabic or entirely Arabic.
Not all of them were published,58 and the approximately two hundred literary and docu­
mentary pieces available in two volumes would benefit from revision.

The literary dossier is of great cultural interest because it includes a glossary of the
Aeneid and fragments of the epic, works one would not have expected in the sixth centu­
ry. Two fragments of a codex concerning the law of succession deserve further attention
from legal historians (P.Ness.11 and 12). The Byzantine portion of the documentary mate­
rial consists of papers of soldiers who belonged to the numerus of the devotissimi Theo­
dosiaci, attested from 505 to 596. A certain Sergius, who came from this milieu, became
the head (hegoumenos) of the local martyrion of saints Sergius and Bacchus at the end of
the sixth century. This role passed afterward to his son, Patrikios, and indeed to further
descendants down to the end of the seventh century. Under the Arabs, Georgios, the next
to the last representative of this family, became the civil and financial administrator of
Nessana, with the result that Greek, bilingual, and Arabic administrative and fiscal docu­
ments became intermixed with the family papers. The archive is thus particularly pre­
cious for the history of the institutions and finances of Palestine under the Umayyads. As
is the case at Dura, interpretation of the Nessana papyri also demands attention to the
Greek and Nabataean epigraphic texts from the site, in which some of the hegoumenoi of
the papyri reappear.59

Palaeography
This aspect of the subject, which requires mastery of many variables and a good knowl­
edge of the Egyptian texts, has been treated by E. Crisci with prudence, perspicacity, and
originality, approaching the subject as a cultural and social phenomenon.60 The handwrit­
ings of the early Hellenistic period, offspring of the bilinear classical majuscule, are not
very differentiated.61 With later political fragmentation, however, regional tendencies
emerge, as in the Parthian documents from Avroman.62 In the Roman period, the Near
East witnessed the development of its own style, which, earlier than in Egypt, tends to ex­
aggerate the contrasts between letters, announcing in this fashion the quadrilinearity

Page 8 of 23
The Papyrology of the Near East

that characterizes Byzantine handwriting.63 Regional differences blur under the Arabs:
Egypt and the East use the same quadrilinear, high-contrast cursive, and, for fiscal texts,
a form of minuscule.

(p. 482) Formats and Diplomatics

One of the oldest, most durable, and most constant traits of Near Eastern documents is
their production in a double copy, the so-called diploma. The full and outwardly visible
copy of a document (scriptura exterior) was headed, in a more limited space, by a cursive,
compact, and often abridged version. This copy was then folded, sewn, and sometimes
sealed, a process that has given it the designation of scriptura interior. The scriptura inte­
rior provided an authoritative copy in case of dispute over any aspect of the scriptura ex­
terior.64

According to the allusions to this format in the Roman papyri of Egypt, like the registra­
tion of a slave sale that had taken place at Bostra,65 its technical name was diplôma hel­
lênikon (“Greek double document”). The ancients thus took a position in favor of a Greek
origin for this format, whereas modern papyrologists do not exclude an encounter be­
tween Greek tradition and Roman customs or even practices of Semitic origin. The diplô­
ma of Bostra must have been on papyrus because, to judge particularly from the finds of
Nahal Hever, most of the diplômata that have come from the Arabo-Nabataean zone
adopted this writing material. The writing is in this case parallel to the height of the origi­
nal roll (transversa charta), in the format that papyrologists call rotulus.66 In the Syro-
Mesopotamian zone, as we may deduce from the sealings of Palmyra and as we see in
documents from Dura, the Middle Euphrates, Avroman, and Bactria, the diplôma is gener­
ally written on rotuli of skin.

The double document has other characteristics as well, such as the number of points at
which the scriptura interior is closed, which varies from five to seven,67 and these points
are often validated on the back by the subscriptions of witnesses.

The Roman calendar is universally used in the Near East (P.Hever, pp. 146–149), with the
consular formula and Macedonian or Roman months, supplemented by local or regional
eras (e.g., era of Arabia, Seleucid era, era of Gaza). Regnal years are given much less reg­
ularly.

Language
The Greek of the Near Eastern papyri is part of the common Greek dialect called koinê.
Although there have been many attempts to explain its distinctive characteristics by a
Semitic substrate,68 the editors of P.Euphr., in keeping with contemporary scholarly
trends, have preferred to connect linguistic anomalies to the general trends of the com­
mon Greek speech of the Roman and Byzantine periods. (p. 483) They also mention the in­
fluence of administrative Latin not only in the vocabulary but also in syntax. Moreover,

Page 9 of 23
The Papyrology of the Near East

this Greek contains words and expressions (e.g., idioms of an undefined origin) not used
in the Greek of Egypt. Naphtali Lewis made many observations in the introduction to
P.Yadin I (notably on the verb καθαροποιέω, in connection with guarantees against evic­
tion), and Hannah Cotton has called attention to the use of χειροχρήστης for ύπογραφεύς
in P.Hever 64 v°). Peculiarities of the same sort have been observed in the dossier of Beth
Phouraia/Appadana, for example, in connection with the term συγκωμήτης (P.Euphr.1 and
4), found only in the Near East and seemingly a calque of the Latin convicanus.

Was koinê a living language? In the Jewish context, Greek seems linked to Roman domi­
nation and perhaps has a somewhat superficial character (P.Hever, p. 146). In Arabia,
similarly, Greek was not adopted until after the disappearance of the Nabataean kingdom
in 106 CE. Moreover, local languages were far from being eliminated, as the Semitic sub­
scriptions to Greek documents from Nahal Hever and the Middle Euphrates demonstrate.
In as important a center as Edessa, Syriac kept its status as the official language under
Roman rule (P.Dura 28), a privilege that helps to explain the city's later distinction in Syri­
ac literature. In the late texts of Petra and Nessana, Arabic surfaces in personal and place
names.

Law
Few scholars would deny the early advance of Roman norms in the domain of provincial
and municipal institutions, as well as public law, well before the Antonine Constitution of
212.69 The study of private law—of areas like marriage contracts, sales, and gifts—is by
contrast dominated by the theme of the interaction of local legal systems, Hellenistic and
Semitic, with Roman law, often with mixed results.70 This approach is logical enough for
the Judeo-Nabataean archives of the second century, given the diversity of the languages
attested in this body of about a hundred documents. According to a recent scholarly
workshop,71 it seems that the juridical concepts used depended on the place and period
in which documents were drawn up,72 as well as on the language. Acts drawn up in Greek
or written in the Roman province of Arabia diverge the most from the legal prescriptions
of the Jewish sources, most notably in matters concerning the legal situation of women.
Thus, in the documents from Nahal Hever, the woman is accompanied by an epitropos in
documents drawn up under the Roman legal system but not in Hebrew, Nabataean, and
Aramaic texts drawn up outside the Roman empire (i.e., in the Nabataean kingdom or un­
der the Bar Kokhba regime).

So far as the diffusion of Roman law itself goes, the editors of P.Yadin I, like those
(p. 484)

of P.Euphr., have noticed the conformity of these private documents to Roman legal pre­
scriptions: thus, the formula of stipulatio, characteristic of Roman legal acts (έπερωτηθεὶς
ώμολόγησα, interrogatus spopondi), the reference to bona fides, and the role of notaries
(librarii, νομικοί) in the drawing up of documents. In addition, P.Yadin I 28–30 provide
three copies of the Greek formula for a procedural action concerning guardianship, in
conformity with the formula for the actio tutelae in the praetor's edict. P.Euphr.1 and 2,

Page 10 of 23
The Papyrology of the Near East

for their part, appear to refer, in the editors' view, to the interdicts of the praetorian edict
(unde vi, uti possidetis).This interpretation has not been challenged to date.

Near Eastern Papyrology and Egyptian Papyrol­

ogy
We have no basis for a systematic comparison of the Hellenistic papyri of Egypt and those
from the Near East, the latter having nothing on the scale of the compact but voluminous
archives of Zenon (for the third century BCE) or those of the agoranomoi of Pathyris (for
the second–first century). It is only for the Roman and Byzantine periods that serious
comparison is possible. For some decades now there has been a consensus that Egypt af­
ter Augustus was fundamentally a Roman imperial province, even if it retained—more in
form than in substance—some characteristics of the preceding Ptolemaic monarchy.73
This province had a special status, created by Augustus, at first sight somewhat reduced
by the near-total absence of civic institutions, but recent research has suggested that a
sort of unofficial municipal life was already in place in Egypt by the second century CE.

Although this view is largely based on the information provided by the papyri, it is con­
cerned with problems of a historical rather than strictly papyrological nature because pa­
pyrologists are in general more interested in studying the formal character of the texts
than their content. For such formal questions, the written documentation from the Roman
Near East is now sufficient in quantity and coherence to allow comparisons with the
Egyptian papyri. This is what the editors of P.Yadin I and P.Euphr. have attempted in seek­
ing to bring out the distinctive characteristics of each corpus and, to the extent that the
written documents reflect their society of origin, regional particularisms. Indeed, the
Near Eastern documentation—which still has many gaps, as, for example, with accounts,
compared to the Egyptian papyri—does not offer a homogeneous typology but numerous
variations of format and diplomatics in different times and places.

“Rome is the same everywhere,” said the editors of P.Euphr.(66).This assertion is

(p. 485)

valid particularly for public documents such as petitions to governors and police officers,
the phraseology and diplomatic format of which (the hypomnêma) are much the same in
Egypt, Syria, and Arabia. Military archives, too, are relatively standardized.

In the domain of private law, however, the differences are more distinctive and, as we
shall see, more significant. Like the rest of the Near East, Hellenistic Egypt used the dou­
ble document,74 but this format disappeared in Egyptian practice after the Roman con­
quest (see the section titled “Formats and Diplomatics”). Where they still occur, one can
almost always show that the “diploma” was drawn up outside Egypt or comes from mi­
lieus with their own legal systems, like the military and Roman citizens. Some scholars
see in these a transposition from the tablet diptychs normally used for Roman wills and
call these Roman double documents.75 Similarly, in the case of sale contracts, the diplo­
matics of the bilateral record of oral proceedings found in P.Euphr.6–10 and known at Du­
ra is unknown in Egyptian documents after the Hellenistic period, even though the formu­

Page 11 of 23
The Papyrology of the Near East

las concerning eviction present in all of the texts do not appear in Egypt until the Byzan­
tine period.76 The Roman stipulation, present already in the second century in Arabia,
does not become common in Egypt until after the Antonine Constitution of 212. Notarial
subscriptions are unknown in Egypt until the Byzantine period, while tabelliones, called li­
brarii or nomikoi, are involved in the drawing up of legal documents in the Roman Near
East, again as early as the second century. Equally significant differences appear in dat­
ing practices: Consular formulas are rare in Egypt before the tetrarchy, while regnal
years served as the standard dating system under the principate.

Some internal matters also deserve mention. The denarius is universally used as the unit
of currency in Near Eastern documents, while in Egypt the drachma remains the stan­
dard unit until 296. This is only an example of the many details that would lead us to the
conclusion that in the sphere of private law Egypt was Romanized more slowly than the
rest of the Near East. Although it was a Roman imperial province, it was in some respects
insular.

From the papyrological point of view, the Byzantine period offers a reversed perspective.
From the fourth century on we see a kind of meeting of the two cultures, each apparently
borrowing traits from the other: Handwriting is unified around a system foreshadowing
the minuscule script; the double document disappears in the Near East except in areas
outside Roman control (e.g., central Asia); Egypt adopts the transversa charta document
format; legal vocabulary from the Near East turns up in Egypt (thus the verb
καθαροποιεîν, which appears only in the sixth century); the notariat is introduced there
as well; denarii and myriads of denarii appear in papyri next to drachmas and talents; and
consular dates are used next to regnal dating and then, after that vanishes, next to the in­
diction year.77 This set of changes undoubtedly has much to do with the integration of
Egypt into the large regional diocese of Oriens, itself under the authority of the praetori­
an prefect (p. 486) of the East, which would have led it increasingly to share the same in­
stitutions and the same civil and administrative law as Syria and the Arabo-Palestinian
provinces. Some differences remained, certainly. For example, to judge from the Petra pa­
pyri, the introduction of the regnal year as the principal dating criterion came about
there soon after the legislation on the subject promulgated in 537 (Nov. Just. 44), while
Egypt lagged by a couple of years. The Petra papyri help to highlight a number of other
Egyptian peculiarities, such as in the taxation of land, where although Egypt shared fun­
damental principles with the rest of the East (taxation by rate rather than by partition), it
did not apply them in the same manner and with the same terminology (Egypt knew noth­
ing of the iugum).It is possible that the administrative separation of Egypt and the Near
East, which occurred in 380, when the diocese of Egypt was created, brought on a new
wave of local particularisms.

Bibliography
Auda, Y., and M.-F. Boussac. 1996. “étude statistique dʼun dépôt dʼarchives à Déios.” In
Archives et sceaux du monde hellénistique: Archivi e sigilli nel mondo ellenistico, ed. M.-
F. Boussac and A. Invernizzi, 511–523. Paris: De Boccard.

Page 12 of 23
The Papyrology of the Near East

Bérard, F., D. Feissel, P. Petitmengin, D. Rousset, and M. Sève. 2000. Guide de

lʼépigraphiste: Bibliographie choisie des épigraphies antiques et médiévales, 3d ed. Paris:
Éditions Rue dʼUlm.

Bernard, P., and C. Rapin. 1994. “Un parchemin gréco-bactrien dʼune collection privée.”
CRAI: 261–294.

Bowman, A. K. 1976. “Papyri and Roman Imperial History, 1960–1975.” JRS 66: 153–173.

Brock, S. 1991. “Remarks on P.Euphr. 20.” Aram 3: 259.

Cadell, H. 1989. “Les archives de Théophanès dʼHermoupolis: Documents pour lʼhistoire.”

In Egitto e storia antica: Atti del Colloquio internazionale, Bologna, 31.8–2.9.1987, ed. L.
Criscuolo and G. Geraci, 315–323. Bologna: Cooperativa Libraria Universitaria Bologna.

Clarysse, W., and D. Thompson. 2007. “Two Greek Texts on Skin from Hellenistic Bac­
tria.” ZPE 159: 273–279.

Clarysse, W., and K. Vandorpe. 1995. Zénon, un homme dʼaffaires grec à lʼombre des pyra­
mides. Leuven: Universitaire Pers Leuven.

Cohen, N. 2006. “New Greek Papyri from a Cave in the Vicinity of Ein Gedi.” SCI 25: 87–
95.

Cohen, N. 2007. “New Greek Papyri from a Cave in Mount Yishai, near Ein-Gedi.” In J.
Frösén, T. Purola, E. Salmenkivi eds., Proceedings of the 24th International Congress of
Papyrology Helsinki, 1–7 August, 2004, vol 1: 191–197, with plate Vof vol. 2. Commenta­
tiones Humanarum Litterarum 122:1. Helsinki: Societas Scientiarum Fennica.

Cotton, H. M. 1999. “Some Aspects of the Roman Administration of Judaea/Syria-Palaesti­

na.” In Lokale Autonomie und römische Ordnungsmacht in den kaiserzeitlichen Provinzen
vom 1. bis 3. Jahrhundert, ed. W. Eck and E. Müller-Luckner, 75–91. Schriften des His­
torischen Kollegs, Kolloquien, 42. Munich: R. Oldenbourg Verlag.

———. 2003. “The Bar Kokhba Revolt and the Documents from the Judaean Desert:
Nabataean Participation in the Revolt (P. Yadin 52).” In The Bar Kokhba War Reconsid­
ered: New Perspectives on the Second Jewish Revolt against Rome, ed. P. Schäfer, 133–
152. Tübingen: Mohr.

———, W. E. H. Cockle, and F. G. B. Millar. 1995. “The Papyrology of the Roman Near
East: A Survey.” JRS 85: 214–235.

Cotton, H. M., and W. Eck. 2002. “P. Murabbaʼat 114 und die Anwesenheit römischer
Truppen in den Höhlen des Wadi Murabbaʼat nach dem Bar Kochba Aufstand.” ZPE 138:
173–183.

Cowley, A. 1919. “The Pahlavi Document from Avroman.” JRAS, n.s., 51: 147–154.

Page 13 of 23
The Papyrology of the Near East

Crisci, E. 1996. Scrivere greco fuori dʼEgitto: Ricerche sui manoscritti greco-ori­
(p. 492)

entali di origine non egiziana dal IV secolo a.c. allʼVIII d.c. Papyrologica Florentina 27.
Florence: Edizioni Gonnelli.

Cumont, F. 1926. Les fouilles de Doura-Europos, 1922–1924. Paris: Librairie orientaliste

Paul Geuthner.

Daniel, R. W. 2001. “P. Petra Inv. 10 and Its Arabic.” In Atti del XXII Congresso Inter­
nazionale di Papirologia, ed. I. Andorlini, G. Bastianini, M. Manfredi, and G. Menci, vol. 1,
331–341. Florence: Istituto Papirologico “G. Vitelli”.

Drexhage, H.-J. 1998. “Ein Monat in Antiochia: Lebenshaltungskosten und Ernährungs-

verhalten des Theophanes im Payni (26. Mai–24. Juni) ca. 318 n.” MBAH 17: 1–10.

Durand, X. 1997. Des Grecs en Palestine au IIIe siècle avant jésus-Christ: Le dossier
syrien des archives de Zénon de Caunos (261–252). Paris: Gabalda.

Edmonds, C. J. 1952. “The Place-names of the Avroman Parchments.” BSOAS 14: 478–482.

Feissel, D., and J. Gascou. 1989. “Documents dʼarchives romains inédits du Moyen Eu­
phrate (IIIe siècle après J.-C.).” CRAI: 535–561.

Fink, R. O. 1971. Roman Military Records on Papyrus. American Philological Association

Monograph 26. Cleveland: Case Western Reserve University Press.

Fournet, J.-L. 2003. Review of P.Petra I. An Tard 11: 398–404.

Gascou, J. 1999. “Unités administratives locales et fonctionnaires romains: Les données

des nouveaux papyrus du Moyen Euphrate et dʼArabie.” In Lokale Autonomie und römis­
che Ordnungsmacht in den kaiserzeitlichen Provinzen vom 1. bis 3. Jahrhundert, ed. W.
Eck and E. Müller-Luckner, 61–73. Schriften des Historischen Kollegs, Kolloquien, 42.
Munich: R. Oldenbourg Verlag.

Geraci, G. 1983. Genesi della provincia romana dʼEgitto. Bologna: Cooperativa Libraria
Universitaria Editrice Bologna.

Gnoli, T. 2000. Roma, Edessa, e Palmira nel III sec. d.c.: Problemi istituzionali: Uno studio
sui papiri dellʼEufrate. Biblioteca di Mediterraneo Antico 1, Pisa-Rome: Istituti Editoriali e
Poligrafici Internazionali.

Grohmann, A. 1963. Arabic Papyri from Hirbet el-Mird. Leuven: Publications universi­
taires.

Guglielmo, L. 2003. “Micae Qumranicae, I manoscritti di Qumran a quasi sessantʼanni

dalla scoperta.” Papyrologica Lupiensia 12: 99–114.

Hadot, P., and C. Rapin. 1987. “Les textes littéraires grecs de la Trésorerie dʼAï
Khanoum.” BCH 111: 225–266.

Page 14 of 23
The Papyrology of the Near East

Haensch, R. 2001. “Zum Verständnis von P. Jericho 16 gr.” SCI 20: 155–167.

Hengstl, J. 2002. “Die byzantinischen Papyri aus Petra: Stand der Bearbeitung und Bitte
um Unterstüatzung.” RIDA 3e s.49: 341–357.

Hesseling, D. C. 1893. “Tabulae ceratae assendelftianae.” JHS 13: 293–314.

Hollis, A. S. 1996. Addendum to Rea, Senior, and Hollis 1994. ZPE 110: 164.

Invernizzi, A. 1996. “Gli archivi pubblici di Seleucia sul Tigri.” In Archives et sceaux du
monde hellénistique: Archivi e sigilli nel mondo ellenistico, ed. M.-F. Boussac and A. Inv­
ernizzi, 131–143. Paris: De Boccard.

Katzoff, R., and D. Schaps. 2005. Law in the Documents of the Judaean Desert. Leiden:
Brill.

Kennedy, D. L. 1994. “The Cohors XX Palmyrenorum at Dura Europos.” In The Roman and
Byzantine Army in the East: Proceedings of a Colloquium Held at the Jagiellonian Univer­
sity, Krakow in September 1992, ed. E. Dabrowa, 89–98. Cracow: Uniwersytet Jagiellon­
skiego.

Koenen, L. 1996. “The Carbonized Archive from Petra.” JRA 9: 177–188.

(p. 493) Kruit, N. 2004. Review of P.Petra I. BibO 61: 98–106.

Lerner, J. D. 2003. “The Aï Khanoum Philosophical Papyrus.” ZPE 142: 45–51.

Lewis, N. 1970. “‘Graeco-Roman Egypt’: Fact or Fiction?” In Proceedings of the Twelfth

International Congress of Papyrology, 3–14. Am.Stud.Pap. VII. Toronto: A.M. Hakkert.

———. 1974. Papyrus in Classical Antiquity. Oxford: Clarendon.

Migliardi Zingale, L. 1999. “Diritto romano e diritti locali nei documenti del Vicino Ori­
ente.” SDHI 65: 217–231.

Millar, F. 1993. The Roman Near East, 31 BC-AD 337. Cambridge, Mass.: Harvard Uni­
versity Press.

Minns, E. H. 1915. “Parchments from the Parthian Period from Avroman in Kurdistan.”
JHS 35: 22–65.

Mitthof, F., and A. Papathomas. 2004. “Ein Papyruszeugnis aus dem spätantiken Karien.”
Chiron 34: 401–424.

Nyberg, H. S. 1923. “The Pahlavi Documents from Avroman.” Le Monde Oriental 17: 182–
230.

Otranto, R. 1998. Review of Crisci 1996. QS 24: 197–206.

Page 15 of 23
The Papyrology of the Near East

Perrot, C. 1963. “Un fragment christo-palestinien découvert à Khirbet Mird.” RBi 70: 506–
555.

Pollard, N. 1996. “The Roman Army as ‘Total Institution’ in the Near East? Dura-Europos
as a Case Study.” In The Roman Army in the East, ed. D. L. Kennedy, 211–227. JRA Sup­
plementary Series 18. Ann Arbor: Journal of Roman Archaeology.

Rapin, C. 1996. “Nouvelles observations sur le parchemin gréco-bactrien dʼAsangôrna.”

Topoi 6: 458–469.

Rea, J. R., R. C. Senior, and A. S. Hollis. 1994. “A Tax Receipt from Hellenistic Bactria.”
ZPE 104: 261–280.

Reed, R. 1972. Ancient Skins, Parchments, and Leathers. New York: Seminar Press.

Reekmans, T. 1998. Review of Durand 1997. CdʼE 73: 144–158.

Saliou, C. 1992. “Les quatre fils de Polémocratès(P. Dura 19): Texte et archéologie.” Syria
69: 65–100.

Schiffman, L. H., and J. C. VanderKam, eds. 2000. Encyclopedia of the Dead Sea Scrolls. 2
vols. New York: Oxford University Press.

Schwabe, M. 1945. “Khirbat Mafjar, Greek Inscribed Fragments.” QDAP 12: 20–30.

Seyrig, H. 1985. “Cachets dʼarchives publiques de quelques villes de la Syrie romaine.” In

Scripta varia: Mélanges dʼarchéologie et dʼhistoire, 417–441. Paris: Librairie orientaliste
Paul Geuthner

; reprinted from
Mélanges de lʼUniversité Saint-Joseph 23 (1940) 85–107.

Sims-Williams, N. 1996. “Nouveaux documents sur lʼhistoire et la langue de la Bactriane.”

CRAI: 633–654.

Stephens, F. J. 1937. “A Cuneiform Tablet from Dura-Europos.” RAssyr 183–189.

Stolte, B. H. 2001. “The Impact of Roman Law in Egypt and the Near East in the Third
Century A.D.: The Documentary Evidence. Some Considerations in the Margin of the Eu­
phrates Papyri (P. Euphr.).” In Administration, Prosopography, and Appointment Policies
in the Roman Empire: Proceedings of the First Workshop of the International Network
Impact of Empire (Roman Empire, 27 B.C.–A.D. 406), ed. L. de Blois, 167–179. Amster­
dam: J. C. Gieben.

Teixidor, J. 1990. “Deux documents syriaques du IIIe siècle après J.C., provenant du
Moyen Euphrate.” CRAI: 144–146.

———. 1991. “Un document de fermage de 242 après J.C.” Semitica 41–42: 195–208.

Page 16 of 23
The Papyrology of the Near East

Tov, E. 1990. The Greek Minor Prophets Scroll from Nahal Hever (8HevXIIgr).
(p. 494)

Seiyâl Collection, vol. 1, with R. A. Kraft and P. J. Parsons. Oxford: Clarendon Press.

———. 2002. “The Discoveries in the Judaean Desert Series: History and System of Pre­
sentation.” In E. Tov, M. Abegg, Jr. et al., The Texts from the Judaean Desert: Indices and
an Introduction to the Discoveries in the Judaean Desert Series. Oxford: Clarendon Press.

———, and S. Pfann. 1993. The Dead Sea Scrolls on Microfiche: A Comprehensive Facsim­
ile Edition of the Texts from the Judean Desert. Leiden: Brill.

van Haelst, J. 1991. “Cinq textes grecs provenant de Khirbet Mird.” AncSoc 22: 297–317.

Youtie, H. C. 1973. “Ostraca from Sbeitah.” Scriptiunculae, vol. 2, 982–988. Amsterdam:

A. M. Hakkert

(reproduced from AJA 40 [1936] 452–459).

Sources not included in the Checklist

I.Ness. = . = H. D. Colt Jr. et al., Excavations at Nessana, vol. 1, 131–210 (by G. E. Kirk, C.
B. Welles, and F. Rosenthal). London 1962: British School of Archaeology in Jerusalem.

P.Daliyeh = = D. Gropp, Wadi Daliyeh. Vol. 2, The Samaria Papyri from Wadi Daliyeh, ed.
E. Schuller et al., in consultation with J. VanderKam and M. Brady, Qumran Cave 4.
XXVIII: Miscellanea, Part 2. Oxford: Clarendon, 2001.

Notes:

(1.) Cotton, Cockle, and Millar (1995, 214–235). These authors take account of texts as­
signable to non-Roman states; they register unpublished texts, documents of uncertain
origin, and Near Eastern texts found in Egypt. For an update to this fundamental work
see Mitthof and Papathomas (2004, 401–424, especially their notes 1–3).

(2.) Ostraca are less numerous than in Egypt, even though editors often include among
them jar inscriptions that papyrologists would class among the instrumenta. An uncom­
mon rigid material, unknown at least in Egypt but widespread under the Umayyads, is
marble found in the ruins of ancient monuments and containing Arabic and Greek texts
sometimes of a very everyday nature. These have been found at Nessana, in the Negev
(published in I.Ness.9–11), at Khirbat al-Mafjar near Jericho (Schwabe 1945), at Qasr al-
Hayr al-Gharbi, and Andarin in Syria (partly unpublished). This practice existed in the Ro­
man period but was probably rare; an example with a Greek letter was found at Izmir.

(3.) See Reed (1972). Editors use the words parchment and leather without technical
study. For the widespread use of skins in the Persian Empire see Lewis (1974, 8–9). But
papyrus was not entirely lacking before the arrival of the Greeks in regions near Egypt
(Phoenicia and Palestine).

Page 17 of 23
The Papyrology of the Near East

(4.) Minns (1915, 22–65). These are rotuli dated respectively to 88/87 and 22/21 BCE, reg­
istering sales of land (vineyards) and using Greek law. They were accompanied by a sale
of a vineyard in Pehlevi in 13/12 BCE or 53 CE, depending on interpretation of the era
date (Cowley 1919, 147–154). See Nyberg (1923) and Edmonds (1952).

(5.) The text, attributable to 167 or 181 BCE, comes from Sangcharak, southwest of Aï
Khanoum, according to Bernard and Rapin (1994); see also Rea, Senior, and Hollis
(1994), with the addendum of Hollis (1996), and Rapin (1996). Two Hellenistic texts from
Afghanistan were published by Clarysse and Thompson (2007). One of the literary manu­
scripts from Aï Khanoum was also on “parchment.”

(6.) Sims-Williams (1996); he describes a body of documents and letters on skin from
northern Afghanistan. Their chronological range is 342 to 781. A number of these are
double documents with five holes closing the scriptura interior as in much earlier texts
from Syria, Palestine, and Nabataean Arabia (see below the section titled Formats and
Diplomatics).

(7.) See below the section titled The Five Main Dossiers.

(8.) See, particularly for Palmyra, Alexandretta, and Dolichê, Seyrig (1985).

(9.) Compare P.Euphr. 19 and 20 and P.Euphr. 6–7.

(10.) For this continuation of Syriac, see P.Dura 28. The official bilingualism of Palmyra is
attested mainly by inscribed monuments, but it can hardly be doubted that Palmyrene
written production had a similar pattern.

(11.) See the section titled The Five Main Dossiers.

(12.) Letters of Toubias, Hellenistic letters from Caria (P.Cair.Zen. I 59036 and 59056, of
257 BCE); letters from soldiers of the Legio III Cyrenaïca recently relocated to Bostra
(P.Mich. VIII 466 [107] and 465 [108]); Roman letter from a military milieu written at An­
tioch (BGU III 794, with an allusion to vows before the Tyche of Antioch); letter of a
woman to her aunt, written at Apamea (P.Bour. 25, fifth century).

(13.) P.Cair.Zen. I 59003 = CPtol.Sklav. 37 (Ammanitis, 259 BCE); BGU 1316 = MChr. 271
= FIRA III 135 (Ascalon, 359). Several slave sales found in Egypt were drawn up in Asia
Minor: in Rhodes (P.Oxy. L 3593 [238–244]), Myra (BGU III 913 [206]), Side (P.Turner 22
[142], BGU III 887 = MChr272 = FIRA III 133 = CPJ III 490 [151]), Seleucia in Pieria (Jur­
Pap 37 = FIRA III 132 = ChLA III 200 = CPL 120 [in Latin, 166]). These pieces provide
valuable information about the institutions and law of the cities in question.

(14.) SB XII 11043, found in the Fayyum and dating to 152, is a report of proceedings be­
fore the procurator of Palestine. An order for payment of annonai and of capita, dated to
293, in Latin with a Greek subscription, mentions Caesarea (of Palestine?). It is not cer­
tain that it was written in Egypt (SB XVIII 13851). In addition, P.Lips.34, a Hermopolite
receipt for vestis militaris, mentions a payment made around 375 in Hierapolis in Syria.

Page 18 of 23
The Papyrology of the Near East

There is a circular of the comes Orientis on the movement of recruits, written at Antioch
around 380 (W.Chr.469).

(15.) For instance, BGU III 895 is a succession agreement from a Syrian milieu. Also
worth mentioning is a Byzantine petition or letter from Aphrodisias (end of the sixth or
start of the seventh century), with an account in grain on the verso (see Mitthof and Pap­
athomas 2004), and P.Münch. III 43, a contract of Bithynian origin dated to 248.

(16.) Clarysse and Vandorpe (1995, 24–25, 90–92). See the collection of Durand (1997)
with the review by Reekmans (1998).

(17.) Cadell ( 1989); Drexhage (1998).

(18.) P Dura 28.

(19.) P Euphr.6–7.

(20.) P Euphr.10.

(21.) P Euphr.1.

(22.) P Petra I 2.

(23.) The extent of our loss can be measured for Seleucia on the Tigris, where Italian ex­
cavations found more than 25,000 seals in the archives building, which was burned after
155/154 BCE, when Babylonia passed from Seleucid to Arsacid control. Most of these
concern a salt tax (halikê). Invernizzi (1996, 133) mentions papyrus and parchment, but
he describes neither the proportions nor how the material was observed. The rescue ex­
cavations at Zeugma have led to the discovery of 65,000 imprints of Hellenistic seals, a
deposit appropriate to this center's importance as a customs house. One may compare
the 16,000 seals found in a house on Delos that was burned in 69 BCE (Auda and Boussac
1996, 511).

(24.) Hadot and Rapin (1987). A fragment of papyrus and two fragments of
“parchment” (mid-third to early second century). The first is a philosophical dialogue on
ideas, the second (in two fragments), iambic trimeters from some dramatic genre. The
philosophical dialogue has been attributed to Aristotle or to Aristotelian circles. See the
discussion and bibliographic update by Lerner (2003). It is not easy to determine whether
the philosophical manuscript on papyrus was imported or comes from a local school (an
analogous problem is raised by the celebrated papyrus containing the Persians of Timoth­
eos).

(25.) One, a petition from 260, has received preliminary publication in Gascou (1999). The
second is a badly damaged marriage contract in two fragments.

(26.) An example of these fruitful, but long and costly, efforts is provided by the reattribu­
tion to Nahal Hever of a part of the texts assigned to Cave 4 at Qumran (P.Hever, pp. 283–
284, introd. to P.Hever 342–361).

Page 19 of 23
The Papyrology of the Near East

(27.) A codex of seven literary tablets of the third century supposed to come from Palmyra
(fables of Babrios; Hesiod) was published by Hesseling (1893) (Pack2 174, 491).

(28.) Youtie (1973), receipts of the sixth century connected to work on a kisterna (SB V
8073–8076).

(29.) Khirbet Mird, a toponym preserving its ancient name, Mar(d)os, is located in the Ju­
daean desert, about twenty kilometers south of Jericho and near the laura of Saint Sabas.
The Greek dossier has Arabic intermixed (Grohmann 1963), as well as Christian Palestin­
ian (Perrot 1963). The Greek part includes letters, biblical and liturgical fragments, and a
school text that span the period from the fifth—sixth century to the eighth—ninth and al­
so a manuscript from the fifth—sixth century. See van Haelst (1991). The first two texts,
letters, are reprinted as SB XX 14188–14189. The first letter seems to express hope for
the recipient's renewed health, reading at the start of line 2 ὑμετέρας ἰάσ[ε] ως instead of
[.]μετε̣ρα ιε̣[.]ω̣ς·

(30.) P.Daliyeh.

(31.) P.Jud.Des.Misc.,11–12.

(32.) These texts come from excavations carried out between 1922 and 1924 for the
Académie des Inscriptions et Belles-Lettres and continued from 1928 to 1937 in collabo­
ration with Yale University. See Cumont (1926) and P.Dura. Besides flexible materials,
several ostraca in Greek and four in Iranian were found (YClSt 14:169, 195–209), along
with a wax tablet and a cuneiform tablet of the second millennium BCE (Stephens 1937).

(33.) See the section titled Formats and Diplomatics.

(34.) These were republished by Fink (1971). The Latin texts from Dura were also reedit­
ed by R. Marichal, ChLA VI–IX. On the military unit see Pollard (1996) and Kennedy
(1994).

(35.) A good example of exegesis of the Dura texts, using all of the available documenta­
tion, is offered by Saliou (1992).

(36.) For a bibliography, see Bérard et al., eds. (2000), Guide de lʼépigraphiste, nos. 372–
374, 2025, 2185–2186.

(37.) Feissel and Gascou (1989).

(38.) P.Euphr. 1–5 (SB XXII 15496–15500); P.Euphr.6–10 (SB XXIV 16167–16171); P.Euphr.
11–17 (SB XXVI 16654–16660). The publication of the two Syriac pieces (P.Euphr. 19 and
20) is the work of Teixidor (1990, 1991); see the remarks of Brock (1991).

(39.) Millar (1993, 129–131,155–156, 478–481).

(40.) Gascou (1999, 61–73).

Page 20 of 23
The Papyrology of the Near East

(41.) For the institutions of the kingdom of Osrhoene and of Roman Mesopotamia, see
Gnoli (2000). For their legal systems, see Stolte (2001).

(42.) See generally Tov and Pfann (1993). The some 920–930 Hebrew and Aramaic texts
from Qumrân (current information kindly provided by M. Bélis) discovered in eleven
caves (especially that called 4Q), from before the first Jewish revolt of 66/73 (more pre­
cisely, April/June 68), are essentially religious and liturgical in content (see the summary
by Guglielmo 2003).

(43.) For further information see Schiffman and VanderKam (2000) and especially Tov
(2002).

(44.) P.Jud.Des.Misc. Although half of this catalogue is devoted to the finds from Ketef
Jericho (northwest of Jericho) (1–19), some of the Dead Sea sites are also represented:
Wadi Sdeir, Nahal Hever, Nahal Mishmar, and Nahal Seʼelim (Wadi Seiyal). The prove­
nance of one text is unknown. The finds from the caves of Jericho include early docu­
ments in Aramaic (fourth century BCE). The balance (Aramaic and Greek) can be as­
signed to the Flavians, Hadrian, and the revolt of Bar Kokhba. On no. 16 see Haensch
(2001). Furthermore, no. 18 seems to me to be the registration of a copy of a marriage
deed (γαμικῆς, 1.1, instead of ταμικῆς). Line 2 is the subscription of the agent who re­
ceived or registered the copy (perhaps the individual mentioned at the start of line 1,
where I would read ὁ δεῖνα χρε]οφύλαξ).

(45.) P.Murabbaʼat. The Greek pieces (nos. 89–157) are partly reprinted in SB X 10300–
10307. For a historical perspective see Cotton and Eck (2002).

(46.) For the documentation of this site, some ten kilometers south of Nahal Hever, see
P.Masada I; for the Graeco-Latin part, see P.Masada II. For a reedition of P.Masada II 741,
see SB XXIV 15988.

(47.) Cohen (2006). The editor places them in the same chronological range as the texts
of Nahal Hever. One element of support is provided by coins of Bar Kokhba. The first text
supposedly mentions arouras, which is not credible in this context. Cohen ( 2007) publish­
es two fragmentary Greek texts from the same area.

(48.) Tov (1990). This text comes from the “cave of horror.” For other biblical pieces from
Nahal Hever see P.Jud.Des.Misc., pp. 133–200.

(49.) Most of the Greek documents are in P.Yadin I; for the documents in languages other
than Greek, see P. Yadin II, a catalogue in which the texts are arranged in a fashion that
differs from what is announced in P.Yadin I and which, despite everything, includes some
important Greek pieces (nos. 52 and 59). P.Hever has some additional Greek pieces; see
especially pp. 131–279 by H. Cotton and the palaeographical study by J. D. Thomas, which
is devoted to Greek documents. This section includes an introduction to the archive of
Salome Komaïse, daughter of Levi.

Page 21 of 23
The Papyrology of the Near East

(50.) P.Yadin II 52 and 59. The first of these invokes the scribe's inability to write in “He­
brew.” For an analysis of the prince's dossier see Cotton (2003).

(51.) The tablet seems to be still unpublished. P.Murabbaʼat also contains pieces in He­
brew and Aramaic connected to the period of the “freedom of Israel,” including letters by
the prince himself (nos. 43 and 44).

(52.) See Katzoff and Schaps (2005).

(53.) See the excellent historical synthesis in P.Hever, pp. 133–165, and Cotton (1999).

(54.) P. Petra I and III. P.Petra I has generated substantial reviews, particularly Fournet
(2003) and Kruit (2004). The stimulating introduction to the Petra papyri by Koenen
(1996) is still worth reading.

(55.) It seems that at the end of the Byzantine period, the Palestinians, like the Egyptians,
preferred to settle disputes by arbitration rather than by lawsuits in the civil courts,
which consequently become rarer in our sources.

(56.) Daniel (2001).

(57.) See P.Nessana I (thirteen texts, including a glossary of the Aeneid and a fragment of
the poem); P.Nessana II.

(58.) Thus, P.Ness.II 77 v, which is of great interest for the status of non-Muslims in an Is­
lamic land.

(59.) See I.Ness. (particularly pp. 131–210, by G. E. Kirk, C. B. Welles, and F. Rosenthal).

(60.) Crisci (1996). See Otranto (1998).

(61.) In fact, J. Rea, in Rea, Senior, and Hollis (1994, 262), did not find anything in the re­
ceipt from Bactria that is not also found, for example, in Egypt; see also the remarks of
Crisci (1996) on the ostracon from Qala-i Sam, in Seistan (Iran/Afghanistan border), pp.
157–158, and on the manuscripts of Aï Khanoum, pp. 162–167.

(62.) Crisci (1996, 160–161) doubts the stylistic influence of Aramaic handwriting.

(63.) The stylistic separation from Egypt is, however, less marked in those countries in di­
rect contact with it (i.e., the Judaean and the Nabataean regions).

(64.) Pauli sententiae 5, 25, 6: ut exteriori scripturae fidem interior servet.

(65.) See P.Oxy. XLII 3054. 9–10; the same expression occurs for a sale at Tripolis in
Phoenicia (P.Oxy. XLII 3053.12) and for a “foreign” sale (P.Vind.Bosw.7.17).

(66.) This direction is that of the slave sale PSI Congr.XX 15 and is one of the reasons the
editor posited a non-Egyptian origin for this contract (see the bibliography given in the in­

Page 22 of 23
The Papyrology of the Near East

troduction). Considering its rather compressed formula, it could be a fragment of a scrip­

tura interior.

(67.) There is an astonishing survival of this custom in Bactria.

(68.) See P.Yadin I 13–16.

(69.) Some nuances are added by Gascou (1999), who notes that on the edges of Syria the
pre-Roman administrative territorial divisions were kept and municipalization was slow
and late.

(70.) See the clear summary article by Migliardi Zingale (1999); the study by Hengstl
(2002), despite its title, takes into account the totality of the juridical problems of the
Near Eastern papyri.

(71.) See Katzoff and Schaps (2005).

(72.) The documents drawn up in the brief reign of Bar Kokhba are in Hebrew and thus in
conformity with halachic law, in agreement with the political and ideological program of
this prince.

(73.) See Lewis (1970), Bowman (1976), and Geraci (1983).

(74.) With particularisms, as in the archives of Pathyris, which do not adopt the transver­
sa charta direction and present the various parts of the document in columns along the
fibers.

(75.) See D. Rathbone, P.Thomas 6, pp. 99–113.

(76.) In their commentaries, the editors of P.Euphr. refer several times to Egyptian papyri
of the sixth century.

(77.) Regnal years are no longer attested in Egypt after 384/385 until their reappearance
in the reign of Justinian.

Jean Gascou

Jean Gascou, Professor of Papyrology, University of Paris-Sorbonne.

Page 23 of 23
Writing Histories from the Papyri

Writing Histories from the Papyri

Todd M. Hickey
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0021

Abstract and Keywords

Papyrology has a long tradition of broader inquiry and historical synthesis that extends all
the way back to the days of Wilcken. Nor has the process itself of writing history from the
papyri altogether escaped inquiry and discussion, with Peter van Minnen and Roger Bag­
nall making the most important contributions in recent years. This article has been writ­
ten in conscious dialogue with their criticisms, prescriptions, and conclusions. Those who
work with the material remains of Graeco-Roman Egypt are in a unique position among
scholars of antiquity: the evidence is such that a much wider range of methods and ap­
proaches is available.

Keywords: Wilcken, papyrology, Roger Bagnall, Peter van Minnen, Graeco-Roman Egypt

This was neither more nor less than the queer extension of her experience, the
double life that, in the cage, she grew at last to lead. As the weeks went on there
she lived more and more into the world of whiffs and glimpses, she found her div­
inations work faster and stretch further It was a prodigious view as the pressure
heightened, a panorama fed with facts and figures, flushed with a torrent of colour
and accompanied with wondrous world-music What it mainly came to at this peri­
od was a picture of how London could amuse itself; and that, with the running
commentary of a witness so exclusively a witness, turned for the most part to a
hardening of the heart.

—Henry James, “In the Cage” (1898)

Because the papyrologist is an artisan working often with intractable material, be­
cause his texts and the inferences he draws from them are presented to the sub­
stantive disciplines as dependable facts, he cannot afford to remain unaware of
the basic assumptions that he uses He is necessarily concerned (p. 496) also with
such rules as have been devised for the detection of error. Unless he operates

Page 1 of 28
Writing Histories from the Papyri

within this framework, however flexible, of principles and rules, he can give no
guarantee of his competence as a maker of facts.

—Herbert Youtie, “The Papyrologist: Artificer of Fact” (1963)

There is no history, there are only historians.

—Lucien Febvre, as quoted in Paul Ricoeur, The Contribution of French Historiog­

raphy to the Theory of History (1980)

Many students’ first encounter with papyrological “method” comes through the writings of the
great editor Herbert Youtie, in the texts that “have established themselves as ‘the papyrologist's
Bible.’ ”1 Though none will deny that we can learn much from these works, some will surely be
repelled by the proposition that papyrology should serve as handmaiden for the “substantive dis­
ciplines.”2 Being an “artificer” of someone else's “facts” will have little appeal; much more in­
triguing—they will readily understand how James's telegraphist was drawn in—will be the
thought of “divining” narratives from the “whiffs and glimpses” of others’ lives.3
Academic backgrounds doubtless play a pivotal role in the reaction. In an influential arti­
cle published in 1988, Deborah Hobson probed her generation of papyrologists, noting:

Thus we decipher papyri not because we want to find out about the history of
Egypt but because we enjoy working with the Greek language; indeed our acade­
mic training for this discipline is primarily in Greek philology and palaeography,
not usually in historiography. In order to justify what we do as a legitimate part of
classical scholarship, we are trained to view Egyptian Greek documents as mani­
festations of Greek or Roman language and culture more than as evidence for life
in Egypt itself. (353)

But what if someone comes to papyrology from history, not philology? For certain kinds of an­
cient historians, the papyri have an obvious appeal; of all the sources available, they seem best
suited to answer the kinds of questions and tell the kinds of stories that interest them.4 The texts
themselves may well be secondary—stimulating, even “seductive” (thus Hobson 1988, 362),
means to an end more than ends in themselves.5
In fact, papyrology has a long tradition of broader inquiry and historical synthesis that ex­
tends all the way back to the days of Wilcken.6 Nor has the process itself of writing histo­
ry from the papyri altogether escaped inquiry and discussion, with Peter van Minnen
(1993) and Roger Bagnall (1995) making the most important contributions in recent
years.7 Though the present chapter cannot retrace the arguments in these works, it has
been written in conscious dialogue with their (p. 497) criticisms, prescriptions, and con­
clusions. Like Bagnall, I have drawn examples from the research (including my own) that
is best known to me; as a result, my coverage of the field (and its problems and possibili­
ties) is uneven. In light of the purpose of this volume—that is, lest this contribution be­
come dated too quickly— I have privileged contemporary work of quality. Younger schol­
ars also have an important place, for it is they who are doing some of the most interesting
and challenging research.

Page 2 of 28
Writing Histories from the Papyri

Looking into the Mirror, Opening Up the Closet

For the last decade of his life, Lucien Febvre's mantra was reported to be “History, sci­
ence of the past, science of the present.”8 History's best practitioners do not need to be
reminded that “the essence of the historical” is “a return of the past in the present
discourse” (Certeau 1986, 214); they write self-consciously, acutely aware of their own
“situatedness,” of their ideologies and limitations. Thus, we find Jacques Revel counsel­
ing, “[H]e [the historian] knows that his information represents a choice in reality, upon
which he superimposes other choices. He can at least attempt to measure the conse­
quences of them and to take advantage of them” (1989, xvi), while Gadamer (influential in
historical circles) advised, “The important thing is to be aware of one's own bias, so that
the text can present itself in all its otherness and thus assert its own truth against one's
own fore-meanings [Vorurteile]” (1989, 269). Self-reflection, however, has never been a
hallmark of papyrology (cf. Bagnall 1995, 1), and this aspect of its disciplinary culture, I
would argue, has been a significant impediment for intradisciplinary historical writing.
Manifestations of this lack of reflection are readily discernable: Despite being in the van­
guard of ancient studies in its application of technology, papyrology remains in some
senses a nineteenth-century discipline.9 The most unfortunate testimonium of this is the
field's long failure to address the illicit trade in papyri,10 which continues to this day—in
contravention of Egyptian antiquities law and the 1970 UNESCO Convention on the
Means of Prohibiting and Preventing the Illicit Import, Export, and Transfer of Ownership
of Cultural Property.11 Yet even when acquisitions have (or appear to have) fulilled the let­
ter of the law, there remain cases that leave one wondering how far removed we are from
Belzoni's declaration, so jarring to the twenty-first-century ear: “The purpose of my re­
searches was to rob the Egyptians of their papyri (1820, 157). Consider, for example, the
Milan Posidippus. The fact that “tomb robbers (or the like—note our rhetorical proximity
to “Indiana Jones”) were its ultimate source was widely (p. 498) reported in the press, yet
scholarly comment on the legal and ethical issues of the papyrus has been virtually
nonexistent (despite the appearance of well more than one hundred publications concern­
ing other aspects of the epigrams). I can only point to Dirk Obbink's forthright comments,
reported in Wilford (2002): “I don't think anyone knows the details of what happened, or
if they do, they're not talking … The Egyptians have laws against looting and might de­
mand the return of the papyrus.” At the very least, ambiguous acquisitions and kindred
acts (or inactions) imply the perpetuation of colonial sentiments, of paternalism and the
notion that the West is the true heir to the ancient Mediterranean world.12 This matters
because these sentiments can subvert the writing of even those with the best intentions.
Thus, Hobson, in an important critique of the “hellenocentricity” of papyrology, could
nonetheless still write the following (1988, 355–356):

When one gets out into the [Egyptian] countryside away from the major centres,
one feels that one has been transported back to the world of the papyri. The mud­
brick houses clustered together, topped by dovecotes, oxen pulling the water
wheel, Egyptians riding into the desert on camels, even toll stations set up along

Page 3 of 28
Writing Histories from the Papyri

the main road, all attest to a life for the peasant which must have remained re­
markably unchanged from the earliest times to the present. 13

In short, Phiroze Vasunia's prescription for mainstream classics is equally relevant for papyrolo­
gy: “What matters is … that they [students of antiquity] genuinely think through what it means
for them to profess and teach the past in the present historical moment, when the legacy of em­
pire continues to do its work openly or in secret” (2003, 95).
That said, how can one expect critical engagement with present practice if the past re­
mains unexamined? When it has been written, the history of papyrology itself has verged
on hagiography: These are the giants upon whose shoulders we stand.14 While the
achievements of pioneers like Grenfell and Hunt remain remarkable and certainly merit
acknowledgment, there seems ample room for the “rest of the story”: for criticisms of
method, to be sure, but also for the contextualization (or, if one prefers, deconstruction)
of scholarship,15 which should include readings of politics and even private lives.16 Some
may cry foul or call it gossip, but where does one draw the line between public and pri­
vate,17 and who is charged with holding the pen? Much more energy needs to be devoted
to the creation and dissemination of disciplinary archives, and the purging of “objection­
able” material from such assemblages (by those other than the principals) should be
viewed in the harshest light. “Archivization produces as much as it records the event”—
something of which those who work with archives of papyri should already be acutely
aware.18

For those “on the outside” who wish to use the papyri to write history, there are addition­
al impediments. Some of these (e.g., the need to learn technical skills to control the evi­
dence) are obvious,19 but a serious one, another nineteenth-century holdover, has re­
ceived inadequate consideration. This is the gold standard for (p. 499) presenting papyro­
logical texts, which has changed little since 1898, the year in which Grenfell and Hunt
published the first volume of The Oxyrhynchus Papyri.20 Although their “Oxyrhynchus
model” was clearly superior to the alternatives of the day,21 it does a great disservice to
the papyri by excessively alienating them from the infrastructures that they reference
and in which they functioned. Publication is a second excavation; like the archaeologist's
brush, it sweeps away ancient traces, and through it, papyri partake ever more of the
“circuit of activities characteristic of the present” and become ever more removed from
the “sequence of operations to which they belonged.”22

One of an editor's objectives should be to mitigate or remediate this process. Under the
“Oxyrhynchus model,” however, Greek literature and Christian texts receive pride of
place at the front of the volume,23 while the documents found with them (to say nothing
of other artifacts) are relegated to the back, sometimes without translations or even com­
plete transcriptions.24 This method of organization is specious, for a literary “work is situ­
ated in history in a way that gives it documentary dimensions, and the document has
worklike aspects.… [B]oth the ‘document’ and the ‘work’ are texts involving an interac­
tion between documentary and worklike components” (LaCapra 1983, 30–31; cf. also
1995, 805; Chartier 1982, 39–40). Non-Greek texts, moreover, almost never appear.25
Philology and “high” (i.e., Athens-Jerusalem) texts (in the traditional sense) are privileged

Page 4 of 28
Writing Histories from the Papyri

over the social and cultural matrix of the papyrus itself—an irony, of course, since papyri
are objects, often refuse26—and, in the interest of “objectivity” and “facts,” speculation—
the lifeblood of history—is avoided. The scholarly fashions of the day should not sully the
papyrus, for which the goal is nothing short of monumentalization (although its text may
be damaged, illegible, and subject to revision).27 The final product, although touted for its
ease of use (principally because translations are included), can be rough going for non­
specialists and students: “Context,” however one chooses to define it, is a first principle
of accessibility.

This is not to imply that editorial practice has been monolithic in the years since P.Oxy. I
appeared. For decades the Dutch and Flemish, for example, have tended to construct
their editions around archives, dossiers, and other kinds of clusters of related texts,28
Katelijn Vandorpe's edition of the “Dryton archive” (P.Dryton), with its combination of the
literary and the documentary, the Greek and the Egyptian, texts on various media (papyri,
ostraca, even graffiti), and a healthy dose of interpretation, is an excellent representative
of this alternative editorial paradigm.29 Momentum for the contextualized study of “texts”
and “objects” has also grown in the last decade or so,30 and in 2005 a call for an integrat­
ed approach to the publication of papyri and other archaeological material was made at a
panel during the annual meeting of Archaeological Institute of America.31

While such programs of scholarship certainly deserve encouragement, it is clear that

more can be done. What is especially striking is the simplicity of some of (p. 500) the pos­
sible improvements; they would not require papyrologists to make a significant invest­
ment in new approaches to their texts (though novel approaches could certainly play a
role). For example, accessibility—which papyrologists should prize out of self-interest if
for no other reason (cf. Bagnall 1995, 113)—could easily be enhanced by the inclusion of
broader synthetic studies within the text volume itself; these essays, which might be writ­
ten by the editor or outsiders (or both in collaboration), could try to answer fundamental
questions like “Why should a nonspecialist reader care about these texts?” or “Why are
these papyri important?”32 Moreover, while addressing the infrastructure of texts is desir­
able, it does not suffice to provide a list of artifacts without interpretation.33
Papyrologists’ general lack of interest in the visual also needs remedying.34 The intersec­
tions of papyri with the “Fayyum” portraits (and their mummies)—indeed, with funerary
art in general—with Ptolemaic royal sculpture, and with the hard-stone heads of priests
(to give just three examples) furnish opportunities too rich to leave unexploited.35

Three Modes: Integration, Scale, Voice

Integration

If the reassimilation of text and infrastructure has been a source of disappointment, the
integration of papyri into the history of the ancient Mediterranean world has proceeded
apace. Though rejections of papyrological (and other) evidence on the basis of Egypt's al­
terity or Sonderstellung continue to be made (and certain omissions of Egyptian material

Page 5 of 28
Writing Histories from the Papyri

continue to baffle; cf., e.g., Horden and Purcell 2000), these seem to be more a function of
“old habits dying hard” than anything else.36 In 1995, Bagnall could point to multiple
works that undermined these notions (see, e.g., pp. 61–68), and the list would be much
longer today. More important, the validity of Egyptian evidence seems to have become
less of an issue in this scholarship; in other words, papyrologists and ancient historians
are moving from a defensive (or an offensive) posture toward one of assumption of rele­
vance. In Cribiore's excellent survey (2001) of Greek education in the Hellenistic and Ro­
man worlds,37 for example, the validity of the Egyptian evidence is addressed in a few
sentences in the introduction to the volume; her presentation seems almost matter-of-fact
when compared to earlier treatments of the issue.38

A recent and rather ambitious attempt at integration is Constantin Zuckerman's Du vil­

lage à I ʼEmpire (2004).39 Here, as the title suggests, there is no question about (p. 501)
the relevance of the Aphroditê papyri with which Zuckerman is working for the broader
stream of history: “This Egyptian village [i.e., Aphroditê] proves to be an accurate sample
of imperial society, a microcosm in which problems that have been among the most debat­
ed of the economic and social history of the Byzantine Empire have left their imprint and,
perhaps, the key to their solution” (12).40 Zuckerman employs his texts, principally a fis­
cal register dating to 526/527, toposit additional Justinianic reforms in Egypt (beyond the
famous Edict XIII; see 52–56); to document the impact of the bubonic plague on Constan­
tinople (207–219); and to argue that fiscal practice in Aphrodite prefigures that in the
Middle Byzantine Empire (238–240). Though parts of the book fail to convince (e.g., the
discussion of the “minus carats” system [cf. 66 ff.] and, more important, the village social
structure proposed [221 ff.]), and while the author's explicit hostility toward models and
apparent reticence about experimenting with his data (cf. 234) cannot be encouraged
(see further below), his confident injection of the papyri into mainstream debates is re­
freshing, as is the combination of serious historical discussion with an edition of a text
critical for the author's arguments.

Written with similar assumptions about the relevance of the papyri (and comparable
bravado) and also concerned with the society and economy of late antiquity is Banaji
(2001).41 There the resemblance ends, however: Banaji's scope is much larger in that he
covers the whole of the Mediterranean (with an eastern emphasis) from the third to the
seventh century, and—far from being averse to models—his materialist paradigm radiates
from the pages. Banaji's ideology aside, his is one of the more important volumes on late
antiquity to have appeared in the last two decades; the breadth of learning on display is
remarkable, and the exposition provokes in a manner that recalls Peter Brown's æuvre.
The combination, moreover, of literary and documentary evidence, archaeology, numis­
matics, and comparative data is unparalleled elsewhere in the field. Banaji is surely cor­
rect about the fourth-century rise of a new elite that had its origins in the military and bu­
reaucracy and about the ever-increasing concentration of resources in the hands of these
grandees. Other arguments—for example, his assertion that wage labor expanded dra­
matically on the “great estates”— are, however, more problematic. There is an irony in
these cases: Despite the blinding array of references to primary sources, Banaji's work
might be described as he has portrayed Weber's: “antithetical to historical detail” (ibid.,
Page 6 of 28
Writing Histories from the Papyri

31). Not all detail, to be sure, but detail that does not suit the author's model of agricul­
tural capitalism. It is the model that has preeminence here.42

This does not mean that models should be avoided (as Zuckerman has done), only that
their heuristic or catalytic function cannot be forgotten; the cart must not come before
the horse. Models from the social sciences have been used with particular effectiveness
by the “Stanford School” of ancient history.43 In Manning (2003), for example, the tradi­
tional (“Polybius,” i.e., post-Raphia) model of Ptolemaic decline is overthrown by argu­
ments based on the sources—an approach, of course, that even the most committed posi­
tivist would find appealing. The novelties in (p. 502) Manning's work are the inclusion of
Egyptian sources (not only papyri but also incised material—the Edfu temple donation
text is particularly important) and, more critically, the substitution of an alternate, institu­
tions-based (“neoclassical”) model of the state.44 Manning acknowledges that his own
construction might itself be overturned (or need modification) with the advent of new
texts, but he believes “that such risks are important. For in order to understand any docu­
ment … one must have a conception of the historical context as well an idea of the struc­
ture of the state” (xi). Two points deserve emphasis here: One is that Manning's model,
like all good models, is flexible, even expendable; the other—even more important—is
that risk and experimentation should always trump the fear of making a mistake in histor­
ical writing. This can pose challenges for the papyrologist, who, in the editorial role, has
been trained to produce accurate texts (and probably to refrain from “excessive” specula­
tion).

Given the wealth of numerical data that the papyri contain, it is no surprise that papyrolo­
gists have long engaged in quantitative research. Generally speaking, however, these
analyses have lacked sophistication, while the data themselves have tended to escape
close scrutiny or (just as bad) conjecture—nothing ventured, nothing gained.45 The sim­
plest calculations can be fruitful if one is willing to experiment with the data; although
precise “answers” may not result, the benchmarks or parameters established are often
just as useful.46 Yet computer-aided modeling is where the possibility of real advances is
greatest. Anyone reading a book like Townsend (1993), an argument for the complemen­
tarity of historical data and the abstract modeling of (neoclassical) economics, cannot
help but think that such models might be profitably applied to data from the papyri. The
hitch is the mathematics, which are likely to be beyond the abilities of all but a few edi­
tors and ancient historians; collaboration would need to occur much more often than not.
Some younger scholars have demonstrated, however, that one person can wear both hats:
Ruffini (2004), for example, applies social network theory—and includes an admirable ex­
planation of it—to the communities of “pagan” scholars in Athens, Alexandria, and Upper
Egypt (or, more precisely, to the Vita Isidori of Damascius). This enables him to conclude
that the Alexandrians were linked more closely to the Athenians than to their counter­
parts elsewhere in Egypt, and to posit alternative explanations for the collapse of the
Alexandrian community in the 480s. More striking is the broad applicability plausibly pro­
posed for the method: “The amount of data is nearly limitless. We could use this approach

Page 7 of 28
Writing Histories from the Papyri

on Oxyrhynchite landholders, the civic [sic] elite of Aphrodito, or the desert fathers of Ni­
tria or Scetis”—to give but three examples.47

The promise of these quantitative methods for historical research is echoed by another
young practitioner, Katja Mueller, whose work has focused on the geography of the
Fayyum.48 A few caveats, however, seem in order. It is rather easy to be wowed by the
complicated modeling employed in these methods and to attribute greater accuracy to
them by virtue of this complexity (and because mathematics seems (p. 503) absolute),
when, in fact, their difficulty likely makes them more susceptible to error than research
conducted using traditional methods. Mueller, for example, dutifully provides her data,
but who has checked them (or will do so)? Since most of her publications have appeared
in unrefereed journals intended for papyrologists and ancient historians, this seems a
valid question.49 Then there is the matter of the continued viability of some of these meth­
ods. Mueller, again, uses the rank-size rule (Mueller 2002), but the discipline of geogra­
phy has largely abandoned such techniques for methods informed by theories that inte­
grate the spatial and the social (Lefèbvre, Foucault, Latour, et al.).50 This does not mean
that Mueller should not have created her model, only that its consumers should be aware
that her method has fallen out of favor in its primary discipline.

Yet a problematic engagement is better than no engagement at all (cf. the remarks of
Bynum 1991, 21, concerning American medievalists). Although this certainly holds true
with respect to quantitative modeling, it is also valid for methods from cultural studies, a
collection of disciplines for which Graeco-Roman Egypt—multicultural, quotidian, ruled
by foreigners, and so on—seems a perfect Wt. In Bagnall's words, “Within the world of
the ancient Mediterranean, no society offers the array of evidence for the workings of cul­
tural interaction in the lives of a wide spectrum of individuals that the Egypt of the papyri
does … Historians working with papyri thus have an opening to many of the liveliest ar­
eas of contemporary thought” (1995, 117). To be sure, others have also recognized this
promise,51 but I am struck by the compelling (and exciting) possibilities that remain. An
ongoing project of my own has benefited considerably from readings in postcolonial and
globalization theory, to such an extent that an extended description seems justiWed.52
Though similar approaches have been utilized for Ptolemaic Egypt,53 I would suggest that
the Roman occupation furnishes a more suitable application.

Excursus: The Priests of Soknebtynis

The vanquished always want to imitate the victor in his distinctive characteristics,
his dress, his occupation, and all his other conditions and customs.

—Ibn Khaldün, al-Muqaddima (trans. Franz Rosenthal)

Kronion and Isidora were deputy prophet and prophetess, respectively, in the Tebtynis temple
during the Wnal quarter of the second century CE.54 The eighty-odd texts in their dossier docu­
ment eight generations of their family (cf. figure 21.1) from the second half of the Wrst century
CE into the third.55 The dossier is scattered around the globe: Components of it ended up in
Berkeley, Cairo, Copenhagen, Florence, Heidelberg, Milan, Lund, New Haven, and Oxford.

Page 8 of 28
Writing Histories from the Papyri

“Thick” documentation of a group of individuals over such a lengthy period is unparalleled in the
(p. 504) corpus of Roman papyri,56 but more remarkable is the fact that the dossier includes lit­

erature in Greek, in addition to documentary texts, and that we also possess four hundred or so
Egyptian literary papyri that these priests were using.57
This bilingual corpus reveals an apparent paradox: Scholarship has tended to view village
priests as insular—a notion that is supported by the marriage practices attested in these
texts—but this particular family was manifestly engaged with “Greek” culture.58 I must
emphasize that this engagement, which occurred through reading and writing (and pre­
sumably conversation) was not a casual one.59 The handwriting of some family members
is quite accomplished, even professional,60 Moreover, while one might have anticipated
priests having an interest in the Greek astronomical and astrological texts, for example,
in the dossier, they clearly were also reading literature that would seem more at home in
an elite “Hellenic” context (like the Hesiodic Catalogue of Women).61

On a more general (i.e., village) level, this phenomenon has already been noted in Van
Minnen (1998), but the explanation he gives there (in an otherwise impressive piece of
scholarship) is unsatisfying. Van Minnen's account is both teleological (i.e., Greek culture
is seen as progressive) and without a satisfactory model of agency (culture is imagined as
simply acting upon the priests). In addition, the notion of exchange or negotiation is lack­
ing (the priests are assumed merely to be recipients); nor is there any acknowledgment
that the process would have been individualized and uneven (the presence of “microcul­
tures”).62 The papyri are well suited to provide the necessary nuance, but even unparal­
leled detail is of limited value without a theoretical framework.

Figure 21.1. The principal relatives of Kronion and

Isidora (working stemma).

“Strategies of power” and resistance, that is, “an experience that constructs and recon­
structs the identity of subjects” (Gupta and Ferguson 1997, 19), are compelling concepts
to place at the center of such a framework. The Roman conquest of Egypt in 30 bce was a
watershed event for Egyptian priests because they became subject to new strategies of
power; the Empire would organize Egypt's differences in ways that varied significantly
from those of the Ptolemies. Most notably, Roman rule led to the rupture of priestly and
royal codependency and to the upheaval of Ptolemaic social structure,63 With the excep­
Page 9 of 28
Writing Histories from the Papyri

tion of the citizens of a few “Greek” cities, the Empire's new subjects all became mere
“Egyptians.”64 Yet (as elsewhere) the realities of rule led the Romans to make certain dis­
tinctions within this subaltern mass. Well known is the Roman construction of the “Hel­
lenic” metropo-lite and gymnasial “classes.”65 Membership in both of these groups (the
latter was actually an elite subset of the former) had a hereditary basis (in theory, at
least), and the ruling power established an administrative apparatus or, rather, system of
examination, the eptkrisis, to control access to them.66 Their members received various
privileges (e.g., lower poll tax), while the gymnasial class, in conformity to the Roman be­
lief that the wealthy and educated were best suited to rule, had access to civic power. The
membership controls and privileges for both groups naturally (p. 505) (p. 506) contributed
to endogamy and, in one part of Egypt at least, to extraordinarily high rates of sibling
marriage (cf. Bagnall and Frier 1994, 130).

Less recognized is the fact that the Romans treated the priests in much the same way.
They clearly saw value in the Egyptian cults and viewed their personnel as an important
component of the social and administrative machinery.67 The Romans did not need to cre­
ate a hereditary priesthood, but they controlled access to the existing body through exam­
inations comparable to those used for the elites in the mêtropoleis (cf. Capponi 2005, 94).
Commentators have traditionally viewed these regulations as oppressive and ignored the
obvious preference being shown.68 The priests also enjoyed privileges comparable to or
better than those of the urban elites, and, as already noted, they too practiced endogamy.
This tendency predated the Romans, but Roman policy certainly did nothing to discour­
age the behavior.

Based on these kinds of similarities between the priests and the metropolite-gymnasial
class, we might anticipate contact between the two groups.69 Though much work remains
to be done on this question, links did exist between the family of Kronion and Isidora and
the descendants of Patron, a wealthy gymnasial clan that owned substantial property in
the meris of Polemon (the division of the Arsinoite nome in which Tebtynis was located).
Though the attested connections are what sociologists call “weak ties,” they are nonethe­
less significant; it is a widely accepted principle of network theory that links of this sort
provide access to social capital otherwise unavailable in one's own group.70 Such connec­
tions also suggest that priestly interaction with metropolite networks played a role—an
important one, possibly—in cultural exchange.

The papyri provide explicit testimony that cultural exchange occurred, but why it oc­
curred is another question. That Greek medical and scientific texts were valued by the
priests as “advances” is obviously unsatisfactory (i.e., teleological) and does not, in any
case, account for the other Greek bookrolls that they were reading. Looking to the prag­
matics of power, to culture as a signifier—to resistance—is helpful. “Practices that are re­
sistant to a particular strategy of power are … never innocent of or outside power, for
they are always capable of being tactically appropriated and redeployed within another
strategy of power” (Gupta and Ferguson 1997, 18–19). Thus, one may argue that the
same liminal position that enabled priests to represent legitimate (divinely sanctioned)
power to a subject population also made them susceptible to Roman inducements and de­

Page 10 of 28
Writing Histories from the Papyri

sirous of signs that would further distinguish them from the mass of ananthrôpoi
(“inhuman”) Egyptians.71

Yet one must take care to avoid oversimplifying what would have been a complex and
shifting situation. Emulation surely does not imply wholesale adoption. On the contrary,
adoption would have been selective, not to mention mutating and inverting,72 and the
priests’ traditional constituencies and power bases would not have been ignored. The in­
digenous (written) language of the priests remained remarkably resistant to Greek loan­
words (e.g., Clarysse 1987); John Ray speaks of demotic as “purified and filtered” (1994,
264). At the same time, the (p. 507) substantial Egyptian literary activity at Tebtynis has
tended to be associated with preserving the past.73 Both phenomena constitute resis­
tance, but perhaps not in the manner that one might think at first glance. Some Egyptolo­
gists view priestly literary activity as a continuation of (implicitly or explicitly “pure”)
pharaonic traditions (e.g., Hoffmann 2000, 196–197), but the imperial context requires us
to pay more and careful attention to the language and content of the Roman-period texts,
which are not identical to earlier versions.74 Then there is the case of the god Soknebty­
nis: Curiously, he was not syncretized with a Greek deity (Kronos) until after the Roman
conquest.75 Typically it is the conqueror who identifies his gods with those of the subject
(interpretatio; cf. Webster 1995), but on this occasion, indigenous tradition seems to have
“reached out” to authority.76 Or did it? To what extent is this an act of conciliation, since
it usurps the conqueror's traditional prerogative? Strikingly, the priests of Tebtynis over­
whelmingly “endorsed” the syncretization with a practice at the heart of identity: the
choice of their own names.77

Scale and Voice

[F]oxes or hedgehogs, butterflies or mole catchers, parachutists or truffle hunters …” So

goes Le Roy Ladurie's taxonomy of historians of literature (1998),78 but the same might
be said of those writing history from the papyri. Yet, despite the references to Braudel—
when papyrologists have engaged with historical theory, the point of engagement has of­
ten been the Annales school (cf. Bagnall 1995, 112 ff.)—or the patently “macro” research
of “parachutists” like Manning or Mueller,79 papyrology has greater affinities with micro­
historia; the evidence it self demands this.80 A good deal of the writing that has drawn on
the papyri is microhistory—of a sort; what is typically lacking is the theoretical frame­
work, the crucial possibility of engaging the larger scales. This seems a missed opportuni­
ty. Microhistory, when done properly, is hardly anecdotal.81 “Structures” may be put at
risk by individuals.82 “To the degree that a presupposition of submission by social agents
goes with a macrohistorical choice of scale, the microhistorical choice leads to the oppo­
site expectation, that of random strategies in which conflicts and negotiations take prece­
dence under the sign of uncertainty” (Ricoeur 2004, 219). This is one of the benefits of re­
ducing the scale of inquiry: Working “level with the ground” (au ras du sol) reveals “not
just familiar objects in miniature but different configurations of the social” (Revel 1995a,
46; cf. Levi 1991, 97).

Page 11 of 28
Writing Histories from the Papyri

Objections that microhistory is dated reveal an ignorance of the approach; as is the case
with the Annales, microhistory continues to be reformulated, and it readily accepts grafts
from other methods.83 Graeco-Roman Egypt is without doubt the theater of the ancient
Mediterranean world that is best suited for serious trials (and reinements) of its prac­
tices. At the very least, microhistory's concept of the “normal exception” (eccezione nor­
male) could help focus research energies on (p. 508) dissonant cultural fragments or, in
other words, on the texts (sensu lato) that “certain forces have attempted to melt down
into the anonymous mass of an unrecognizable culture” (the words of Derrida 1989, 821);
these present our best opportunities to challenge existing conceptions.84

Closely connected to microhistory is the use of narrative,85 which historians have adopted
for a number of reasons: to emphasize the subjectivity of the historical process, to high­
light the contradictions and heteroglossia of fragmentary sources (and representations of
these sources in the historical present), or because the contingent and discontinuous
facts of the past become intelligible only when woven together as stories” (Lowenthal
1985, 218). Once again, the evidence from Graeco-Roman Egypt seems eminently suited
for exploration,86 yet initiative has been lacking: Only the work of James Keenan stands
out for its narrative (and microhistorical) sensibilities.87 While a papyrological Alabi's
World is beyond our evidence—even the “multivocal” katochoi archive—and we lack
“thick” enough documentation to support a Stories of Scottsboro, a Death of Woman
Wang (with Aphroditê and environs in place of the Tʼan-chʼeng district, and the poet/no­
tary Dioscoros taking the roles of both the author Pʼu Sung-ling and the magistrate
Huang Liu-hung?) might be possible.88 Less controversially—Woman Wang was criticized
for its fictions”—an examination of storytelling in Graeco-Roman Egypt along the lines of
Fiction in the Archives could certainly be written; our petitions are not that far removed
from the lettres de remission that Natalie Zemon Davis employed in that volume (1987).89
This is not to suggest that papyrologists ape these outside paradigms but to encourage
them to extend the boundaries of their reading and to allow the fruits of this reading to
inform their work. This is one trait of Keenan's research, and it also characterizes Sebast­
ian Richter's recent study of the corpus of Coptic child-donation contracts. As one might
expect (given the unusual content of these texts), this is well-trodden ground, but
Richter's readings in cultural studies (in particular narratology) have led him to a novel
and compelling conclusion: that the children involved were troublesome cases (for med­
ical or other reasons) and that they were donated by overburdened parents who were in
“a complex dilemma of emotional, social, and religious components.” The stereotypical
narrationes in these texts, group autobiographies” owing to the many voices involved,
bore the therapeutic energy” of coherent narrative; they allowed those involved to make
sense out of nonsense. The interaction and communication between the monastery [re­
ceiving the children] and the issuers of child donation documents, resulting in the child
donation narratio, would have been a medium providing the parents themselves and their
community” with a means of coping with “disturbing, isolating experiences.”90

The objective of this short contribution has not been to prescribe but to encourage and il­
luminate possibilities. Those who work with the material remains of Graeco-Roman Egypt
are in a unique position among scholars of antiquity: Our evidence is such that a much
Page 12 of 28
Writing Histories from the Papyri

wider range of methods and approaches is available (p. 509) to us. By virtue of this, we
ought to be the pioneers in our departments—and innovation is certainly the best argu­
ment for our continued existence. Embracing our opportunities and having the courage to
experiment (and speculate) boldly need not (and should not) require any dilution of the
technical rigor that has characterized papyrology from its earliest days.91 At the same
time, we are obliged to begin working much more self-consciously—if philosophers are
not always papyrologists, it would be desirable, at least, that papyrologists be philoso­
phers92—and we must do so with the same scrupulousness that Youtie advised for editing.
It is only through looking carefully in the mirror (and opening up our closet) that growth
will occur, that our histories will merit the gifts that time, climate, and chance have be­
queathed us.

Acknowledgments
This chapter assumes some familiarity with the fundamental texts and schools of histori­
ography. For a good survey of these (written by a historian of late antiquity, no less), see
Clark (2004).

I thank James Keenan, Leslie Kurke, Joe Manning, Elisabeth O'Connell, and the partici­
pants of the Columbia Summer Seminar in Papyrology for their remarks on earlier ver­
sions of this chapter; the deficiencies that remain are my own. I am responsible for all un­
attributed translations. I offer this chapter to the memory of Jock Weintraub in gratitude
for those occasions when I was able to sit at his table.

Bibliography
Abu-Lughod, L. 1991. “Writing against Culture.” In Recapturing Anthropology: Working in
the Present, ed. R. G. Fox, 137–162. Santa Fe, N.M.: School of American Research Press.

Appadurai, A. 1996. Modernity at Large: Cultural Dimensions of Globalization. Minneapo­

lis: University of Minnesota Press.

Bagnall, R. S. 1992. “Landholding in Late Roman Egypt: The Distribution of Wealth.” JRS
82: 128–149.

———. 1995. Reading Papyri, Writing Ancient History. London: Routledge.

———, and B. W. Frier. 1994. The Demography of Roman Egypt. Cambridge: Cambridge
University Press.

Baines, J. 2004. “Egyptian Elite Self-Presentation in the Context of Ptolemaic Rule.” In An­
cient Alexandria between Egypt and Greece, ed. W. V. Harris and G. Ruffni, 33–61. Lei­
den: Brill.

Banaji, J. 2001. Agrarian Change in Late Antiquity: Gold, Labour, and Aristocratic Domi­
nance. Oxford: Oxford University Press.

Page 13 of 28
Writing Histories from the Papyri

Begg, D. J. I. 1998. “ ‘It Was Wonderful, Our Return in the Darkness with … the Baskets of
Papyri!’ Papyrus Finds at Tebtunis from the Bagnani Archives, 1931–1936.” BASP 35:
185–210.

Belzoni, G. B. 1820. Narrative of the Operations and Recent Discoveries within the Pyra­
mids, Temples, Tombs, and Excavations in Egypt and Nubia; and of a Journey to the Coast
of the Red Sea in Search of the Ancient Berenice, and of Another to the Oasis of Jupiter
Ammon. London: John Murray.

Bisson, T. N. 1998. Tormented Voices: Power, Crisis, and Humanity in Rural Catalonia,
1140–1200. Cambridge, Mass.: Harvard University Press.

Bowman, A. K. 2002. “Recolonising Egypt.” In Classics in Progress: Essays on Ancient

Greece and Rome, ed. T. P. Wiseman, 193–223. Oxford: Published for the British Academy
by Oxford University Press.

———, and D. W. Rathbone. 1992. “Cities and Administration in Roman Egypt.”

(p. 516)

JRS 82: 107–127.

Braudel, F. 1969. écrits sur lʼhistoire. Paris: Flammarion.

Bynum, C. W. 1991. Fragmentation and Redemption: Essays on Gender and the Human
Body in Medieval Religion. New York: Zone Books.

Canfora, L. 2005. Il papiro di Dongo. Milan: Adelphi.

Capponi, L. 2005. Augustan Egypt: The Creation of a Roman Province. London: Rout­
ledge.

Certeau, M. de. 1986. Heterologies: Discourse on the Other, trans. Brian Massumi. Min­
neapolis: University of Minnesota Press.

———. 1995. “History and Mysticism.” In Histories: French Constructions of the Past, ed.
J. Revel and L. Hunt; trans. A. Goldhammer et al., 437–447. New York: New Press.

Chartier, R. 1982. “Intellectual History or Sociocultural History? The French Trajecto­

ries.” In Modern European Intellectual History: Reappraisals and New Perspectives, ed.
D. LaCapra and S. L. Kaplan, 13–46. Ithaca, N.Y.: Cornell University Press.

Clark, E. 2004. History, Theory, Text: Historians and the Linguistic Turn. Cambridge,
Mass.: Harvard University Press.

Clarysse, W. 1987. “Greek Loan-Words in Demotic.” In Aspects of Demotic Lexicography:

Acts of the Second International Conference for Demotic Studies, ed. S. P. Vleeming, 9–
33. Leuven: Peeters.

———, and K. Vandorpe. 1997. “Viticulture and Wine Consumption in the Arsinoite
Nome.” Ancient Society 28: 67–73.

Page 14 of 28
Writing Histories from the Papyri

Colin, F. 2002. “Les prêtresses indigènes dans lʼégypte hellénistique et romaine: Une
question à la croisée des sources grecques et égyptiennes.” In Le rôle et le statut de la
femme en Égypte hellénistique, romaine, et byzantine: Actes du colloque international,
Bruxelles-Leuven, 27–29 novembre 1997, ed. H. Melaerts and L. Mooren, 41–122. Leu­
ven: Peeters.

———. 2003. “La documentation plurilingue de lʼégypte hellénistique: Un laboratoire des

interactions linguistiques et culturelles.” In lʼOrient méditerranéen de la mort
dʼAlexandre au Iersiècle avant notre ère: Anatolie, Chypre, Égypte, Syrie, ed. M.-T. Le Di­
nahet, 257–280. Nantes: Éditions du Temps.

Cribiore, R. 2001. Gymnastics of the Mind: Greek Education in Hellenistic and Roman
Egypt. Princeton: Princeton University Press.

Davis, N. Z. 1987. Fiction in the Archives: Pardon Tales and Their Tellers in Sixteenth-
Century France. Stanford: Stanford University Press.

Derrida, J. 1989. “Biodegradables: Seven Diary Fragments.” Trans. Peggy Kamuf. Critical
Inquiry 15(4): 812–873.

———. 1995. Mal dʼarchive: Une impression freudienne. Paris: Galilée.

Dieleman, J. 2005. Priests, Tongues, and Rites: The London-Leiden Magical Manuscripts
and Translation in Egyptian Ritual (100–300 CE). Leiden: Brill.

Fabre, G. 2002–2003. “Medea Norsa ebrea?” Analecta Papyrologica 14–15: 337–350.

Feierman, S. 1999. “Colonizers, Scholars, and the Creation of Invisible Histories.” In Be­
yond the Cultural Turn: New Directions in the Study of Society and Culture, ed. V. E. Bon­
nell and L. Hunt, 182–216. Berkeley: University of California Press.

Fournet, J.-L. 2004. “Entre document et littérature: La pétition dans lʼAntiquité tardive.”
In La pétition à Byzance, ed. D. Feissel et J. Gascou, 61–74. Paris: Association des amis du
Centre dʼhistoire et civilisation de Byzance.

Frankfurter, D. 1998. Religion in Roman Egypt: Assimilation and Resistance. Princeton:

Princeton University Press.

(p. 517) Franko, G. F. 1988. “Sitometria in the Zenon Archive: Identifying Zenon's Personal
Documents.” BASP 25: 13–98.

Gadamer, H.-G. 1987. “The Problem of Historical Consciousness,” trans. J. L. Close. In In­
terpretative Social Science: A Second Look, ed. P. Rabinow and W. Sullivan, 82–140.
Berkeley: University of California Press.

———. 1989. Truth and Method, 2d rev. ed., trans. J. Weinsheimer and D. G. Marshall.
New York: Crossroad.

Page 15 of 28
Writing Histories from the Papyri

Gagos, T., J. E. Gates, and A. T. Wilburn. 2005. “Material Culture and Texts of Graeco-Ro­
man Egypt: Creating Context, Debating Meaning.” BASP 42: 171–188.

Gibbon, E. 1814. The Miscellaneous Works of Edward Gibbon, Esq., with Memoirs of His
Life and Writings, vol. 4. London: John Murray.

Gigante, M. 1986. Girolamo Vitelli e la nuova filologia. Santa Croce del Sannio: Istituto
storico Giuseppe Maria Galanti.

Ginzburg, C. 1988. “Clues: Roots of an Evidential Paradigm.” In Clues, Myths, and the
Historical Method, trans. J. and A. Tedeschi, 96–125. Baltimore: Johns Hopkins University
Press.

Glare (now Fewster), P. 1993. “The Temples of Egypt: The Impact of Rome.” Ph.D. diss.,
Cambridge University.

Goodman, J. 1994. Stories of Scottsboro. New York: Pantheon.

Grafton, A. 2006. “History's Postmodern Fates.” Daedalus 135(2): 54–69.

Granovetter, M. S. 1973. “The Strength of Weak Ties.” American Journal of Sociology 78:
1360–1380.

Grantham, G. 2004. Review of Banaji 2001. http://www.eh.net/bookreviews/library/

0784.shtml (accessed September 22, 2008).

Grendi, E. 1972. “Microanalisi e storia sociale.” Quaderni Storici 7: 506–520.

Gupta, A., and J. Ferguson. 1997. “Culture, Power, Place: Ethnography at the End of an
Era.” In Culture, Power, Place: Explorations in Critical Anthropology, ed. A. Gupta and J.
Ferguson, 1–29. Durham, N.C.: Duke University Press.

Hanson, A. E. 2001. “Text and Context for the Illustrated Herbal from Tebtunis.” Atti del
XXII Congresso Internazionale di Papirologia, ed. I. Andorlini, G. Bastianini, M. Manfredi,
and G. Menci, 585–604. Florence: Istituto Papirologico G. Vitelli.

Herbin, F.-R. 2002. Padiimenipet, fils de Sôter: Histoire dʼune famille dans lʼégypte ro­
maine. Paris: Réunion des musées nationaux/Louvre.

Hickey, T. M. 2001. “A ‘Public House’ but Closed: ‘Fiscal Participation’ and Economic De­
cision Making on the Oxyrhynchite Estate of the Flavii Apiones.” Ph.D. diss., University of
Chicago.

Hobson, D. 1988. “Towards a Broader Context for the Study of Greco-Roman Egypt.”
échos du monde classique 32, n.s., 7: 353–363.

Hochschild, A. 1998. King Leopoldʼs Ghost: A Story of Greed, Terror, and Heroism in Colo­
nial Africa. New York: Houghton Mifflin.

Page 16 of 28
Writing Histories from the Papyri

Hoffmann, F. 2000. Ägypten: Kultur und Lebenswelt in griechisch-römischer Zeit: Eine

Darstellung nach den demotischen Quellen. Berlin: Akademie Verlag.

Holm, C. E. 1936. Griechisch-ägyptische Namenstudien. Göteborg: Wettergren & Ker­

bergs.

Hombert, M. 1933. “Le commerce des papyrus en Égypte.” Chronique dʼÉgypte 15: 148–
154.

Horden, P., and N. Purcell. 2000. The Corrupting Sea: A Study of Mediterranean History.
Oxford: Blackwell.

Jones, A. H. M. 1964. The Later Roman Empire, 284–602: A Social, Economic, and
(p. 518)

Administrative Survey. Norman: University of Oklahoma Press.

Keenan, J. G. 1992. “A Constantinople Loan, A.D. 541.” BASP 29: 175–182.

———. 1993. “Papyrology and Byzantine Historiography.” BASP 30: 137–144.

———. 1995. “The Aphrodito Murder Mystery: A Return to the Scene of the Crimes.”
BASP 32: 57–63.

Kulikowski, M. 2005. Review of C. Kelly, Ruling the Later Roman Empire. http://
ccat.sas.upenn.edu/bmcr/2005/2005-02-12.html (accessed September 25, 2008).

LaCapra, D. 1983. Rethinking Intellectual History: Texts, Contexts, Language. Ithaca,

N.Y.: Cornell University Press.

———. 1995. “History, Language, and Reading: Waiting for Crillon.” American Historical
Review 100(3): 799–828.

Le Roy Ladurie, E. 1998. “Michelet: Le poète de lʼhistoire.” Le Figaro littéraire (June 25).

Levi, G. 1991. “On Microhistory.” In New Perspectives on Historical Writing, ed. P. Burke,
93–113. University Park, Penn.: Pennsylvania State University Press.

Lowenthal, D. 1985. The Past Is a Foreign Country. Cambridge: Cambridge University

Press.

Manning, J. G. 2003. Land and Power in Ptolemaic Egypt: The Structure of Land Tenure.
Cambridge: Cambridge University Press.

Martin, A. 1994. “Archives prunes et cachettes documentaires.” Proceedings of the 20th

International Congress of Papyrologists, ed. A. Bülow-Jacobsen, 569–577. Copenhagen:
Museum Tusculanum Press.

Merton, R. K. 1965. On the Shoulders of Giants: A Shandean Postscript. New York: Free
Press.

Page 17 of 28
Writing Histories from the Papyri

Moioli, M. L. 1987. “La famiglia di Sarapion alias Apollonianus, stratego dei nomi Arsi­
noites ed Hermopolites.” Acme 40: 123–146.

Montserrat, D. 1996. Sex and Society in Graeco-Roman Egypt. London: Kegan Paul Inter­
national.

Morelli, F. 2002. “La raccolta dei P.Bingen.” Chronique dʼÉgypte 77: 312–321.

Mueller, K. 2002. “Ptolemaic Settlements in Space: Settlement Size and Hierarchy in the
Fayum.” Archiv für Papyrusforschung 48: 107–125.

———. 2003. “Mastering Matrices and Clusters: Locating Graeco-Roman Settlements in

the meris of Herakleides (Fayum/Egypt) by Monte-Carlo-Simulation.” Archiv für Papyrus­
forschung 49: 218–254.

———. 2005a. “Geographical Information Systems (GIS) in Papyrology: Mapping Frag­

mentation and Migration Flow to Hellenistic Egypt.” BASP 42: 63–92.

———. 2005b. “Redisricting the Ptolemaic Fayum, Egypt: From Nomarchies and
Toparchies to Weighted Voronoi Tessellation.” Archiv für Papyrusforschung 51: 112–126.

———, and W. Lee. 2005. “From Mess to Matrix and Beyond: Estimating the Size of Set­
tlements in the Ptolemaic Fayum/Egypt.” Journal of Archaeological Science 32: 59–67.

OʼConnell, E. 2007. “Recontextualizing Berkeley's Tebtunis Papyri.” In Proceedings of the

24th International Congress of Papyrologists, ed. J. Frösén, T. Purola, and E. Salmenkivi,
807–826. Helsinki: Societas scientiarum Fennica.

Price, R. 1990. Alabi's World. Baltimore: Johns Hopkins University Press.

Rathbone, D. W. 1989. “The Ancient Economy and Graeco-Roman Egypt.” In Egitto e sto­
ria antica dallʼellenismo allʼetà araba: Bilancio di un confronto. Atti del colloquio inter­
nazionale, ed. L. Criscuolo and G. Geraci, 159–176. Bologna: Cooperativa Libraria Univer­
sitaria Editrice Bologna.

———. 1990. “Villages, Land, and Population in Graeco-Roman Egypt.” Proceed­

(p. 519)

ings of the Cambridge Philological Society 216, n.s., 36: 103–142.

Ray, J. D. 1994. “How Demotic Is Demotic?” Egitto e Vicino Oriente 17: 251–264.

Reid, D. M. 2002. Whose Pharaohs? Archaeology, Museums, and Egyptian National Identi­
ty from Napoleon to World War I. Berkeley: University of California Press.

Revel, J. 1989. “lʼhistoire au ras du sol.” In Le pouvoir au village: Histoire dʼun exorciste
dans le Piémont du XVIIe siècle, G. Levi; trans. M. Aymard, i–xxxiii. Paris: Gallimard NRF.

———. 1995a. “Introduction.” In Histories: French Constructions of the Past, ed. J. Revel
and L. Hunt; trans. A. Goldhammer et al., 1–63. New York: New Press.

Page 18 of 28
Writing Histories from the Papyri

———. 1995b. “Microanalysis and the Construction of the Social.” In Histories: French
Constructions of the Past, ed. J. Revel and L. Hunt; trans. A. Goldhammer et al., 492–502.
New York: New Press.

Richter, T. S. 2005. “What's in a Story? Cultural Narratology and Coptic Child Donation
Documents.” Journal of Juristic Papyrology 35: 237–264.

Ricoeur, P. 2004. Memory, History, Forgetting, trans. K. Blamey and D. Pellauer. Chicago:
University of Chicago Press.

Riggs, C. 2005. The Beautiful Burial in Roman Egypt: Art, Identity, and Funerary Religion.
Oxford: Oxford University Press.

Ruffini, G. 2004. “Late Antique Pagan Networks from Athens to the Thebaid.” In Ancient
Alexandria between Egypt and Greece, ed. W. V. Harris and G. Ruffini, 241–257. Leiden:
Brill.

———. 2005. “Social Networks in Byzantine Egypt.” Ph.D. diss., Columbia University.

———. 2006. “Genealogy and the Gymnasium.” BASP 43: 71–99.

Ruggiero, G. 1993. Binding Passions: Tales ofMagic, Marriage, and Power at the End of
the Renaissance. Oxford: Oxford University Press.

Sahlins, M. D. 1985. Islands of History. Chicago: University of Chicago Press.

Scheidel, W. 2001. “Progress and Problems in Roman Demography.” In Debating Roman

Demography, ed. W. Scheidel, 1–81. Leiden: Brill.

Schubert, P. 2000. Vivre en Égypte gréco-romaine: Une sélection de papyrus. Vevey,

Switzerland: éditions de lʼAire.

Sewell, W. H., Jr. 1999. “The Concept(s) of Culture.” In Beyond the Cultural Turn: New Di­
rections in the Study of Society and Culture, ed. V. E. Bonnell and L. Hunt, 35–61. Berke­
ley: University of California Press.

———. 2005. Logics of History: Social Theory and Social Transformation. Chicago: Uni­
versity of Chicago Press.

Spence, J. D. 1978. The Death of Woman Wang. New York: Viking.

Stanwick, P. E. 2002. Portraits of the Ptolemies: Greek Kings as Egyptian Pharaohs.

Austin: University of Texas Press.

Stephens, S. A. 2003. Seeing Double: Intercultural Poetics in Ptolemaic Alexandria.

Berkeley: University of California Press.

Stone, L. 1979. “The Revival of Narrative: Reflections on a New Old History.” Past and
Present 85: 3–24.

Page 19 of 28
Writing Histories from the Papyri

Tait, W. J. 1992. “Demotic Literature and Egyptian Society.” In Life in a Multi-Cultural So­
ciety: Egypt from Cambyses to Constantine and Beyond, ed. J. H. Johnson, 303–310.
Chicago: Oriental Institute of the University of Chicago.

———. 1994. “Some Notes on Demotic Scribal Training in the Roman Period.” Proceed­
ings of the 20th International Congress of Papyrologists, ed. A. Bülow-Jacobsen, 188–192.
Copenhagen: Museum Tusculanum Press.

Thompson, D. J. 1988. Memphis under the Ptolemies. Princeton: Princeton Univer­

(p. 520)

sity Press.

Townsend, R. M. 1993. The Medieval Village Economy: A Study of the Pareto Mapping in
General Equilibrium Models. Princeton: Princeton University Press‥

van Minnen, P. 1993. “The Century of Papyrology (1892–1992).” BASP 30: 5–18. http://
scriptorium.lib.duke.edu/papyrus/texts/history.html (September 21, 2008).

———. 1994. “House-to-House Enquiries: An Interdisciplinary Approach to Roman Kara­

nis.” ZPE 100: 227–251.

———. 1998. “Boorish or Bookish? Literature in Egyptian Villages in the Fayum in the
Graeco-Roman Period.” Journal of Juristic Papyrology 28: 99–184.

———. 2002. “αἱ ἀπὸ γυμνασίου: ‘Greek’ Women and the Greek ‘Elite’ in the Metropoleis
of Roman Egypt.” In Le rôle et le statut de la femme en Égypte hellénistique, romaine et
byzantine: Actes du colloque international, Bruxelles-Leuven, 27–29 novembre 1997, ed.
H. Melaerts and L. Mooren, 337–353. Leuven: Peeters.

Vasunia, P. 2003. “Hellenism and Empire: Reading Edward Said.” Parallax 9: 88–97.

Verhoogt, A. M. F. W. 1998. “Family Papers from Tebtunis: Unfolding a Bundle of Papyri.”

In The Two Faces of Graeco-Roman Egypt: Greek and Demotic and Greek-Demotic Texts
and Studies Presented to P. W. Pestman by Alumni of the Papyrological Institute, ed. A. M.
F. W. Verhoogt and S. P. Vleeming, 141–154. Leiden: Brill.

Walker, S. E. C. 1997. “Mummy Portraits in their Roman Context.” In Portraits and Masks:
Burial Customs in Roman Egypt, ed. M. L. Bierbrier, 1–6. London: British Museum Press.

Wallerstein, I. 1995 “Annales as Resistance.” In Histories: French Constructions of the

Past, ed. J. Revel and L. Hunt; trans. A. Goldhammer et al., 367–370. New York: New
Press.

Webster, J. 1995. “ ‘Interpretatio’: Roman Word Power and the Celtic Gods.” Britannia 26:
153–161.

Wilfong, T. G. 2007. “Gender and Society in Byzantine Egypt.” In Egypt in the Byzantine
World, 300–700, ed. R. S. Bagnall, 309–327. Cambridge: Cambridge University Press.

Page 20 of 28
Writing Histories from the Papyri

Wilford, J. N. 2002. “A Mummy's Bequest: Poems from a Master.” New York Times
(November 26).

Will, E. 1985. “Pour une ‘anthropologie coloniale’ du monde hellénistique.” In The Craft of
the Ancient Historian: Essays in Honor of Chester G. Starr, ed. J. W. Eadie and J. Ober,
273–301. Lanham, Md.: University Press of America.

Youtie, H. 1963. “The Papyrologist: Artificer of Fact.” Greek, Roman and Byzantine Stud­
ies 4: 19–32.

———. 1974. The Textual Criticism of Documentary Papyri, 2d rev. ed. London: Institute of
Classical Studies.

Zuckerman, C. 2004. Du village à lʼEmpire: Autour du registre fiscal dʼAphroditô

(525/526). Paris: Association des amis du Centre dʼhistoire et civilisation de Byzance.

Notes:

(1.) Keenan, this volume, chapter 3, on Youtie (1963, 1974).

(2.) Cf. the criticism in Van Minnen (1993,12): “Papyrology is not an ancillary discipline, a
view often erroneously held even by papyrologists. Papyrology is a primary discipline.’’

(3.) I owe the “In the Cage” metaphor (but not its application to papyrology) to Levi
(1991, 106). Admittedly, the response of the papyrologist–historian to these narratives is
more likely to be one of wonder, not a “hardened heart.”

(4.) Cf. Bagnall ( 1995, vii): “Because the papyri are by far our best ancient source for
many aspects of these areas of history [i.e., social and economic history].”

(5.) Cf. Bynum (1991, 15): “Behind whatever smooth surface they [i.e., her essays] may
present to the modern reader lies the hard grappling with the texts—reading them, dat­
ing them, ascribing them—that is the ordinary work of the medievalist” (emphasis added).

(6.) See Keenan, this volume, chapter 3.

(7.) I would be remiss if I did not also mention Bowman (2002), a thoughtful essay.

(8.) Reported in Braudel (1969, 61); cf. also Wallerstein (1995). For this issue, see, for ex­
ample, Gadamer (1987).

(9.) For the heyday of papyrological discovery, the Victorian Age, see Keenan, this vol­
ume, chapter 3.

(10.) Cf. already the remarks of Morelli (2002, 313). (I thank Fabian Reiter for bringing
this review to my attention.)

Page 21 of 28
Writing Histories from the Papyri

(11.) The UNESCO Convention is available at http://portal.unesco.org/en/ev.php-

URL_ID=13039&URL_DO=DO_TOPIC&URL_SECTION=201.html (accessed September
29, 2008). For an English translation of Egyptian law 117 of 1983 see http://
unesdoc.unesco.org/images/0006/000666/066629eo.pdf(accessed September 29, 1983).
Since this chapter was written there have been welcome developments. In June 2007, the
American Society of Papyrologists (ASP) overwhelmingly approved a resolution concern­
ing the illicit trade in papyri. Two months later the leadership of the Association Interna­
tionale de Papyrologues (AIP) took up the issue at the International Congress of Papyrolo­
gy in Ann Arbor; they sponsored an educational plenary session and created an ad-hoc
committee to study the commerce further.

(12.) Cf. Reid (2002) on Western domination and control of Egypt's antiquities during the
“long nineteenth century.” The sequel to this volume will cover the years up to Nasser.

(13.) Contrast, for example, Abu-Lughod (1991, 154) (“homogeneity, coherence, and time­
lessness”).

(14.) “Shoulders of giants” is a reference to R. K. Merton's celebrated work (1965) con­

cerning “Newton's” aphorism. Merton's argument that a unified consensus (or meta-nar­
rative) is being built through scholarship is (unconsciously) central to current papyrologi­
cal method.

(15.) “Criticism of method”: Van Minnen (1993) gives us a good model. “Contextualization
of scholarship”: Hombert (1933) (discussed by Keenan in chapter three of this volume)
makes for ironic reading if one has any familiarity with the history of the Belgian Congo;
one wonders whether parts of it could have been written without the “benefit” of the First
World War; cf. Hochschild (1998, 292–306).

(16.) The Italian interest in the history of the discipline is admirably exceptional, and in
some cases, “difficult” issues have been addressed; see, for example, Fabre (2002–2003)
(on Medea Norsa's “racial” identity under Mussolini) and now Canfora (2005), kindly
brought to my attention by my colleagues at the Istituto Papirologico “G. Vitelli.” Cf. also
Gigante (1986) (a reference I owe to James Keenan).

(17.) Papyrologists are confronted with the same problem in their ancient texts; see, for
example, Franko (1988).

(18.) The quote is from Derrida (1995, 34) (speaking of the transformations of “technical
structures”—fax machines, e-mail, etc.—but equally relevant to human agency).

(19.) For such “formidable entry fees,” cf. Bagnall (1995, 110–111).

(20.) What follows should in no way be considered an indictment of the quality of the
scholarship in the series The Oxyrhynchus Papyri.

(21.) Cf. Keenan (1993, 142–143) and chapter three of this volume (more critically).

Page 22 of 28
Writing Histories from the Papyri

(22.) I borrow the words of Certeau (1995, 444). For comparable observations, see Gagos,
Gates, and Wilburn (2005); I thank the authors for sharing a draft of their publication.
Note as well Van Minnen (1993, 12), observing Von Wilamowitz-MoellendorV's ability to
see the “big picture” despite the volumes’ organization.

(23.) Cf. Van Minnen (1993,11) on literary papyrology: “New literary papyri are as a rule
studied in splendid isolation from the culture that produced them.’’

(24.) The typology employed in a recent volume: “New Literary Texts (a. Tragedy and
Comedy; b. Prose),” “Known Literary Texts,” “Subliterary Texts (a. Scribal Practice and
Draft; b. Magic),” “Documentary Texts.” On occasion, the ancients’ own attitudes seem to
stand in stark contrast; cf. P.Mich.Vl 390.

(25.) For the non-Greek material from Oxyrhynchus, cf. Bowman (2002, 218).

(26.) I do not mean to suggest that Grenfell and Hunt completely ignored material con­
texts (even in volumes besides P.Fay.). For the exploitation of the information that they do
provide, see, for example, Verhoogt (1998). OʼConnell (2007) shows how much can be
done with the limited records that have been preserved.

(27.) In the preface to P. Oxy. L, the general editors write of their late colleague Edgar Lo­
bel: “At the same time he inherited and continued a particular style: no parade of scholar­
ship, no clutter of bibliography; no shrinking from ‘I,’ when the pronoun properly repre­
sented the subjective limitations of eyesight and intuition; an insistence on fact and preci­
sion, a distaste for easy solutions and grandiose speculations. This is a tradition of which
we are proud; we hope that our next fifty volumes will maintain it.’’ “Subject to revision”:
There are currently eleven volumes of corrections to documentary texts alone!

(28.) For the distinction between archives and dossiers, cf. Martin (1994).

(29.) Although not every text is translated, and some pieces (through no fault of Vandor­
pe) appear as descripta, I have used this volume successfully with freshmen (i.e., it is ac­
cessible).

(30.) In papyrological circles, Van Minnen (1994) led the way, relying on data from the
reasonably well-documented Michigan excavations at Karanis. In the wake of more recent
“scientiWc” excavations, the possibilities have expanded. For the opportunities afforded
by the excavations that occurred before World War I, cf. note 26 above.

(31.) Gagos, Gates, and Wilburn (2005) derives from this panel.

(32.) Such essays will be included in three of the new volumes of The Tebtunis Papyri (VII-
IX). The practice, however, has precedents; cf. Rea on the Oxyrhynchite “corn
dole” (sitêresion) in P.Oxy. XL and even (as James Keenan reminds me) Rostovtzeff's com­
mentary on P.Tebt. III 703. Cf. also Zuckerman (2004) (though I would have preferred a
text that was better annotated).

Page 23 of 28
Writing Histories from the Papyri

(33.) As I did in P.Zauzich 13–14. Yet not all houses “work” as well as that of Sokrates at
Karanis (cf. Van Minnen 1994 and note 84 below).

(84.) Eccezione normale: Grendi (1972), though it is not entirely clear what he meant; cf.
Revel (1989, xxx–xxxi). I concur with Levi (1991, 109); for another interpretation see Ri­
coeur (2004, 216). Again, these conceptions may be methodological (e.g.: I cannot prove
it, but the house of Sokrates at Karanis [see Van Minnen 1994] strikes me as an eccezione
normale; i.e., I wonder whether the finds from other Karanis houses would have impacted
the way that papyrologists view archaeological material).

(34.) In his sensitivity to the visual, Montserrat (cf. 1996) was unique among those who
have edited papyri.

(35.) “Fayyum portraits”: The connection to the gymnasial “class” (cf., e.g., Walker 1997)
has not been adequately explored. “Funerary art in general”: Riggs (2005) should reveal
opportunities for those working with texts; the Soter tomb in Western Thebes (see most
recently Herbin 2002 and Riggs 2005, 182 V.) in particular cries out for an integrated ap­
proach, though not all of the papyri from the tomb have been published. Stanwick (2002)
provides a Wne synthetic (“text”-”object”) treatment of Ptolemaic royal sculpture but is
not the last word. Cf. also Baines (2004, esp. p. 35), excellent on elite self-presentation
during the Ptolemaic period. One might also note the central position of the visual in the
Egyptian religio-cultural system. (I thank Chris Hallett for bringing the works by Stan­
wick and Baines to my attention.)

(36.) Sonderstellung: The notion goes back to Tac. Hist. 1.11. “Old habits”: There were al­
ways exceptions, of which Jones (1964) is a prominent one.

(37.) Her subtitle, Greek Education in Hellenistic and Roman Egypt, belies the real
breadth of the book.

(38.) Contrast Rathbone (1989), a classic assault on the historiography of the ancient
economy.

(39.) My understanding of this work has been enhanced by discussions with Roger Bag­
nall and James Keenan.

(40.) The Sonderstellung Ägyptens has always been less of a factor in the historiography
of late antiquity, yet it certainly is present: “One is nevertheless hesitant to extend the
generalization [about landholding in late antique Egypt] to other parts of the Empire.
Egypt was always a special case” (Grantham 2004).

(41.) Zuckerman's failure to engage Banaji in any significant fashion is thus puzzling. Ku­
likowski (2005) speaks of a “Banaji eVect”: Banaji's work is ignored because his “sociolo­
gy and his economics are monuments of opacity, painfully inaccessible to non-specialists.”
I do not share this opinion.

(42.) See further Hickey (2001, 106–107).

Page 24 of 28
Writing Histories from the Papyri

(43.) “Stanford School”: The term is my coinage. Not all of the members have published
on Graeco-Roman Egypt (e.g., Ian Morris), nor are they all at Stanford (e.g., Dennis Ke­
hoe). Nor are they a monolith: Scheidel (cf. 2001), for example, is more pessismistic about
ancient sources than Manning (cf. 2003).

(44.) Drawn from the work of economist Douglass C. North; see Manning (2003, 10–11)
for elaboration and references.

(45.) Cf. Bagnall (1995, 73–75). He provides, incidentally, one of the exceptions to the rule
in, for example, Bagnall (1992).

(46.) Cf. Hickey (2001, 78 ff). For the “power of estimates,” see also Clarysse and Vandor­
pe (1997) and the seminal Rathbone (1990).

(47.) Ruffini (2004, 245). Ruffini himself addresses the Oxyrhynchite landholders and the
village elite of Aphrodite in his doctoral dissertation (Ruffini 2005).

(48.) See, for example, Mueller (2003, 250; 2005b, 121) for statements concerning the ap­
plicability of her methods.

(49.) Mueller (2005a) was certainly refereed. Mueller and Lee (2005) may have been ref­
ereed since the Journal of Archaeological Science asks contributors to name Wve possible
referees. (After this chapter was completed, two informaticians published a critique of
Mueller's methods: P. Hoffman and B. Klin, “Careful with That Computer,” Journal of Juris­
tic Papyrology 36 [2006] 67–90.)

(50.) For criticism of the rank-size rule and central place theory in a work of ancient his­
tory, cf. Horden and Purcell (2000, 102–105), who claim, curiously, that these methods
are “of increasing interest to historians as well as to geographers and economists.”

(51.) See, for example, Wilfong (2007), a vade mecum for the study of gender in late an­
tique Egypt; Dieleman (2005) on the sociocultural milieux of two bilingual papyri from the
“Theban magical library” at the various stages of their existence (composition, compila­
tion, use, and reuse); and Stephens (2003) on the bicultural poetics of Ptolemaic Alexan­
dria (a “classics” book that displays knowledge of the nonliterary evidence).

(52.) I must caution the reader that I have not annotated this description as I would have
a fully developed (and freestanding) work.

(53.) Cf. Will (1985) (with the criticisms of Bagnall, e.g., 1995, 101). Cf. also Stephens
(2003, ch. 4) and Baines (2004, 34).

(54.) See Colin (2002) for indigenous priestesses in the Hellenistic and Roman periods.

(55.) I do not give a precise number because new texts continue to be discovered (and the
relationship of certain papyri to the dossier is still being researched). In this sketch, I
have not counted (or considered) some sixty papyri that recently came to light in Flo­

Page 25 of 28
Writing Histories from the Papyri

rence, Lund, New Haven, and Oxford and which have not yet received sufficient study.
Last treatment of the dossier: Hanson (2001, 602–604).

(56.) The somewhat later Oxyrhynchite dossier of Sarapion aka Apollonianos (also eight
generations) comes to mind, but it (at least as presented in Moioli 1987) cannot compare
to the Tebtynis assemblage in either quantity or, more important, richness. If the literary
material from Kom Gamman could be connected to Sarapion, it might be another story.

(57.) If the structure in which the dossier (or at least the bulk of it) was found was used
for wastepaper or served as a geniza (or as the temple library), only the scale or granular­
ity, not the critical mass, of the argument here is (potentially) impacted.

(58.) “Insular”: Cf., for example, Dieleman (2005, 208). Frankfurter (1998, 222), however,
distinguishes the “culture … of the large Sobek priesthoods in the Fayyum” from that of
priests in other villages. Marriage: The family of Kronion and Isidora was narrowly endog­
amous; only individuals from the priesthood (frequently the top levels of the hierarchy)
were admitted. “Greek”: I have set this in quotation marks because a serious definition of
the culture in question is well beyond the scope of this discussion.

(59.) Pace Tait (1992, 310): “not cut off from Greek culture, but their concern with it was
curiously limited.”

(60.) The hands of Kronion and Isidora (et al.) are extant (the former being more accom­
plished than the latter), as are letters in Greek from both. For Roman-period letters from
Tebtynis priests “in an artificial literary demotic hand,” see Tait (1992, 307 and n. 20).

(61.) Note also Begg (1998, 190): “a page of Homer” (perhaps indicative of a pedagogical
context, as indeed might the Catalogue be).

(62.) Cf. Van Minnen (1998, 112–113,169).

(63.) I agree with the argument in Manning (2003) that the displacement of the temples
as economic institutions began under the Ptolemies; I am thinking of the realia behind
ideology.

(64.) For example, Capponi (2005, 92) and Baines (2004, 34): “By contrast, the Roman
conquest was followed by a reduction in local autonomy and increased discrimination
against those who were culturally non Greco-Roman [sic].” As opposed to the old regime:
“In Ptolemaic Egypt people changed their ethnic affiliation by changing their clothes
(almost)” (Van Minnen 2002, 348). Of course, the Roman transformations were a process,
not an overnight development.

(65.) For example, Capponi (2005, 92), Bowman and Rathbone (1992,120 V.). Needless to
say, the “Greek” ancestry of these populations was “diluted.’’

(66.) The last three sentences of this paragraph are distilled from Bowman and Rath-bone
(1992,120 V.). Van Minnen (2002) has some value for this topic, but I do not accept his re­
visionist claims: that the gymnasial class was not an elite, that it was not more privileged
Page 26 of 28
Writing Histories from the Papyri

or less numerous than the metropolite class, and so on. I cannot address these arguments
here. Ruffini (2006) critiques other issues; I thank him for sharing his work in advance of
publication.

(67.) Cf. Glare (1993). I thank the author for allowing me access to her doctoral disserta­
tion. Now also see P.Dime II, pp. 9–14, for administrative involvement.

(68.) “Preference shown'': cf. Glare (1993). For the traditional view, cf. Dieleman (2005,
208–211).

(69.) Cf. http://tebtunis.berkeley.edu/ancientlives/bagnall.html (September 22, 2008).

(70.) Granovetter (1973) is the foundational text.

(71.) To borrow the expression of the Ammonios responsible for P.Oxy. XIV 16815–7 (third
century CE): βάρβαρόν τινα ἤ Αἱγνπτιον ἀνάνθρωπον, “a barbarian or an inhuman Egypt­
ian.”

(72.) Cf. Appadurai (1996, 89–113), a classic text.

(73.) E.g., Tait (1994, 190, also 191): “The object of an education in Demotic (and Hierat­
ic) must have been the maintenance of the native literate culture, both religious and secu­
lar.” Priests’ antiquarian interests (and literature as history): P.Carlsb. VI, p. 18. One must
be cautious not to make the priests too prescient …

(74.) See, for example, the Tebtynis “Onch-Sheshonqy” in P.Carlsb. III.

(75.) Stated with the caveat that a lot of Tebtynis material remains inaccessible to me. Al­
so see note 76.

(76.) Cf. Colin (2003, 278): “From 37 CE … all the offcial documents issued by the clergy
of the god Soknebtynis (Fayum) identify this divinity with the Greek Kronos; it is tempting
to think, in this case, that the sudden novelty was the product of a decision by the Council
of the temple.”

(77.) There is an explosion of Kronos and Geb names in the documentation from Roman
Tebtynis that seems unlikely to be a Xuke; for the family of Kronion and Isidora, see
Wgure 21.1. (Note that Geb was typically assimilated with Kronos; therefore, “Pakebkis,”
i.e., “The one belonging to Geb,” and “Kronion” are equivalents of a sort.) Holm (1936,
70) notes that the priests favored the indigenous Geb names.

(78.) I assume that Le Roy Ladurie derived his first pair (which goes back to Archilochus,
fr. 201, of course) from Isaiah Berlin's well-known essay on Tolstoy.

(79.) Though, as noted earlier, Manning is certainly cognizant of the potential impact of
“the micro.”

(80.) That papyri are fragmentary is also relevant; cf. Ginzburg (1988).

Page 27 of 28
Writing Histories from the Papyri

(81.) “Anecdotal”: Against this charge, cf. Ricoeur (2004, 213); see also Levi (1991, 96,
109) and Revel (1989, xiii; 1995b, 501). Sewell (2005, 75–77) finds such arguments prob­
lematic. For a general account of the criticisms of microhistory, see Grafton (2006, 63–
67).

(82.) A classic formulation of Sahlins (cf. 1985, 136 ff.); see Sewell (2005, 197 ff.) for dis­
cussion and modification. Cf. also Revel (1995b, 500). These at-risk structures include
those of the historical present, for instance, our own categories of analysis; cf. Feierman
(1999, 206).

(83.) Not even quantitative methods (a favorite tool of macrohistorians) are incompatible.

(85.) Cf. Ruggiero (1993, 18–20) as well as Stone (1979, 3–4).

(86.) Bagnall (1995, 38) seems too pessimistic.

(87.) See Keenan (1992), one of the most important papyrological articles ever published
(in general, Keenan's scholarship—often ahead of the curve—has not received the atten­
tion that it merits); cf. also Keenan (1995). See also Schubert (2000, 9–13) (“Une histoire
pour commencer”), which I typically translate during the first meeting of my undergradu­
ate courses on Graeco-Roman Egypt.

(88.) Alabi: Price (1990); katochoi archive: See the excellent synthesis in Thompson (1988,
ch. 7); Scottsboro: Goodman (1994); Woman Wang: Spence (1978).

(89.) For the petitions of late antiquity, a start has been made; cf. Fournet (2004). James
Keenan informs me that the Catalan “memorials of complaint” (cf. Bisson 1998) are a bet­
ter fit than Davis's documents.

(90.) Richter 2005, 260–261. I thank the author for giving me access to his work in ad­
vance of publication.

(91.) Cf. the emphasis on texts in Bynum (1991, 22).

(92.) Thus Gibbon (1814, 66), but of historiens (since he was writing in French).

Todd M. Hickey

Todd M. Hickey, Assistant Professor of Classics, University of California, Berkeley.

Page 28 of 28
Geography and Administration in Egypt (332 BCE–642 CE)

Geography and Administration in Egypt (332 BCE–642

CE)
Maria Rosaria Falivene
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Egyptology
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0022

Abstract and Keywords

This article outlines the administrative geography of Egypt under the Graeco-Macedonian
regime and as it evolved over the next millennium. This information comes from papyri
and ostraca—sources that differ from the material evidence of archaeological finds yet
are not so distant from these as literary evidence often is. What makes papyrological texts
so informative is their being on the same timescale as the actors involved: their speaking,
as it were, in everyday words. There are some downsides as well. Mahaffy's observation
applies most aptly to the difficulties we encounter when trying to reconstruct the admin­
istrative framework of Graeco-Roman Egypt and having to deal with the intricacies of a
territory and its management as it evolved over time and under changing regimes.

Keywords: administrative geography, Egypt, archaeological finds, papyrological texts, Mahaffy, Graeco-Macedon­
ian regime

THIS chapter outlines the administrative geography of Egypt under the Graeco-Macedon­
ian regime and as it evolved over the next millennium. This information comes from pa­
pyri and ostraca—sources that obviously differ from the material evidence of archaeologi­
cal finds yet are not so distant from these as literary evidence often is.1 What makes pa­
pyrological texts so informative is their being (p. 522) on the same timescale as the actors
involved: their speaking, as it were, in everyday words. As invariably happens with quali­
ties, this one also has its downside, which was described as follows by a pioneer of papy­
rological studies: “For in all such documents, the most necessary assumptions are those
which every contemporary reader took for granted as obvious, whereas we have to infer
or detect them from stray and casual allusions” (Mahaffy 1896, xxxvii). Mahaffy's obser­
vation applies most aptly to the difficulties one encounters when trying to reconstruct the
administrative framework of Graeco-Roman Egypt and having to deal with the intricacies
of a territory and its management as it evolved (conforming to long-established, seeming­
ly timeless patterns) over time and under changing regimes (in the millennium that con­
cerns us here: Ptolemaic, Roman, and Byzantine).

Page 1 of 22
Geography and Administration in Egypt (332 BCE–642 CE)

Geography
Following pharaonic tradition, Egypt under the Ptolemies was divided into an upper
(northern) and lower (southern) country (anô and katô chôra, as they were called with ref­
erence to the south-north course of the Nile), and these in turn were partitioned into an
approximately equal number of districts, or nomes (nomoi, demotic tš: Müller-Wollermann
1992, 243–244), their total number fluctuating over time around an average of forty.
These districts also varied in size and importance; on the whole, however, and in view of
the very long span of time concerned, they remained remarkably stable both as adminis­
trative units (until the early Byzantine period) and (even longer) as cultural systems (Kees
1961; Helck 1974; Bagnall 1993, 333–335). At its core, each district had a temple dedicat­
ed to the principal local divinity (e.g., Arsaphes/Herakles in the Herakleopolite) and a
totemic animal of its own, as illustrated in the seventeenth book of Strabo's Geography
(Strab. 17.1.39–40; see also 17.2.4, and Yoyotte and Charvet 1997, 152–154). Strabo de­
voted this entire book to Egypt; he was an expert witness, having lived and traveled in
Egypt in the 20s BCE in the entourage of the first (Cornelius Gallus) and second (Petron­
ius) Roman prefects.

As a rule, the Greeks named the districts and their capitals (mêtropoleis) after either the
respective presiding deities or their totemic animals, thereby performing an act of inter­
pretatio graeca, as well as an administrative one: The Herakleopolite district, for in­
stance, was thus named on the basis of an equation between the Egyptian Somthous/Ar­
saphes (a god of youthful vigor, or andreia) and Herakles, while the neighboring Kynopo­
lite received its Greek name from the canine god Anoubis (cf. Strabo 17.1.40). The
Fayyum derives its modern name from ancient (p. 523) Egyptian Piom or Phiom (“the
lake”), which was translated as ἡ λίμνη in Greek, but under first Graeco-Macedonian and
then Roman rule it was also called Arsinoe's district (Arsinoites) or Krokodilopolites, from
the (Graecized) name of its metropolis, “the city of the Crocodiles” (the Leuven Fayyum
Villages Project now offers a wealth of information on the geography of this district: see
http://www.trismegistos.org). This renaming procedure, however, went little further than
nome capitals. Villages (kômai, the smallest administrative unit in the land) usually kept
their Egyptian names, which in the Greek documents were simply transliterated and
adapted to Greek case endings, with the exception of new settlements and a few other
places that for some reason (e.g., their good “connectivity degree”: Mueller 2003a,
2003b; Ruffini 2007) acquired importance to newcomers.

Water was brought into the Fayyum by the Bahr Yusuf, a side branch diverging from the
Nile more than three hundred kilometers to the south, on the border with the Thebaid
(the southernmost part of Egypt, which comprised several nomes). Here Ptolemy I found­
ed and named a city after himself: Ptolemais, as Alexandria was named after the Mace­
donian conqueror of Egypt. Just as Alexandria (founded early in 331 BCE on a site the
Egyptians called Rhakotis: Fraser 1972, 3–7; Depauw 2000) was to counterbalance Mem­
phis in the lower country, Ptolemais could be seen as the Graeco-Macedonian counterpart
to Thebes (the ancient capital of Upper Egypt and later on a center of stubborn resis­
tance against the foreign rulers). The Nile and the Bahr Yusuf ran approximately parallel
Page 2 of 22
Geography and Administration in Egypt (332 BCE–642 CE)

through several nomes: Proceeding downstream, the best attested in papyri include Pa­
nopolite, Lykopolite, Hermopolite, Oxyrhynchite, Kynopolite, and Herakleopolite; the Anti­
noopolite district (a small area carved out of the Hermopolite) deserves special mention
because Antinoopolis was the (enduring) creation of Hadrian in memory of Antinoos, who
drowned in the Nile during the Roman emperor's visit to Egypt (130 CE). In the Roman
period this part of Egypt was called “the Seven Nomes,” as attested in both documentary
(BGUII 646 and P.Oxy. XXXIV 2705: two letters from praefecti to the strategoi of the
“Seven Nomes and the Arsinoite,” dating from, respectively, 193 and ca. 225) and literary
sources from around the same time: Claudius Ptolemaeus, the well-known mathemati­
cian, musicologist, astronomer, and geographer (born in Ptolemais), writes in his Geogra­
phy that “the regions to the South of the Great Delta and of the Northern country are
called the Seven Nomes, or Heptanomis” (4, 5, 55).

As the Bahr Yusuf flowed into the Fayyumic depression, the Nile followed its northward
course across the Aphroditopolite and the Memphite and on to the Delta region, which
was also divided into several nomes. East and west of the cultivable land (the chôra, as it
was called in Greek) was the desert. Oases in the western desert (called Libya) were also
inhabited: Strabo records that three of these were adjacent to Egypt and pertained ad­
ministratively to it (17.1.5). The eastern desert bore the name of Arabia.

A list of the nomes of Egypt in the early Ptolemaic period is found (twice, with
(p. 524)

some variation) in the Revenue Laws papyrus, a collection of documents providing guide­
lines for the farming out of taxes, finally revised against the originals of the dioiketes
Apollonios on September 1, 259 BCE, the twenty-seventh year of Ptolemy II (P.Rev. 38, 1;
first edition: Grenfell 1896; new edition: Bingen 1952, and see Bingen 1978). Besides the
eastern and western desert areas (respectively, Arabia and Libya), this list details the
nomes of the delta region and then, starting with the Memphite, those in between the
delta and the Thebaid, which is a collective name for all of the districts of southern Egypt
(the anô chôra; see Bagnall 1993, 333–335, for a list of the districts that compose this re­
gion). There is a remarkable degree of coincidence between the P.Rev. list and the infor­
mation provided by Strabo (Geogr. 17, 1, 18–50; see Grenfell 1896, 50–51) Pliny (Natu­
ralis Historia 5, 49–50), Claudius Ptolemaeus (Geogr. 4, 38–73), and the itinerary given in
P.Oxy.XI 1380 (invocation of Isis, second century CE).

The Administrative Geography of Ptolemaic

Egypt
The process by which the foundations of the Ptolemaic administration of Egypt were
marked out on a geographical grid was not a straightforward one: It developed gradually
and not at the same time even within the same nome. Particularly in the earliest phase of
the Ptolemaic regime, different grids applied to the same territory, depending on the spe­
cial interests of the administrators, who operated in different spheres: This may perplex
modern interpreters of the relevant papyrological material. An administrative bilingual­
ism took time to develop and refine, and this may well explain some apparent confusion,
Page 3 of 22
Geography and Administration in Egypt (332 BCE–642 CE)

which is perhaps a reflection of misleading translations from Egyptian (the language in

which certain functions and titles had been originally conceived) into Greek rather than
an aspect (or problem) of the actual administrative practice in early Ptolemaic Egypt.
Much of the managing and writing—including, one suspects, the invention of appropriate
Greek titles for traditionally Egyptian functions—had to be entrusted to a class, or rather
a caste (ethnos in Greek: Thompson 2001, 1261), of “Hellenes.” These included the Hell­
enizing segment of the Egyptian population: Trained according to a Greek system of edu­
cation (best exemplified by a third-century-BCE school-book from the Fayyum: Guéraud
and Jouguet 1938), they would be able to speak and write both Egyptian and Greek; their
willingness to cooperate was crucial especially in the first decades of Ptolemaic rule,
when much of the administration was in fact recorded in demotic or bilingual documents.
By learning the language (p. 525) of the new pharaoh, while at the same time mastering
the mother tongue of the native population, this caste (or perhaps we should call it a lob­
by) clung as best it could to traditionally privileged positions in the “managing” and “writ­
ing” sectors of the administration of Egypt under the Ptolemies. Connections and indeed
overlaps between this caste and the priestly establishment must have been a matter of
course: Strabo's assertion that “the priests were conversant with the kings” (17, 1, 3; also
17, 1, 5) bolsters this statement.

Having lost the Chremonidean War, which lasted from the summer of 268 to 263/262, and
(just before the war began) his loving (Philadelphos) sister and wife, Arsinoe, who had ac­
tively promoted it (Habicht 1997,143–149), Ptolemy II set out to consolidate his Egyptian
kingdom. Ptolemaic control over the southern Aegean was now at risk, and the Seleucid
menace loomed large and near in Syria and Phoenicia (Bagnall 1976). The Second Syrian
War, pitting Ptolemy II against Antiochus II, broke out around 260 and lasted until 253.
Strict, clear-headed military and economic measures were needed; it is perhaps not a co­
incidence that a demotic ostracon from Karnak (dated between October 27 and Novem­
ber 25, 258), summing up instructions for a general survey of land (Bresciani 1978, 1983;
Burstein 1985,122–123n97), appeared only one year after the Revenue Laws papyrus.
Even if these two documents were not an attempt at fiscal or economic codification (Bin­
gen 1952, 3), they nevertheless presupposed or at least aimed at some kind of standard
practice in the surveying of cultivable land and in the farming of taxes, which in turn as­
sumed a well-defined geographical and administrative grid.

Different spheres of the Ptolemaic administration may be defined with reference to the
following Greek verbal roots:

ag- (lengthened by vowel mutation: hence êg-), for the “leading,” in the first place,
of army troops; hence the titles strategos and hegemon for the commanders of, re­
spectively, larger and smaller units in the Ptolemaic army;

arch-, as found, for example, in the title nomarches (“nomarch” in its Anglicized
form) for the management of all kinds of economic activities, foremost among
them agricultural production and the collection of taxes;

Page 4 of 22
Geography and Administration in Egypt (332 BCE–642 CE)

graph-, as in grammateus (“scribe”) and antigrapheus (“checking clerk”) for the

writing or recording of facts and figures, an essential operation in view of the cor­
rect assessment of exploitable resources;

oik- asin oikos (“house, household”), the whole of Egypt being conceived of as
Ptolemy's household, to be governed by a dioiketes in Alexandria, whose functions
were replicated by the oikonomos (one in each district).

In the demotic documents it was normal practice for the Egyptian officials who wrote
them to employ the Egyptian equivalent for official titles (Clarysse 1987, 12): It thus
seems possible to indicate the equivalents for the Greek arch- (“managing”; sḥn:
“ordonner, équiper, protéger, confier”: Héral 1990, 305; and see Sethe and Partsch 1920,
(p. 526) 163) and the Greek graph-(“writing”; sẖn, as found in sẖ n Pr-‘3, the “pharaoh's

scribe,” or basilikos grammateus; Kruse 2002, 11–22). On the other hand, the titles of
strategos (demotic s3trks, in various spellings), hegemon (demotic hgmn), and oikonomos
(demotic 3knwms, in various spellings) are as a rule simply transliterated (Clarysse 1987,
21–32): This observation is consistent with the initial predominance of Graeco-Macedon­
ian immigrants in the military and financial spheres.

At the central level, in Alexandria, the dioiketes acted as the king's chief finance and inte­
rior minister; the office of the idios logos (“special account”) was in charge of the admin­
istration of nonrecurring income (Bagnall 1976, 5; Swarney 1970); a hypomne­
matographos and an epistolographos superintended the royal chancery (P.Tebt. III 703, in­
troduction, 68–70): Various laws, regulations, orders, and instructions emanating from
the king (nomoi, diagrammata, prostagmata, programmata: Lenger 1980,1990) would
originate from here, quite often as a reaction to petitions (enteuxeis) from his subjects.

The key figures in the Ptolemaic administration at the level of the nome are all listed in
the address of an ordinance (prostagma) of November–December 263, introducing the
edict (programma) by which the one-sixth tax (apomoira) on the produce from vineyards
and orchards was directed to the cult of Arsinoe Philadelphos (traditionally, the native
Egyptian temples were entitled to this share of the agricultural produce): “King Ptolemy
to all strategoi, hipparchs, captains (hegemones), nomarchs, toparchs, oikonomoi, audi­
tors (antigrapheis), royal scribes (basilikoi grammateis), libyarchs, and chiefs of police
(archiphylakitai), greeting (P.Rev. 37, 2–5). In this list, the first officials named are the
army leaders (strategoi, hipparchs, hegemones), who oversaw the military control of the
conquered territory, and last mentioned are the archiphylakitai, who were in charge of
the day-to-day policing of all economic activities in the nome (Thompson 1997b). Between
them are listed the civil officials (including those in charge of the oases of western Egypt:
the libyarchs), who attended to routine administrative matters.

Page 5 of 22
Geography and Administration in Egypt (332 BCE–642 CE)

Table 22.1. The Basics of Ptolemaic Administration

Branches of Administration

Administrative units οἶκος ἄρχειν (sḥn) γράϕειν (sẖn) ἄγειν

Alexandria/Rhakotis dioiketes hypomnematographos,

epistolographos

Nomos (tš), oikonomos nomarches p3 sḥn dnj.t basilikos grammateus, Strategos s3trks
metropoleis antigrapheus

Meris (dnj.t) meridarches p3 sḥn

dnj.t

Topos/toparchy (qḥ) toparches p3 sḥn t3 qḥ. topogrammateus

Kôme komarches komogrammateus

Page 6 of 22
Geography and Administration in Egypt (332 BCE–642 CE)

(p. 527) The basic structure of early Ptolemaic administration is summed up in table 22.1.

The basilikos grammateus (“royal scribe”) was the pivotal figure in the administration of
the nome (Kruse 2002); his immediate superior was the strategos. The task of a basilikos
grammateus was to write down and keep written records of the facts and figures relating
to all of the transactions in the nome; foremost among these were the census (by house­
hold, oikia, or fiscal category, ethnos: Clarysse and Thompson 2006) and the survey of cul­
tivable and cultivated land. An accurate survey of the land (after the Nile flood) and of the
produce (before the harvest) was an obvious prerequisite for the working out of the
country's fiscal system (Verhoogt 1998, 2005). Important financial responsibilities there­
fore accrued to the “writing sector” of the early Ptolemaic administration, and this ex­
plains how the basilikos grammateus eventually came to replace the oikonomos (Kruse
2002, 890). No reference to an oikonomos is found in the late Ptolemaic and early Roman
documents from the Herakleopolite published in the BGU series (BGU IV, VIII, XIV, XVI),
with the exception of BGU XIV 2370, a somewhat earlier document (dated around 84/83
BCE) that provides an interesting terminus post quem for the disappearance of the
oikonomos.

At the lower administrative levels of meris, topos, and kômê (“division, region,” “place,
area” and “village,” respectively), meridarchai (“managers of a meris”), topogrammateis
(“scribes of the topos”), and toparchai (“managers of the topos”), as well as komogram­
mateis (village scribes) and komarchai (“managers of the village”) are found. A topos
comprised a varying number of villages (kômai) and could be referred to by the name of
its main center: “in the area of” (Greek perì) + village name; alternatively, its name might
reflect geographical conditions: Katô, Mesê, Anô, with reference to the northern (“low­
er” ), central or southern (“upper”) location (e.g., in the Oxyrhynchite nome: Pruneti
1981) or even derive from historical circumstances: Agêma, for instance, was the name of
a topos in the Herakleopolite nome that had been destined for occupation by soldiers of
the homonymous special contingent of the Macedonian army (Falivene 1998, 37–39). In a
third-century BCE demotic document from al-Hiba (CGC 50148, 6–7), the designation p3
sḥn t3 qḥ applies to the same official (Petosiris), who was referred to by the Greek title
toparches in P.Hib. 75.2–3 (Müller-Wollermann 1992, 244); accordingly, qḥ should be tak­
en as the Egyptian equivalent of topos (in this document, the topos Koites is meant), just
as sḥn corresponds to archein. The term toparchy (toparchia), on the other hand, proba­
bly referred to a toparch's managing function (toparches), soon to be applied, as it were
by metonymy, to the area in which he exercised this authority (one or more topoi, or parts
thereof). In other words, one and the same official could be the toparches of more than
one topos; consequently, a toparchy could be a larger entity than a topos; by the same to­
ken but on a smaller scale, a komarches could be in charge of more than one village
(kômê).

(p. 528) As military conquest turned into military occupation of Egypt, a large part of the
Graeco-Macedonian army was effectively demobilized. In lieu of their wages, army men
were granted land: This was to all effects a payment in kind, one that turned them into
landholders (or cleruchs: klerouchoi) and rentiers and made them “at home” (katoikoi), fi­

Page 7 of 22
Geography and Administration in Egypt (332 BCE–642 CE)

nally settled in the Egyptian chôra. Significantly, the Egyptian language had no equivalent
for this new institution; the word katoikos, therefore, was simply borrowed from Greek
and transliterated in demotic characters (gtwks, and at least one more variant spelling:
Clarysse 1987, 25). Cavalry settlers (katoikoi hippeis) received the largest land holdings—
up to one hundred arouras for the “hundred-aroura men” (hekatontarouroi). The names of
these original settlers (hoi prôtoi: “the first” Graeco-Macedonian settlers in a nome:
Uebel 1968, 167n3; Kramer 1991, 78) recur in land-survey documents long after they had
left their landholdings (Falivene 2007), still grouped by their army unit, sometimes under
the names of their respective commanders, the hegemones (Kramer 1991, 74–80; see also
Boswinkel and Pestman 1982, 34–55). Crowning the same move from conquest to occupa­
tion of the land, a general (strategos) came to be at the head of the nome (Mooren 1984).
Thus, the Ptolemaic army controlled and shaped the administrative system, which was in
fact mapped onto the Egyptian land, resulting in a kind of cadastral reference system that
lasted into the Roman period. This is what Strabo may have meant when he wrote that
the smallest administrative divisions in the nome (merides) were the arourai (17.1.3); that
is, rather than referring strictly to the Egyptian land measure, he was perhaps thinking of
the names that remained attached to the original land holdings, or “fossil klêroi” (Zucker
1964; Falivene 1998, 273–288; 2007), long after they had been split into smaller units and
variously reapportioned to new grantees.

The apportionment of land to new settlers and afterward the organization of land cultiva­
tion cannot have been immune to at least potential tensions with the preexisting organi­
zation of the Egyptian territory. The standard procedure for the granting of land to new
settlers required that the land be apportioned in large tracts of ten thousand arouras ei­
ther to a single grantee (in the mid-third century BCE, this was the rather exceptional
case for the dôrea of the dioiketes Apollonios: see Clarysse and Vandorpe 1995) or for
cleruchs: An agent known as a muriarouros (a “ten-thousand-aroura man”) was in charge
of preparing the land for distribution and cultivation (Clarysse 1997, 75). The superior of
a muriarouros was in all likelihood the nomarch (nomarches, “manager of the nome”):
Their special relation is shown in P.Petr.II 42, where they appear together, along with var­
ious other officials; it is further confirmed by the fact that the Ghoran cartonnage papyri
concerning a muriarouros originate from the archives of nomarchai (Falivene 2000). As­
suming that there was traditionally (and possibly still in the earliest stages of the Ptole­
maic administration) just one nomarches in a nome, in time two changes apparently af­
fected this official. On the one hand, a general of the invading army (a strategos) was put
at the head of the nome; in addition, the nomarch's authority was reduced by dividing it
among several (p. 529) nomarchs, each of whom was in charge of just a
“portion” (Egyptian dnj.t, and in Greek, moira, meris; Héral 1990, 308–311; also Müller-
Wollermann 1992, 245) of the nome; as many as seven nomarchs are attested for the Arsi­
noite nome in the third century BCE (SB XXIV15937; Clarysse 1997, 70–72). The Egyptian
title of these “plural” nomarchs was p3 sḥn dnj.t, and a more fitting Greek rendering
would be meridarches (the Greek root arch- corresponds to the Egyptian root sḥn, just as
meris is rendered as dnj.t). Meridarchai are in fact attested for the Herakleopolite nome
by several documents dating from the first century BCE, including BGU XIV 2370; of the

Page 8 of 22
Geography and Administration in Egypt (332 BCE–642 CE)

three meridarchai referred to in this document, at least one was responsible for more
than one toparchy (Brashear 1980, 10). This shows that at the time the Herakleopolite
was divided into merides (and these in turn into toparchies), just like the Arsinoites,
where the three merides of Herakleides, Polemon, and Themistos are attested beginning
in the third century BCE.

Some relatively late attestations for nomarchs in the Arsinoite district (dating from the
230s BCE; Clarysse 1997, 72–76) may be accounted for by supposing that the smaller,
therefore more numerous, merides of the nomarchs were incorporated into three, gradu­
ally larger divisions not all at once but rather one nomarch at a time, insofar as each
nomarch's task within the overall reclamation and resettlement plan of the Fayyum (see
Thompson 1999a, b) was completed. This also is consistent with the fact that muriarouroi
are not attested in documents after 247/246 (Clarysse 1992–1993, 216). Harimouthes, an
Egyptian nomarch attested in the lower (i.e., northern) toparchy of the Oxyrhynchite
nome at an early date (P.Hib.I 85, of 261 BCE) and reappearing in P.Hib.I 44 (of 253 BCE)
in the position of toparch of the same toparchy, may be a case in point: While his job de­
scription remained apparently much the same, the shift in terminology may reflect a
change in the administrative hierarchical structure; perhaps Harimouthes now had a su­
perior, a meridarches at the head of a newly instituted meris that embraced more than
one toparchy, and was thus made redundant as a nomarch. A process of this kind may ex­
plain the overlap (Clarysse 1997) between nomarchs and toparchs in the Fayyum in the
230s. The possibility for a nomarch or a toparch to function in each other's place is in fact
explicitly stated in P.Rev. 41,14–17, where either the nomarch or the toparch, in the ca­
pacity of “official in charge of the nome,” is entitled to obtain a seed loan from the
oikonomos “before the season comes for sowing the sesame and croton … if they so
wish.” Later on, “when the season comes for gathering the sesame, croton, and cnecus,
the cultivators shall give notice to the nomarch and the toparch, or where there are no
nomarchs or toparchs to the oikonomos” (P.Rev. 42, 3–7). Once more, nomarch and
toparch (or just one of them) may be present—or indeed both may be absent, in which
case their superior, the oikonomos, will act in their place; ideally (if theoretically), all
three should preside on the assessment of the crops.

Even though there were Egyptians among the nomarchs (Pathembris and Horos are at­
tested among the seven officials with this title in SB XXIV 15937), by the end of the
process of reclamation and resettlement all three divisions of the (p. 530) Arsinoite district
were named after Greek officials: Once again, the apportioning of land to Greek settlers
appears to have been coterminous with its renaming in Greek. According to much the
same principle, the whole Fayyum came to be called “Arsinoe's district,” although this had
not yet occurred in the Revenue Laws papyrus (where it is still called he Limne, “the
lake”), but all in all, the impression one gets from this document is that by the 250s the
Graeco-Macedonian newcomers (first among them, their king) felt that they had gained
command of the geographical and human landscape of Egypt and that, although they still
needed (as they always would) cooperative Hellenizing Egyptians, they now knew how to
adequately impose their requirements. If and when required, a Greek could now assume
the position of head manager” in the nome and perform this task effectively: At this stage,
Page 9 of 22
Geography and Administration in Egypt (332 BCE–642 CE)

Egyptians may have been demoted from positions of authority they had previously (and
traditionally) held as nomarchs. Indeed, the very title nomarches vanishes from our docu­
ments.

Again see table 22.1 for the basics of Ptolemaic administration.

Administration and the Ptolemaic Fiscal Sys­

tem
Adapting a practice that had been established in step with the development of a monetary
economy within the Athenian empire (Préaux 1939, 450–459), the Ptolemies farmed out
their taxes: During a yearly audit, tax farmers (telonai) would bid and pledge themselves
for a given revenue, thereby acquiring the right to keep for themselves any surplus that
might occur. They could also operate as a company, consisting of a chief farmer (ar­
chones) and his associates (metochoi). It was possible to farm one or more taxes for one
or more districts. A double surety was required in that somebody else must go bail (en­
guetes, “surety”) for the tax farmer in case he became unable to compensate all or part of
a deficit in the expected tax income.

Banks (trapezai) and granaries (thesauroi: basically, royal banks for the economy in kind),
respectively under a trapezites (“banker”) and a sitologos (“grain accountant”), were cru­
cial in ensuring the deposit, exchange, and circulation of money and grain according to a
complex agenda that we can reconstruct in detail on the basis of the Revenue Laws
papyrus. Trapezai were farmed out (Bogaert 1994). The underlying principle (but not the
scale and scope of activity, of course) was the same as that applied in the farming out of
the exclusive trading license (i.e., a monopoly: monopôlia) for other, less ambitious com­
mercial enterprises, such as the exclusive right to sell lentil soup in Choinotbis, a village
in the Herakleopolite district (Uebel 1964). What a contractor of the trading license for
oil (the elaikê) acquired, for instance, was (p. 531) an exclusive right to buy sesame, cro­
ton, and the produce of other oil plants at wholesale prices from the grower while these
crops were still on the threshing floor; he would then sell the resulting oil at varying
prices in Alexandria, the Libyan oases, and “all over the country in all the towns and vil­
lages” (P.Rev. 40,14,18).

With the right, however, came the obligation to buy and sell fixed quantities at fixed
prices. The specification of prices is the focus of the Revenue Laws papyrus insofar as
these texts represent an application of the law, not the laws themselves. As for expected
quantities, these were established on the basis of a preliminary assessment of the pro­
duce. Before harvest time, the contractor and the cultivator appraised the crops in the
presence of the oikonomos and his checking clerk, the antigrapheus: This was in fact the
second inspection, following an earlier one to determine the number of arouras to be
sown with oil plants under the responsibility of the nomarch and/or the toparch, the
oikonomos, and the antigrapheus, all of whom were subject to a double fine (to both the
royal treasury and the contractor) if, at the time of the second assessment, “they find that

Page 10 of 22
Geography and Administration in Egypt (332 BCE–642 CE)

the right number of arouras has not been sown.” The oikonomos and the antigrapheus
might thus find themselves in the somewhat peculiar situation of fining themselves, hence
the order that “the dioiketes shall exact the payment from them” (P.Rev. 41, 5–13).

The nomarch and the toparch, though liable for the fine for incorrect land assessment,
are conspicuously absent during the second produce assessment, which tells us some­
thing about the limitations of their authority. They would apportion land to cultivators and
designate it for the production of certain crops, and they were responsible for the results
of these undertakings; in fact, they acted as the interface between the cultivators and the
other parties, whether contractors or officials other than themselves, but they had no
part in the evaluation of prospective produce, not to mention the fixing of prices for the
various crops; in short, the nomarch and the toparch had no say in the financial decisions.

The same pattern applied at village level: Here only the komarch could allow the produce
to leave the village, but only after the contractors had given him a sealed receipt that list­
ed what they had bought from each cultivator (P.Rev. 40, 1–8) at the value decreed in the
legal tariff (P.Rev. 39, 16–17). The komarch controlled the transactions between the culti­
vator and the contractor and ensured that the produce was sold by the cultivators to the
contractors and to nobody else; he also made certain that the contractors paid the pre­
scribed amount to the cultivators. But just how much produce was to be obtained and
sold and at what prices was for others (the oikonomos and the antigrapheus, upon agree­
ment with the contractor and the cultivator) to decide.

On the whole, the Ptolemies punctiliously pursued a “can't lose” fiscal and economic
strategy aimed at ensuring that the Crown would in any case receive the expected rev­
enue; any deficit had to be compensated by a class of middlemen. This class comprised
not just the contractors but also the cultivators (georgoi), who were not actually workers
of the land (the ancient counterpart of the fellahin) but rather (p. 532) their employers,
who leased large tracts of land from absentee landholders (Bingen 1978, 1983).

Whether this sophisticated system was a viable one and how long it lasted are different
issues: It could evidently work only as long as the middlemen found it advantageous. A
well-known document (P.Tebt. III 703; transl. Austin 1981, 432, no. 256) from the end of
the third century BCE, possibly a circular from the dioiketes to an oikonomos being “sent
to the nome” (1.258), hints at problems arising at various stages in the very same proce­
dure envisaged by the Revenue Laws papyrus and recommends (11.222–230) the follow­
ing: “Take especial care that no act of extortion or any other misdeed is committed. For
everyone who lives in the country must clearly know and believe that all such acts have
come to an end and that they have been delivered from the previous bad state of affairs.”
The Ptolemies may have been unable to conceive of a different system for exploiting the
land and the people of Egypt (Bingen 1978, 11), but they also probably had no choice be­
cause of the limited number of their own people in Egypt, as well as the resistance they
encountered (or would have met) on the part of the local population and establishment
(especially the priestly class), had part of the Egyptian elite not been involved in the com­
plex and controlled “outsourcing” that I have outlined.

Page 11 of 22
Geography and Administration in Egypt (332 BCE–642 CE)

Later Developments: the Roman Period

Nomes were sometimes split, leading to the creation or reestablishment of smaller dis­
tricts: The Koite and Neilopolite, for instance, were at different times (late Ptolemaic and
early Roman period, respectively) carved out of the southern and the northern Herak­
leopolite, respectively. The Antinoopolite was created by Hadrian. Between 137 and 254
CE the three divisions of the Arsinoite nome were reduced to two, those of Themistos and
Polemon being under the same strategos and the meris of Herakleides under a different
one (this arrangement lasted for about a century; thereafter, the whole Arsinoite was
again under one strategos; Bastianini-Whitehorne 1987, 46–55). But despite these
changes, the administrative map of Egypt—the articulation of the land into nomes,
merides, toparchies, and villages, as well as the titles of the officials in charge of their ad­
ministration—remained basically stable under Roman rule.

Beneath the appearance of continuity, however, important shifts in authority led to a re­
distribution of responsibility and power among old and new officials—first among these
was the emperor's deputy, praefectus Alexandreae et Aegypti (Bowman 1986, 65–68)—
and, at a more fundamental level, among the various components of the population of
Egypt. The keyword here is municipalization, a “process … definitely (p. 533) and deliber­
ately begun in the Augustan period with the creation of urban communities with ‘Hel­
lenic’ landowning élites,” leading to the formal constitution of town councils (boulai) in
Egypt by Septimius Severus in 200/201 CE (Bowman and Rathbone 1992, 108). The boulê
was to function as a compensation chamber between the central administration and the
revenue-producing land: It did this by ensuring that the fundamental administrative oper­
ations of tax collecting and recordkeeping were adequately performed by local liturgical
officials (“some from its own membership, some not, depending on the task”: Bowman
1986, 71) under the supervision of the strategos and the basilikos grammateus, who for
their part became career bureaucrats who were appointed by the government and served
only outside their home district (Kruse 2002, 44).

In the reign of Philip the Arabian (244–249 CE), a reform of the taxation and administra­
tive system took place that entrusted a commission of the “ten first men” (dekaprotoi) in
the boulê with the administration of the nome. At the same time, the office of the basilikos
grammateus was abolished (Kruse 2002, 940–952), and in fact, the whole bureaucratic
line of “writing officials” appears to have been dismissed. As the dekaprotoi operated at
the toparchy level, the topogrammateis must also have become redundant (latest dated
attestation: P.Laur.1 4, 246 CE), while the komarches took over the functions of the komo­
grammateus (Thomas 1975, 115; Borkowski and Hagedorn 1978, 781–783). The limited
number of dekaprotoi, combined with the heavy burden that went with their office, ex­
plains the rule of collegial responsibility for these officers, who were to operate by
toparchy in boards of at least two (toparchies were paired off so that a team of at least
four dekaprotoi was in charge at any one time; Bagnall 1978). The same principle of colle­
gial responsibility apparently applied to toparchs and komarchs as well (Vitelli 1906 on

Page 12 of 22
Geography and Administration in Egypt (332 BCE–642 CE)

P.Flor.I 2, 265 CE; see also Oertel 1917, 163). Some rearrangement of the toparchies (dis­
cussed later) must also have occurred in this reorganization.

“What the Romans had been trying to do for the past century, namely to make the local
property-owning class responsible for the orderly and complete collection of the taxes
owed to the central government” (Bagnall 1993, 55) was now accomplished. The dekapro­
toi, however, soon found themselves caught in an “administrative web …. Responsible in
every particular to their superiors, they were forced to make up losses by
oppression” (Turner 1936, 11). In about sixty years, a new reform was carried out whose
beginnings went back to Diocletian's reign (284–305); apart from the dioiketes himself
(Hagedorn 1985), its “victims” included the dekaprotoi, the strategos, and indeed the
nome itself as an administrative entity. The dekaprotoi were now to serve for a five-year
term coincident with the five-year cycle of the newly instituted tax schedule (epigraphê)
issued each year in the early summer (Bagnall and Thomas 1978, 186); in return, they
were excused from renomination. Eventually, “at a date between 3 May and 2 July 302,”
the dekaprotoi ceased to function altogether (Thomas 1974, 68), and from 307/308 on,
numbered toparchies were replaced by numbered pagi, each headed by a praepositus pa­
gi.

(p. 534) Later Developments: The Byzantine Period

The process that led to the creation of the new geographical and administrative grid by
pagi may be summed up as follows. First, the number of toparchies was increased and
their size reduced, with a view to making them more “manageable”; the resulting new
toparchies were referred to by number, though naturally the traditional names persisted
for a while: A few Fayyum documents from the third decade of the second century CE
attest to this intermediate stage, when “toparchies are at the same time numbered and
named after a village” (Derda 2003, 45). Second, the creation in the third century CE of
the office of the dekaprotoi (whose area of responsibility was defined by toparchies) must
have reinforced and stabilized the new grid of numbered toparchies. Third, at the begin­
ning of the fourth century, as the dekaprotoi were being phased out, pagi became the new
administrative unit, possibly overlapping the numbered toparchies, as distinguished from
the older ones (those named after a village or with reference to either their location in
the nome or some other special characteristic of their territory); this is suggested by the
occasional interchangeability of the terms pagus and toparchia in certain fourth-century
documents (Sijpesteijn and Worp 1978, 9–10); it also goes a long way toward explaining
the often-observed discrepancies between the number of old toparchies and of pagi in the
Oxyrhynchite (Pruneti 1989), Hermopolite (Sheridan 1998, 106–134), and Arsinoite
nomes (Derda 2001).

The very existence of the nomes as administrative units was curtailed as a result of the in­
troduction of the pagi. The cultural persistence of the nomes, however, is still apparent in
the Notitia Dignitatum, a “register of civil and military offices of the Roman/Byzantine
empire compiled in the 390s,” in which “the towns of the delta are schematically repre­

Page 13 of 22
Geography and Administration in Egypt (332 BCE–642 CE)

sented and from them the traditional symbols of the various nomes emerge” (Bowman
1986, 81–82). In the capitals of the nomes (the mêtropoleis of old, now civitates) the strat­
egos, now divested of many of his powers, acquired the new title of exactor civitatis,
while his other responsibilities were diverted to the logistes (or curator civitatis) and later
to the defensor civitatis (Thomas 2001, 1253; Kruse 2002, 952–953).

After 293/294, following the suppression by the junior emperor Galerius of a revolt in
Coptos, the Thebaid became administratively separate from the rest of Egypt. This divi­
sion was a lasting one and was carried over in the several successive attempts—through­
out the fourth century and down to Justinian's Edict XIII (ca. 537/538)—at reorganizing
the administrative structure of Egypt (Rouillard 1928; Lallemand 1964, 41–57; Bowman
1986, 78–79).

The inauguration of the new fifteen-year tax cycles (in which each year was called an in­
diction), which reckoned retrospectively from the year 312, tallied with (p. 535) the
restoration of church property (313), which followed immediately upon the cessation of
the official prosecution of Christians as soon as Constantine finally came to power. The
church promptly affirmed itself as an alternative source of authority; it was within the
(much-divided) Church of Egypt that the conflict between Alexandria and Constantinople
acted itself out (Haas 1997; Heinen 1998). The foundation of Constantinople (formally cel­
ebrated on May 11, 330) at the same time undermined Alexandria's position as the princi­
pal city of the Greek-speaking east and reoriented Egypt eastward. The bishop, or patri­
arch of Alexandria, in fact became the most powerful figure in Egypt (Rouillard 1928,
229–239): Unlike officials who were but deputies of the distant emperor, he could claim
and exercise direct authority in both Alexandria and the whole country by means of the
bishops whom he appointed in the mêtropoleis and also through the village clergy (pres­
byters and deacons, the bishop's appointees). Inland, the country witnessed the “collapse
of the institutional basis of separate Egyptian culture, the temples,” and, running parallel
to it, the dismantling of village institutions: “By the end of the fourth century it is normal
to find presbyters and deacons as the representatives of the village population” (Bagnall
1993, 316). On the other hand, the existence of large landowners effectively favored “the
integration of the villages into the city economy and society.” Increasingly, these wealthy
families were made responsible for village tax collection, thereby “leading to the privati­
zation of public business and the bureaucratization of the private” (ibid., 318): in sixth-
century Oxyrhynchus, the Apion family provides a well-known example of this double shift
(Gascou 1985; Mazza 2001).

New ways of life and changed bureaucratic patterns are reflected in the administrative
language, too: From the seventh century on, the term home, for “village,” is replaced by
chôrion, with an emphasis on a village as part of the countryside (the chôra), that is, the
territory now integral to a mêtropolis. The pagarch (pagarches or pagarchos) first ap­
pears in our sources toward the end of the fifth century: He was responsible for collect­
ing taxes from all subjects who were not exempted—on the strength of autopragia, or (fis­
cal) “self-determination”—from his authority (Liebeschuetz 1990). Since this function (pa­
garchia) was apparently attested before the officer who exercised it (Mazza 1995), one

Page 14 of 22
Geography and Administration in Egypt (332 BCE–642 CE)

may suppose that the praepositus pagi, having lost control of too large a part of his pagus,
was superseded by a new official, the pagarch (Rouillard 1928, 52–62), who also took on
what had been a komarch's responsibilities.

Arab pagarchs are attested well into the eighth century CE, when they are found address­
ing, still in Greek, the local communities (Gonis 2004). The new conquerors—as Alexan­
der and the early Ptolemies had been—were well aware of the importance of avoiding a
breakdown of the administration (Rouillard 1928, 248; Grohmann 1959, 33–34; Christides
1993) if they were to consolidate their occupation of the Egyptian land.

Bibliography
Austin, M. M. 1981. The Hellenistic World from Alexander to the Roman Conquest: A
Selection of Ancient Sources in Translation. Cambridge: Cambridge University Press.

Bagnall, R. S. 1976. The Administration of the Ptolemaic Possessions outside Egypt. Lei­
den: Brill.

Bagnall, R. S. 1978. “The Number and Term of the Dekaprotoi.” Aegyptus 58: 160–167.

———. 1993. Egypt in Late Antiquity. Princeton: Princeton University Press.

———, and D. W. Rathbone, eds. 2004. Egypt from Alexander to the Copts: An Archaeolog­
ical and Historical Guide. London: British Museum Press.

Bagnall R. S., and J. D. Thomas. 1978. “Dekaprotoi and Epigraphai.” Bulletin of the Ameri­
can Society of Papyrologists 15: 185–189.

Baines, J., and J. Málek. 2000. Cultural Atlas of Ancient Egypt, rev. ed. New York: Facts on
File.

Bastianini, G., and J. E. G. Whitehorne. 1987. Strategi and Royal Scribes of Roman Egypt:
Chronological List and Index. Papyrologica Florentina 15. Florence: Edizioni Gonnelli.

Bingen, J. 1952. Papyrus Revenue Laws: Nouvelle édition du texte (Sammelbuch Griechis­
cher Urkunden aus Ägypten, Suppl. 1). Göttingen: Hubert & Co.

———. 1978. “The Third-century B.C. Land Leases from Tholthis.” Illinois Classical Stud­
ies 3: 74–80.

———. 1983. “Les cavaliers catoeques de lʼHéracleopolite au Ier siècle.” In Egypt and the
Hellenistic World: Proceedings of the International Colloquium, Leuven, 24–26 May 1982,
ed. E. Vanʼt Dack, 1–11. Studia Hellenistica 27. Leuven: Studia Hellenistica.

Bogaert, R. 1968. Banques et banquiers dans les cités grecques. Leiden: A. W. Sijthoff.

———. 1994. Trapezitica Aegyptiaca: Recueil de recherches sur la banque en Égypte gré­
co-romaine. Papyrologica Florentina 25. Florence: Gonnelli.

Page 15 of 22
Geography and Administration in Egypt (332 BCE–642 CE)

Borkowski, Z., and D. Hagedorn. 1978. “Amphodokomogrammateus: Zur Verwaltung der

Dörfer Ägyptens im 3. Jh. n.Chr.” In Le monde grec: Pensée, littérature, histoire, docu­
ments. Hommages à Claire Préaux, ed. J. Bingen, G. Cambier, and G. Nachtergael, 775–
783. Brussels: éditions de lʼUniversité de Bruxelles.

Boswinkel, E., and P. W. Pestman, eds. 1982. Les archives privées de Dionysios, fils de
Kephalas (P.Lugd.Bat. 22): Textes grecs et démotiques. Leiden: Brill.

Bowman, A. K. 1986. Egypt after the Pharaohs. 332 BC-AD 642: From Alexander to the
Arab Conquest. Berkeley: University of California Press.

———, and D. W. Rathbone. 1992. “Cities and Administration in Roman Egypt.” Journal of
Roman Studies 82: 107–127.

Boyaval, B. 1973. “Papyrus ptolémaïques inédits de Ghôran et Magdôla.” Cahiers des

Recherches de lʼInstitut de Papyrologie et dʼégyptologie de Lille 1: 215–216.

Brashear, W. M., ed. 1980. Ägyptische Urkunden aus den Staatlichen Museen Berlin:
Griechische Urkunden. Vol. 14, Ptolemäische Urkunden aus Mumienkartonage. Berlin:
Staatliche Museen Preussischer Kulturbesitz.

Bresciani, E. 1978. “La spedizione di Tolemeo II in Siria in un ostrakon demotico inedito

da Karnak.” In Das ptolemäische Ägypten: Akten des internationalen Symposions 27–29
September 1976 in Berlin, ed. H. Maehler and V. M. Strocka, 31–36. Mainz: Philipp von
Zabern.

———. 1983. “Registrazione catastale e ideologia politica nellʼEgitto tolemaico.” Egitto e

Vicino Oriente 6:15–31.

Burstein, S. M. 1985. Translated Documents of Greece and Rome. 3. The Hellenis­

(p. 537)

tic Age from the Battle of Ipsos to the Death of Kleopatra VII. Cambridge: Cambridge Uni­
versity Press.

Calderini, A., and S. Daris. 1935–2007. Dizionario dei nomi geografici e topografici
dellʼEgitto greco-romano. 1.1. Cairo: Società Reale di Geografia dʼEgitto-1.2

.
Madrid: Consejo Superior de Investigaciones Cientificas-2–5

,
Suppl. 1. Milan: Cisalpino-La Goliardica

.
Suppl. 2. Bonn: Habelt

.
Suppl. 3. Pisa: Giardini

Page 16 of 22
Geography and Administration in Egypt (332 BCE–642 CE)

Suppl. 4. Roma: Fabrizio Serra.

Christides, V. 1993. “Continuation and Change in Early Arab Egypt as Reflected in the
Terms and Titles of the Greek Papyri.” Bulletin de la Société dʼArchéologie dʼAlexandrie
45: 69–75.

Clarysse, W. 1987. “Greek Loan-Words in Demotic.” In Aspects of Demotic Lexicography:

Acts of the Second International Conference for Demotic Studies, Leiden, 19–21 Septem­
ber 1984, ed. S. P. Vleeming, 9–33. Leuven: Peeters.

———. 1992–1993. “A New Muriarouros in a Bilingual Text.” Enchoria 19/20: 215–217.

———. 1997. “Nomarchs and Toparchs in the Third-century Fayyum.” In Archeologia e pa­
pyri nel Fayyum: Atti del Convegno Internazionale. Siracusa 24–25 maggio 1996, 69–76.
Syracuse, Italy: Istituto internazionale del papiro.

———, and D. J. Thompson. 2006. Counting the People in Hellenistic Egypt

.
Vol. 1, Population Registers (P.Count.)

.
Vol. 2, Historical Studies. Cambridge: Cambridge University Press.

Clarysse, W., and K. Vandorpe. 1995. Zenon, un homme dʼaffaires grec à lʼombre des
Pyramides. Leuven: Presses Universitaires.

Depauw, M. 2000. “Alexandria, the Building Yard.” Chronique dʼégypte 75: 64–65.

Derda, T. 2001. “Pagi in the Arsinoite Nome: A Study in the Administration of the Fayum
in the Early Byzantine Period. With an Appended Edition of P.Aberd. 164 descr. by Niko­
laos Gonis.” Journal of Juristic Papyrology 31: 17–31.

Derda, T. 2003. “Toparchies in the Arsinoite Nome: A Study in Administration of the

Fayyum in the Roman Period,” Journal of Juristic Papyrology 33: 27–54.

Drew-Bear, M. 1979. Le nome Hermopolite: Toponymes et sites. Missoula, Mont.: Schol­

ars Press.

Falivene, M. R. 1991. “Government, Management, Literacy: Aspects of Ptolemaic Admin­

istration in the Early Hellenistic Period.” Ancient Society 22: 203–227.

———. 1998. The Herakleopolite Nome: A Catalogue of the Toponyms with Introduction
and Commentary. American Studies in Papyrology 37. Atlanta: Scholars Press.

———. 2000. “Sullʼorigine del P.L.Bat. XX Suppl. A (Progetto di dighe e canali per la doreà
di Apollonios).” In S. Russo, ed. Atti del V Convegno Nazionale di Egittologia e Papirolo­
gia, Firenze, 10–12 dicembre 1999, 115–121. Florence: Istituto papirologico G. Vitelli.

Page 17 of 22
Geography and Administration in Egypt (332 BCE–642 CE)

———. 2007. “Patterns of the Greek Settlement in Egypt during the Ptolemaic Period:
‘Old Settlers’ in the Herakleopolite Nome.” In B. Palme. ed., Akten des 23. Interna­
tionalen Kongresses für Papyrologie, Wien, 22.–28. Juli 2004, 207–214. Vienna: Verlag der
Österreichischen Akademie der Wissenschaften.

Fraser, P. M. 1972. Ptolemaic Alexandria. Oxford: Clarendon Press.

Gardiner, A. H., ed. 1948. The Wilbour Papyrus I–III (IV, Index, by R. O. Faulkner, 1952).
London: Oxford University Press.

Gascou, J. 1985. “Les grands domaines, la cité, et lʼétat en Égypte byzantine.” Travaux et
Mémoires 9: 1–90.

Gomaà, F., R. Hannig, and H. Pohlmann. 1991. Tübinger Atlas des Vorderen Orient. Vol. 3,
1, ägypten zur Zeit des neuen Reiches: Nordteil, Südteil. Wiesbaden: Reichert.

Gonis, N. 2004. “Another Look at Some Officials in Early Abbasid Egypt.” ZPE
(p. 538)

149: 189–195.

Grenfell, B. P., ed. 1896. Revenue Laws of Ptolemy Philadelphus. Edited from a Greek Pa­
pyrus in the Bodleian Library, with a Translation, Commentary, and Appendices. Oxford:
Clarendon Press.

———, A.S. Hunt, and J. G. Smyly. 1902. The Tebtunis Papyri I. London: Oxford University
Press.

Grohmann, A. 1959. Studien zur historischen Geographie und Verwaltung des frühmitte­
lalterlichen Ägypten. Wien: R. M. Rohrer.

Guéraud, O., and P. Jouguet. 1938. Un livre dʼécolier du IIIe siècle avant J.C. Cairo: Im­
primerie de lʼInstitut français dʼarchéologie orientale.

Haas, C. 1997. Alexandria in Late Antiquity: Topography and Social Conflict. Baltimore:
Johns Hopkins University Press.

Habicht, C. 1997. Athens from Alexander to Antony, trans. D. L. Schneider. Cambridge,

Mass.: Harvard University Press.

Hagedorn, D. 1985. “Zum Amt des dioiketes im römischen Aegypten.” Yale Classical Stud­
ies 28: 167–210.

Heinen, H. 1998. “Das spätantike Ägypten (284–646 n.Chr.).” In ägypten in Spätantik-

Christlicher Zeit: Einführung in die koptische Kultur, ed. M. Krause, 35–56. Wiesbaden:
Reichert.

———, W. Schlömer, and H. Pohlmann. 1989. Tübinger Atlas des Vorderen Orient. Vol. 5,
21. ägypten in hellenistisch-römischer Zeit. Wiesbaden: Reichert.

Helck, W. 1974. Die altägyptische Gaue (TAVO Beiheft B15). Wiesbaden: Reichert.

Page 18 of 22
Geography and Administration in Egypt (332 BCE–642 CE)

Héral, S. 1990. “Deux équivalents démotiques du titre de nomarches.” CdÉ 65: 304–320.

Hombert, M. 1925. “Quelques papyrus des collections de Gand et de Paris.” Revue Belge
de Philologie et dʼHistoire 4: 660–662.

Kees, H. 1961. Ancient Egypt: A Cultural Topography, ed. T. G. H. James; trans. I. F. D.

Morrow. Chicago: University of Chicago Press.

Kramer, B., ed. 1991. Corpus Papyrorum Raineri. Vol. 18. Griechische Texte, 13, Das Ver­
tragsregister von Theogenis (P.Vindob. G 40618). Vienna: Brüder Hollinek.

Kruse, Th. 2002. Der königliche Schreiber und die Gauverwaltung: Untersuchungen zur
Verwaltungsgeschichte Ägyptens in der Zeit von Augustus bis Philippus Arabs (30 v. Chr.
—245 n. Chr.). 2 vols. Munich: K. G. Saur.

Lallemand, J. 1964. lʼadministration civile de lʼÉgypte de lʼavènement de Dioclétien à la

création du diocèse (284–382). Brussels: Académie Royale de Belgique.

Lenger, M.-Th. 1980. Corpus des Ordonnances des Ptolémées (C. Ord. Ptol.). Brussels:
Académie Royale de Belgique.

———. 1990. Corpus des Ordonnances des Ptolémées: Bilan des additions et corrections
(1964–1988). Compléments à la bibliographie. Brussels: Fondation égyptologique Reine
Elisabeth.

Liebeschuetz, J. H. W. G. 1990. From Diocletian to the Arab Conquest: Change in the Late
Roman Empire. Aldershot: Ashgate.

Mahaffy, J. P. 1896. “Introduction to the Revenue Papyrus.” In B. P. Grenfell, ed., Revenue

Laws of Ptolemy Philadelphus. Edited from a Greek Papyrus in the Bodleian Library, with
a Translation, Commentary, and Appendices, Oxford: Clarendon Press: xvii–lv.

Mazza, R. 1995. “Ricerche sul pagarca nellʼEgitto tardoantico e bizantino.” Aegyptus 75:
169–242.

———. 2001. lʼarchivio degli Apioni. Terra, lavoro e proprietà senatoria nellʼEgitto tar­
doantico. Bari: Edipuglia.

Mooren, L. 1984. “On the Jurisdiction of the Nome Strategoi in Ptolemaic Egypt.”
(p. 539)

In Atti del XVII Congresso Internazionale di Papirologia. Vol. 3, 1217–1225. Naples: Cen­
tro Internazionale per lo Studio dei Papiri Ercolanesi.

Mueller, K. 2003a. “Mastering Matrices and Clusters: Locating Graeco-Roman Settle­

ments in the Meris of Herakleides (Fayyum/Egypt) by Monte-Carlo-Simulation.” APF 49:
218–254.

———. 2003b. “Places and Spaces in the Themistou Meris (Fayyum/Graeco-Roman

Egypt): Locating Settlements by Multidimensional Scaling of Papyri.” Ancient Society 33:
103–125.
Page 19 of 22
Geography and Administration in Egypt (332 BCE–642 CE)

Müller-Wollermann, R. 1992. “Demotische Termini zur Landesgliederung Ägyptens.” In

Life in a Multi-cultural Society: Egypt from Cambyses to Constantine and Beyond, ed. J.
H. Johnson, 243–247. Studies in Ancient Oriental Civilization 51. Chicago: Oriental Insti­
tute of the University of Chicago.

Oertel, F. 1917. Die Liturgie: Studien zur Ptolemäischen und Kaiserlichen Verwaltung Ae­
gyptens. Leipzig: Teubner.

Palme, B. 1989. Das Amt des ἀπαιτητής in Ägypten. Vienna: Hollinek.

Parsons, P. J. 1967. “Philippus Arabs and Egypt.” Journal of Roman Studies 57: 134–141.

Pestman, P. W. 1990. The New Papyrological Primer. Leiden: Brill.

Préaux, C. 1939. lʼéconomie royale des Lagides. Brussels: Fondation égyptologique Reine
Elisabeth.

Pruneti, P. 1981. I centri abitati dellʼOssirinchite: Repertorio toponomastico. Florence:

Gonnelli.

———. 1989. “Toparchie e pagi: Precisazioni topografiche relative al nòmo Ossirinchite.”

Aegyptus 69: 113–118.

Rathbone, D. W. 1990. “Villages, Land, and Population in Graeco-Roman Egypt.” In Pro­

ceedings of the Cambridge Philological Society 216, n.s., 36: 103–142.

Rouillard, G. 1928. lʼadministration civile de lʼÉgypte Byzantine, 2d ed. Paris: P. Geuthner.

Ruffini, G. R. 2007. “New Approaches to Oxyrhynchite Topography.” In Proceedings of the

XXIV International Congress of Papyrology, Helsinki 1–7.8.2004, ed. J. Frösén, T. Purola,
and E. Salmenkivi, 965–978. Helsinki: Finnish Society of Sciences and Letters.

Samuel, A. E. 1966. “The Internal Organization of the Nomarch's Bureau in the Third
Century B.C.” In Essays in Honor of C. Bradford Welles. American Studies in Papyrology
1. New Haven, Conn.: American Society of Papyrologists: 213–229.

Sethe, K., and J. Partsch. 1920. Demotische Urkunden zum Ägyptischen Bürgschaft­
srechte vorzüglich der Ptolemäerzeit. Leipzig: Teubner.

Sheridan, J. A., ed. 1998. Columbia Papyri IX: The Vestis Militaris Codex. American Stud­
ies in Papyrology 39. Atlanta: Scholars Press.

Sijpesteijn, P. J., and K. A. Worp. 1978. Zwei Landlisten aus dem Hermupolites
(P.Landlisten). Zutphen, the Netherlands: Terra.

Sottas, H. 1921. Papyrus démotiques de Lille. Paris: P. Geuthner.

Swarney, P. R. 1970. The Ptolemaic and Roman Idios Logos. American Studies in Papyrol­
ogy 8. Toronto: Hakkert.

Page 20 of 22
Geography and Administration in Egypt (332 BCE–642 CE)

Talbert, R. J. A., ed. 2000. Barrington Atlas of the Greek and Roman World. Princeton:
Princeton University Press.

Thomas, J. D. 1974. “The Disappearance of the Dekaprotoi in Egypt.” Bulletin of the Amer­
ican Society of Papyrologists 11: 60–68.

———. 1975. “The Introduction of Dekaprotoi and Comarchs into Egypt in the
(p. 540)

Third Century A.D.” Zeitschrift für Papyrologie und Epigraphik 19: 111–119.

———. 1982. The Epistrategos in Ptolemaic and Roman Egypt. Part 2, The Roman Epis­
trategos. Opladen: Westdeutscher Verlag.

———. 2001. “The Administration of Roman Egypt: A Survey of Recent Research and
Some Outstanding Problems.” In Atti del XXII Congresso Internazionale di Papirologia,
ed. I. Andorlini, G. Bastianini, M. Manfredi, and G. Menci, 1245–1254. Florence: Istituto
papirologico G. Vitelli.

Thompson, D. J. 1997a. “The Infrastructure of Splendour: Census and Taxes in Ptolemaic

Egypt.” In Hellenistic Constructs: Essays in Culture, History, and Historiography, ed. P.
Cartledge, P. Garnsey, and E. Gruen, 242–257. Berkeley: University of California Press.

———. 1997b. “Policing the Ptolemaic Countryside.” In Akten des 21. Internationalen Pa­
pyrologenkongresses, Berlin 1995, ed. B. Kramer, 961–966. Stuttgart: Teubner.

———. 1999a. “Irrigation and Drainage in the Early Ptolemaic Fayyum.” In Agriculture in
Egypt: From Pharaonic to Modern Times, ed. A. K. Bowman and E. Rogan, 107–122. Ox­
ford: Published for the British Academy by Oxford University Press.

———. 1999b. “New and Old in the Ptolemaic Fayyum.” In Agriculture in Egypt: From
Pharaonic to Modern Times, ed. A. K. Bowman and E. Rogan, 123–138. Oxford: Published
for the British Academy by Oxford University Press.

———. 2001. “Ethne, Taxes, and Administrative Geography in Early Ptolemaic Egypt.” In
Atti del XXII Congresso Internazionale di Papirologia, ed. I. Andorlini, G. Bastianini, M.
Manfredi, G. Menci, 1255–1263. Florence: Istituto papirologico G. Vitelli.

Turner, E. G. 1936. “Egypt and the Roman Empire: The Dekaprotoi.” JEA 22: 7–19.

Uebel, F. 1968. Die Kleruchen Ägyptens unter den ersten sechs Ptolemäern. Berlin:
Akademie Verlag.

———. 1964. “MONOΠΩΛI A ΦAKH Σ: Ein bisher unbezeugtes Handelsmonopol frühp­

tolemäischer Zeit in einem Jenaer Papyrus (P.Ien. inv. 900).” In J. Wolski, ed. Actes du Xe
Congrès International de Papyrologues, 165–181. Warsaw: Comité des Sciences de la cul­
ture antique, Académie polonaise des Sciences.

Verhoogt, A. 1998. Menches, Komogrammateus of Kerkeosiris: The Doings and Dealings

of a Village Scribe in the Late Ptolemaic Period (120–110 B.C.). Leiden: Brill.

Page 21 of 22
Geography and Administration in Egypt (332 BCE–642 CE)

———. 2005. Regaling Officials in Ptolemaic Egypt: A Dramatic Reading of Official Ac­
counts from the Menches Papers (P.L. Bat. 32). Leiden: Brill.

Vitelli, G., ed. 1906. Papiri fiorentini: documenti pubblici e privati dellʼetà romana e bizan­
tina. I( 1–105). Milano: Hoepli.

Yoyotte, J., and P. Charvet. 1997. Strabon: Le Voyage en Egypte. Un regard romain. Paris:
NiL Éditions.

Zucker, F. 1964. “Beobachtungen zu den permanenten Klerosnamen.” In Studien zur Pa­

pyrologie und antiken Wirtschaftsgeschichte Fr. Oertel zum achtzigsten Geburtstag
gewidmet, ed. H. Braunert, 101–106. Bonn: Rudolf Habelt Verlag.

Notes:

1. Many topics foundational to this chapter are not treated here. I refer the reader to
Egypt from Alexander to the Copts: An Archaeological and Historical Guide (Bagnall and
Rathbone 2004) for an introduction to what modern Egyptian landscape can still tell us
about the appearance of the country and its ways of life in the millennium that passed be­
tween Alexander's invasion (332 BCE) and the Arab conquest (642 CE). Bowman (1986)
offers a narrative synthesis of the same period. In a longer historical perspective, the Cul­
tural Atlas of Ancient Egypt (Baines and Málek 2000) is also very useful. More on the geo­
graphical side (in greater detail, too, and on a larger scale), the relevant maps of the Bar­
rington Atlas (Talbert 2000, maps 74–80) and the Tübinger Atlas des Vorderen Orients
(Heinen, Schlömer, and Pohlmann 1989;see also Gomaà, Hannig, and Pohlmann 1991),
besides the Dizionario dei nomi geografici (Calderini and Daris 1935–2007) provide essen­
tial reference tools. All of these are best used in combination with information available
on the Internet, particularly at the sites of the Duke Databank of Documentary Papyri and
the Heidelberger Gesamtverzeichnis der griechischen Papyrusurkunden Ägyptens.

Maria Rosaria Falivene

Maria Rosaria Falivene is Associate Professor of Papyrology, Faculty of Letters and

Philosophy, University of Rome Tor Vergata.

Page 22 of 22
Law in Graeco-Roman Egypt: Hellenization, Fusion, Romanization

Law in Graeco-Roman Egypt: Hellenization, Fusion, Ro­

manization
Uri Yiftach-Firanko
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Law, Greek and Roman Papyrology
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0023

Abstract and Keywords

The Ptolemaic kings of Egypt ruled a variety of ethnic groups that were diverse in lan­
guage, culture, religion, and legal practices. The main themes were tolerance and even
the protection of particular legal traditions. By the beginning of the Roman period,
changes were under way. The autonomous courts of law had by then ceased to exist. The
second century CE witnessed the abandonment of demotic script in legal documents and
the emergence of a new law, “the law of the Egyptian”, which was applied by the entire
population and consisted of Greek and Egyptian elements alike. In the late third century
BCE, agoranomeia were established in the throughout Egypt to allow the state to monitor
foreclosure on assets placed as security for debts. In the Roman empire, Roman citizens
in Egypt followed major elements of the Roman law of succession, family, and personal
status.

Keywords: Hellenization, Romanization, demotic, Ptolemaic kings, autonomous law courts, Roman empire, agora­
nomeia

Introduction
A wide variety of activity related to law, such as contracts, appeals, court proceedings,
and legislative acts, were frequently documented on papyri. In Graeco-Roman Egypt
these were recorded in a variety of languages over the millennium and a half from the
Saite period to the early centuries of Arab rule. From the sixth century bce to the first
century CE demotic was used. With the Macedonian occupation of 332 bce, Greek set in
and was still in use long after the Arab occupation of 641. With the Roman occupation of
30 bce Latin emerged for some purposes, and Coptic was used in legal documents from
the sixth century on, becoming dominant after the Arab conquest.

In the period stretching from Alexander to the Arab conquest, then, a person who was
documenting a legal activity could frequently choose from among more than one lan­

Page 1 of 20
Law in Graeco-Roman Egypt: Hellenization, Fusion, Romanization

guage. This choice of language was neither a mere technicality nor a (p. 542) matter of in­
difference, however. Very frequently, legal documents were authored by professional
scribes, who in their writing adhered to long-standing traditions: Greek scribes used for­
mulae imported from the Greek motherland by immigrants who had followed Alexander's
conquest, while their Egyptian counterparts employed local formulas that predated the
Greek occupation. The formulas used in either case were naturally different. Accordingly,
the choice of language affected the nature of the institution.

The selection of a language also carried cultural implications. One would most naturally
employ the legal institutions of one's ancestors and record them in the same language
they had used. If one chose differently, there must have been important reasons to do so.
In the early Roman period a will drawn up by a Roman citizen had to be written in Latin
and follow strict formulaic rules to be valid. Accordingly, the acquisition of Roman citizen­
ship entailed a change from Greek to Latin as the language of wills (Kaser 1971, 687n10;
Wolff 2002, 157–159). A language might also be changed for reasons of expediency: In the
Ptolemaic period a petition to a Greek official such as the king, would be more effective if
written in Greek than in Egyptian, so Greek petitions to the king are predominant from
the start (Depauw 1997, 137).

Neither rule nor expediency necessarily governed other legal activities, however. In the
Ptolemaic period the composition of legal documents in demotic was never banned. The
contents of these documents were also, under certain conditions, enforceable in a court
of law. Accordingly, in the Ptolemaic and much of the early Roman period, Greek and
Egyptian scribal traditions coexisted and were eventually used by parties of Greek and
Egyptian origin alike.

Still, in the course of the first century ce demotic disappeared from legal documents. This
was a development of far-reaching consequence, for the demise of the demotic document
went hand in hand with that of the formulaic tradition that found expression in its claus­
es. So far as the law of contract goes, in the following centuries Egypt became a Greek
land, and those living there almost exclusively used Greek types of contracts authored by
Greek scribes and following Greek formats. One of the objectives of this chapter is to dis­
cuss how and why this Hellenization took place. I maintain that it accompanied the rise of
the office of the Greek public notaries—most conspicuously the agoranomeion—in the
Egyptian chôra in the Ptolemaic and early Roman periods. Accordingly, I examine impor­
tant landmarks in the evolution of the agoranomeion in that time frame.

The recognition that the contractual practices underwent their crucial phase of Helleniza­
tion in the Wrst century ce highlights a striking paradox: With regard to the format of le­
gal documents, never in the three centuries that Egypt was ruled by a Greek (i.e., Mace­
donian) dynasty was the country as Greek as it was when it became a Roman province.
This recognition raises an important question: Did the Hellenization of Egypt advance at
the same pace in other areas of life and society? This question will not, of course, be
treated in this framework. Yet some positive (p. 543) results can be reached. Minutes of
court proceedings from the second century CE sometimes refer to a certain manual called

Page 2 of 20
Law in Graeco-Roman Egypt: Hellenization, Fusion, Romanization

the “Law of the Egyptians,” which served as a guideline in court proceedings that in­
volved peregrines (i.e., those who did not hold Roman citizenship). The precepts con­
tained in that guideline are of diverse origin—Greek and Egyptian alike—and were ap­
plied by and to the local subjects regardless of their ethnic origin. Thus, the contents of
this manual at least do not evince the triumph of the Greek over the Egyptian element in
the second century CE but instead reflect a fusion of the two into a new “code” of law that
was applied by and to everyone in the province.

In 30 BCE Egypt became a Roman province and remained under Roman rule for almost
seven hundred years. What effect did this Roman presence have on the legal landscape of
Egypt? In 212 the Antonine Constitution turned the provincial population into Roman citi­
zens. Formally, it subjected all its inhabitants to the precepts of Roman law. Yet did this
change in status also mean a profound change in the legal practices in Egypt? To answer
this I dedicate a short section to the question of Romanization. We now turn to the begin­
nings of the agoranomeia.

Hellenization
Security and Sale

The agoranomos emerges in the Greek papyri for the first time in the mid-third century
BCE. Some of the earliest attestations connect him with the foreclosure of securities for
loans.1 In loans, the creditor gives money and requires some security for its recovery, so
the debtor grants him temporary title (though not necessarily ownership) to some of his
assets until he repays the debt. What happens if the borrower does not return the money
on time? The Ptolemaic lawgiver shows much interest in regulating this issue. According
to a royal decree (diagramma), presumably from around 275 BCE, the creditor cannot
simply seize the asset (Wolff 2002, 51–52).

When the loan is granted, the debtor is expected to register the security with the agora­
nomos.2 The entry is to report the nature of the object, its location, the identity of the
creditor, the amount of money lent, and the date of expiry. If the debt is not settled on
time or treated in a new contract, the agoranomos moves on to the epikatabolê, an official
act that sets in motion the foreclosure procedure (Rupprecht 1995a, 426–428; 1997, 291–
302). In addition, it is quite plausible that the decree that regulated foreclosures and de­
fined the role of the agoranomos is also responsible for the very establishment of the ago­
ranomeia in the chôra in the third (p. 544) century BCE. The agoranomoi were established
in the chôra, then, in order to monitor foreclosures. Soon, however, they assumed new
tasks.

When the Greeks first arrived in Egypt, there were no organized archives for Greek docu­
ments, so the settlers had to secure their transactions by other means. In the third centu­
ry BCE, legal documents commonly assumed the form of a double document. Identical
versions of the contract were recorded on the upper and lower halves of the same pa­

Page 3 of 20
Law in Graeco-Roman Egypt: Hellenization, Fusion, Romanization

pyrus. The upper text was folded and then closed with the seals of the parties and the wit­
nesses who were present at the act, while the lower half remained visible. Any change in
the wording of the open part after the conclusion of the contract could be easily detected
by breaking open the upper text and reading its contents. The instrument would then be
deposited with the syngraphophylax, a private individual whose high social standing
would ensure its safekeeping (Wolff 1978, 57).

By the late third century things began changing. Copies of privately composed legal docu­
ments were assembled in local (perhaps official) collections (Rupprecht 1995b, 37–49).
Far more important for the present discussion is the aforementioned creation of the ago­
ranomeia. According to my hypothesis, the agoranomoi were originally established in the
chôra to monitor loans and securities. For that purpose they developed advanced bureau­
cratic skills. The general population was familiar with these skills and realized the poten­
tial advantages of registering contracts at this newly established public archive. Accord­
ingly, once an agoranomos was present, other (especially valuable) transactions were
recorded in his files as well. For the agoranomos, the registration meant a new source of
income (CPR XVIII, pp. 30–31), so he likely welcomed the new practice.

This practice is attested in two spheres in particular: the law of marriage and the law of
sale. A newly married bride frequently brought into her husband's house highly valuable
assets—very frequently her only ones. The husband was to manage these assets with care
in the course of the marriage and to return them to the wife or her family if the marriage
ended. It was in the wife's best interest to ensure that he did so. Accordingly, when the
agoranomeia emerge in the second half of the third century bce, a new clause appears in
marriage documents that obligates the husband to record the marital arrangements in
the new archive within a specified period after his wife requested that he do so (Yiftach-
Firanko 2003, 55–79).

An identical phenomenon is evident in the case of land sales. For the purchasers, a sale
was a precarious matter. They paid good money for an object that could later prove defec­
tive in substance or title. In consequence, sale contracts from the late third century BCE
contain an anticipatory clause similar to that found in contemporaneous marriage docu­
ments. The vendor is instructed to “give” (dotô) the purchaser a “deed of sale” (ônê)atan
agoranomeion within ten days of the latter's request.3 The purchaser's security needs
would be satisfied by the registration of the act at any nearby agoranomeion. Accordingly,
in SB XIV11376 (239 BCE, Tholthis)—our earliest evidence of the practice from the Egypt­
ian chôra4—the (p. 545) vendor is free to register the sale at any of the nearby agora­
nomeia of Oxyrhynchus and Herakleopolis.

At first, the registration was of no consequence for the legal position of the purchaser. All
the buyer gained was probably yet another means of proving the conclusion of the con­
tract and the act of payment, should his title later be challenged or prove defective. Still,
the agoranomoi also issued a special certificate that reported the registration.5 Since the
same papyrus would frequently contain a confirmation of the payment of the conveyance
tax (enkyklion),6 its issuance could well corroborate the assumption that the purchaser

Page 4 of 20
Law in Graeco-Roman Egypt: Hellenization, Fusion, Romanization

had procured title to the asset. Accordingly, when the state wished to secure the inalien­
ability of an asset, one way was to prohibit the agoranomoi from performing the registra­
tion (SB XVI 12519.5, II bce, Arsinoite).

In general, however, the state was interested in promoting the new practice. A complete
and well-organized registry of sales would allow it to take cognizance of real rights and to
increase its revenues from the conveyance tax. The registry would also enhance the legal
certainty, for it would give potential buyers some indication of the legal position of the
prospective object of sale. In order to achieve these goals, it was necessary first to stan­
dardize the registration. In SB XIV 11376 of 239 BCE, we recall, the vendor is at liberty to
record the sale at any nearby agoranomeion. For the parties' security requirements this
was quite sufficient. When necessary, they simply went back to the office where they had
registered the sale and extracted proof of its registration. Yet this liberty prevented the
state and members of the parties' community from keeping track of the act. For their
needs, registering all of the acts performed in a given area in one registry seemed imper­
ative. Accordingly, in P.Köln V 219—a memorandum issued by an unknown official in 209
or 192 BCE—the inhabitants of an unidentified village in the northeastern Arsinoite who
wish to register sales are instructed to appear before a certain Herakleides in Philadel­
phia so that certificates (chrêmatismoi) may be issued “as customary.”

The papyrus does not relate the position of Herakleides in Philadelphia or explain how he
was supposed to treat the acts of sale. He may have been a collector of the conveyance
tax, who was to take cognizance of sales before their registration by the agoranomos and
to ensure thereby that the tax would be paid after the sale was registered.7 Herakleides
could also have been an agoranomos or an official with similar capacities who was posted
in Philadelphia inter alia to make the registration procedure available to the inhabitants
of its surroundings. On either hypothesis, P.Köln V 219 illustrates the great interest of the
Ptolemaic state, in the late third or early second century BCE, in furthering and regulat­
ing the registration of sales in Egypt and in utilizing it as a means of taking cognizance of
land conveyances in the chôera.

Most of our mid- and late Ptolemaic evidence on the registration of sales stems from the
Pathyrite nome. The Pathyrite source material yields some sixty registration (p. 546) cer­
tificates issued between 134 and 88 bce. The text of the certificates and their physical
layout are exceedingly uniform. Just like P.Köln 219, this uniformity reveals a high degree
of standardization, possibly by the state, of the registration of sales. The Pathyrite source
material is important for yet another reason. Much of the source material relates to Egyp­
tians. For these Egyptians, the composition by a Greek notary of a Greek instrument with
its distinct Greek terminology meant a close, perhaps first encounter with Greek scribal
traditions and through them also with Greek law. Accordingly, the creation of the agora­
nomeia played a key role in the dissemination of Greek law in the Ptolemaic chôra.8 We
now turn to the anagraphê.

Page 5 of 20
Law in Graeco-Roman Egypt: Hellenization, Fusion, Romanization

The Anagraphê: Official List of Contracts

Written documentation of a contract played a key role in any legal dispute over its terms
(Wolff 1978, 146–154). The record reported the legal act itself and also conveyed the
terms on which it was made. For this purpose it usually applied routine clauses that went
back to the Greek or Egyptian legal traditions in accordance with the language in which
the document was issued. If the document was introduced in court, the judge had to be
familiar with these clauses in order to consider and interpret them properly. Accordingly,
one of the most important measures in the formation of the court system in early Ptole­
maic Egypt was to direct disputes among Greeks, with their Greek documents, to the
Greek court of the dikastêrion and those among Egyptians to the Egyptian board of the
laokritai (Wolff 1970, 37–53).

The existence of an interpretative tradition in the courts also required the maintenance of
some uniformity by the scribes; for the expedient administration of justice with regard to
leases, for example, the judge had to find a similar scheme and content in every lease
contract. Accordingly, BGU VI 1214 and P.Ryl. IV 572 exhibit a considerable effort by the
state, in the early and mid-Ptolemaic period, to create a limited body of Egyptian scribes
whose selection would be tightly controlled by the state. What is especially interesting for
our purpose is that, at least according to a sound restoration in P.Ryl. IV 572.35, the
laokritai were to be involved in the selection. In other words, the Egyptian judges were to
exercise control over the nomination of the very scribes whose product would later guide
them in their verdicts.

This system could be upheld only as long as the laokritai maintained their exclusivity in
dealing with and interpreting Egyptian documents. If they had ever possessed such exclu­
sivity, in the second century BCE it was largely lost.9 Cases brought by Greeks and Egyp­
tians alike were now freely judged by various state officials and judicial boards (i.e., the
chrêmatistai, who gradually emerged as a regular court of law in the early second centu­
ry BCE). In the new instances the judges were rarely Egyptians, and one wonders
whether they could have dealt with (p. 547) a legal document in demotic without transla­
tion (Wolff 1970, 81–85). One solution was to have the Egyptian documents translated by
the litigants themselves before submitting them to the court.10 But a translation done pri­
vately could be prejudiced; thus, the state needed to take action.

By the mid-second century the agoranomeia had become established throughout Egypt;
for roughly a century they had kept lists of documents, witnessed declarations by parties
to transactions, and issued official papers recording these statements. They were now
prepared for an additional task, perhaps the most important one. In the minutes of the le­
gal hearing P.Tor.Choiach. 12.4.13–15 = MChr 31 = UPZ II 162 (117 BCE, Thebes), a
prostagma is cited as having declared that Egyptian contracts that have not undergone
anagraphê are “without authority” (akyroi). Akyros implies not only invalidity but also—
especially in the context of P. Tor.Choach. 12—inadmissibility as evidence in a court of law
(Préaux 1965, 188–189). Accordingly, the prostagma states that an Egyptian document
that was not subject to an anagraphê is not admissible in court.

Page 6 of 20
Law in Graeco-Roman Egypt: Hellenization, Fusion, Romanization

The most direct Ptolemaic evidence of the anagraphê is provided by UPZ I p. 596 from
145 BCE. In this letter, Paniskos, a notary in the region of Peri Thêbas, reports to his su­
perior the details of a procedure that was introduced in the preceding year: When Egypt­
ian documents are introduced by an Egyptian temple scribe (monographos), a Greek ex­
tract is issued that summarizes the key elements of the transaction. Then follows the ana­
graphê—the registration of the transaction on a special list. Finally, a subscription is writ­
ten on the Egyptian document that confirms the act of registration (Wolff 1978, 40–44;
Pestman 1985a, 24–25; Rupprecht 1995b, 49–50).

Once the procedure was completed, the original document was returned to the parties,
who then, and only then, could submit it as evidence before a court of law. From the per­
spective of the language barriers, the new measure did not remove the earlier difficulties
entirely: All that a Greek-speaking judge now had was probably an Egyptian document
with a short note in Greek that the document had undergone anagraphê, with few partic­
ulars of the transaction noted. More details, however, were potentially available to the
judge from the Greek extract, whose creation is mentioned by UPZ I p. 596 as well. Per­
haps this potential availability was enough for the litigants to give a reliable verbal ac­
count of the contents of the document they had submitted. Or were the litigants expected
to submit the Greek extract as well? Either way, the prostagma placed any discussion of
Egyptian legal transactions by a non-Egyptian court on much firmer ground than before.

For the agoranomeia, the prostagma had two major consequences. First, in UPZ I, p. 596
the Greek extract of the Egyptian document was issued by the notary and his assistants
themselves. This necessitated, perhaps for the first time, the employment of a native
speaker who also had some experience with Egyptian legal schemes and terminology. The
same employee was also expected to be familiar with the Greek formulaic tradition and to
be able to find the appropriate Greek (p. 548) rendering of the Egyptian terms. In the
decades following the prostagma we find in Upper Egypt some “Egyptians” reaching the
zenith of the scribal career, the office of the agoranomos itself (Wolff 2002, 71n1; 95; Pest­
man 1978, 310).11 This was perhaps an outcome of the new requirement.

Second, in order to register demotic legal documents issued anywhere in Egypt, it was
necessary to dispatch state scribes to the Egyptian hinterlands to carry out the task. The
need became ever stronger when Greek double documents also became subject to the
anagraphê around 120. Later on, around the beginning of the Roman period, the pres­
ence of the scribes in every corner of the chôra played a key role in the widespread dis­
semination of the Greek legal document and the consequent Hellenization of the law of
contracts in Egypt (cf. below). In order to understand the change we need to return first
to the second century BCE.

The “Agoranomic” Document

Through the anagraphê of Greek and Egyptian documents, the agoranomoi became ex­
posed to a variety of documents and schemes that they now regularly read, excerpted,
and registered. It was merely a matter of time until they started composing entire legal

Page 7 of 20
Law in Graeco-Roman Egypt: Hellenization, Fusion, Romanization

documents on their own. At first, the agoranomoi issued primarily those that related to
their original scope of business (i.e., loans and securities). 12 By the late Ptolemaic period,
however, they were in some areas composing every type of legal document, all of which
they also stored in archives under their supervision. Earlier security measures were
therefore considered redundant: Agoranomic documents do not report the presence of
witnesses and only occasionally makes use of the double-document form (Wolff 1978,
83).13

Yet the appearance of the agoranomic document signifies more than just a shift from one
form of document to another. In Pathyris, where most of the late Ptolemaic agoranomic
documents stem from a milieu of Hellenized Egyptians, the change was probably not from
one Greek form of documentation to another but from an Egyptian to a Greek documenta­
tion method. Accordingly, the availability of an agoranomeion played a key role in the
gradual promotion of Greek law, perhaps at the expense of the Egyptian. 14

Still, the immediate consequences of the change were rather modest. In the late Ptolema­
ic period the agoranomeia left their mark in limited areas. Best attested is the activity of
the agoranomeion in Upper Egyptian Pathyris and Krokodilopolis, as well as in the city of
Hermopolis, as the papyri of the archive of Dionysios, son of Kephalas (P.Dion.), indicate:
Legal documents from that archive stemming from the metropolis itself were composed in
the agoranomeion (nos. 23–31). The archive of Dionysios also shows, however, the limits
of the new position; once we leave the metropolis the old formats—Greek and demotic
alike—resurface (P.Dion. 1–8, 9–22, 32–34).

It was the Romans who paved the way for the triumph of the Greek documenta­
(p. 549)

tion methods. Demotic legal documents were still drawn up in the early Roman period; in
some areas, such as family law, they were common enough to attract the attention of two
governors in the late first and early second century CE.15 In general, however, in the first
days of their presence in Egypt, the Romans introduced new measures that in the long
run brought about the demise of demotic as a language for legal documents (Depauw
2003, 89n157). If in the Ptolemaic period demotic documents were supposed to be regis­
tered in situ (cf. p. 547), the Roman documents authored by Egyptian synallag­
matographoi were to be registered in the archive of the Nanaion, the temple of Isis in
Alexandria (Wolff 1978, 51–52). Upon a plausible interpretation, the petition SB I 5232
(Soknopaiou Nesos, 15 ce)—the earliest account of the procedure—shows that a con­
veyance of real right was considered ineffective if it was documented by a demotic record
that had not been registered (Rupprecht 2003, 484, 488; Jördens 2005, 45–46). The new
complications generally made the composition of demotic documents unrewarding (Wolff
2002, 73).16

While demotic documentation methods became increasingly complex in the early Roman
period, their Greek counterparts steadily became more accessible to the public. In the
Ptolemaic era, the very few public notary offices that existed were perhaps located pri­
marily in the metropoleis. Still, in the late Ptolemaic period some officials were register­
ing privately composed documents in the villages (cf. p. 548). Their presence there pre­

Page 8 of 20
Law in Graeco-Roman Egypt: Hellenization, Fusion, Romanization

pared the ground for the establishment, at the beginning of the Roman period, of inde­
pendent, permanent notary offices in villages throughout the Egyptian hinterlands. In the
Arsinoite nome, village notaries (grapheia) became established in the more populated vil­
lages of Tebtynis, Theadelphia, and Karanis, but also in remote localities like Soknopaiou
Nesos (Wolff 1978, 18–19; Jördens 2005, 46–48).

The case of Soknopaiou Nesos is particularly interesting, for the community remained an
Egyptian priestly stronghold throughout the early Roman period (Hobson 1981, 402; Jör­
dens 2005, 44–45, 50, 56). Here, the creation of the grapheion made Greek documenta­
tion methods more accessible to the public than they ever were in the Ptolemaic period.
The grapheion scribes are known to have composed demotic documents as well. But in
the case of the demotic document they also had to add a Greek subscription (hypographê)
to the legal instrument in order to validate it. Consequently, by the end of the first centu­
ry demotic was abandoned in legal documents (Depauw 2003, 104–105).

The foregoing does not mean that, in the centuries to come, the Greek public notaries re­
mained the only places for the composition of legal documents. Quite the opposite is the
case. Whenever the trouble and expenses arising from the involvement of the public no­
tary were disproportionate to the importance of the transaction, this involvement was
avoided. This was the case, for example, with leases through much of the Roman period
(Yiftach-Firanko 2007). Yet even when the public notary was sidestepped, the old demotic
forms were not reinstated. In (p. 550) those cases the authors of the document preferred
Greek schemes that did not require the direct involvement of a public notary (Wolff 1978,
112–113, 117–119, 125). When in the late second century the grapheia were losing
ground, it was these alternative Greek schemes, not the demotic ones, that were increas­
ingly adopted for every type of transaction. As far as the law of contract goes, Egypt had
now become a Greek land. Yet did Egyptian law disappear completely? By no means. As
the next section explains, in some spheres of legal activity Egyptian practices were still
followed in the second century CE.

Fusion of Legal Traditions: The Case of “The

Law of the Egyptians”
The term “the law of the Egyptians” (ho tôn Aigyptiôn nomos) emerged in the Roman pe­
riod. It is mentioned in no more than six papyri, most of them from the second century
CE.17 Still, as a survey of the most instructive of these sources shows, the importance of
the term surpasses its modest attestation. In those sources, the law is reported to have al­
lowed fathers to dissolve their daughter's marriage and take away her dowry against her
will. The right, which is extensively discussed in the second-century CE petition P.Oxy. II
237, is recorded in literary sources from as early as fourth-century-BCE Athens (Harrison
1968, 30–32; Mélèze-Modrzejewski 1970, 332) and is generally assumed to be of Greek
origin. As indicated by another papyrus, the court proceedings SPP XX 4 (124 CE, Ptole­
mais Euergetis?), the “law of the Egyptians” also allowed fathers to revoke wills made by
their deceased sons and to inherit their property. The origin of this latter right is borne
Page 9 of 20
Law in Graeco-Roman Egypt: Hellenization, Fusion, Romanization

out by a third papyrus referring to the law—the collection of extracts of court proceed­
ings from early second-century P.Oxy. XLII 3015 (after 117 CE, Oxyrhynchus).

Greek and demotic papyri amply record the Egyptian practice of parents “selling” their
estate in advance to their children. After the sale the parents are allowed to use the as­
sets but not to dispose of them without the children's consent. The unity of the family as
an economic entity is thereby secured (Kreller 1919, 204–207; Pestman 1995, 80–81). The
extent of this right is discussed in the extracts of P.Oxy. XLII 3015, in which the judges
consider whether, after such a sale, the parents are still allowed to divide their property
by will among their children and prospective intestate heirs. For this purpose they con­
sult “the law of the Egyptians”.

This is the background of the father's right to the estate of his predeceased son as dis­
cussed in SPP XX 4. In order to maintain the family unity as an economic entity it is nec­
essary to limit not only the parents' right to dispose of their (p. 551) property but also that
of their children: They, too, have to leave their assets within their family of origin. In or­
der to meet this end, everything that predeceased children leave behind should devolve
first upon their father and then, after his death, his closest next of kin. The family fortune
will thus remain intact. Accordingly, while the rule regarding the daughter and her dowry
(as discussed in P.Oxy. II 237) is of Greek origin, that relating to the estate of the de­
ceased son reflects an Egyptian tradition. “The law of the Egyptians”, then, is a collection
of local practices of different ethnic backgrounds.

In fact, it is much more than this. P. Oxy. II 237 is one of the most important sources on
“the law of the Egyptians”. It is a petition submitted in 186 by Dionysia to the governor of
Egypt against her father, Chairemon, who tried to dissolve her marriage and take away
her dowry against her will. Dionysia claims that he did so unlawfully. To support her claim
she adduces four precedents, three of which are minutes of court proceedings in which a
Roman judge prevents a father from dissolving a marriage.

In these minutes the father's supposed authority is generally ignored: The wife alone
should decide the course of her own marriage. This is the case in a sentence passed in
128 by the governor, T. Flavius Titianus, and in a ruling by the epistrategos Paconius Felix
in 134. Yet, unlike Titianus, who simply ignores the law for its “inhumanity,” Paconius Fe­
lix first allows its contents to be read in court. The fact that the “law of the Egyptians”
could be read indicates that it was a written, tangible piece of evidence (Wolff 2002,
75n18). It was a collection of precepts of different origin—both Greek and Egyptian—that
were adhered to in early Roman Egypt. In the aforementioned instance it is read in a
court presided over by a Roman judge. This may well have been the reason for the cre­
ation of the law in the first place: to make provincial practices accessible to Roman
judges whenever they heard cases involving non-Roman litigants.

Still, the law is more than just a written account of provincial practices. The fourth prece­
dent in Dionysia's dossier is a response by a local legal expert, Ulpius Dionysodoros by
name, to an official who was hearing a case in which a father was calling for the dissolu­
tion of his daughter's marriage. Dionysodoros, who is asked whether the father's action is
Page 10 of 20
Law in Graeco-Roman Egypt: Hellenization, Fusion, Romanization

lawful, recalls that fathers are allowed to dissolve their daughters' marriage only if they
(i.e., the fathers) are united with their wives in an “unwritten marriage.” Precisely the
same restriction is made in SPP XX 4 regarding the father's right to his dead son's assets;
the father acknowledges in this papyrus that he is entitled to the estate only because he
(the father) was united with the son's mother in an “unwritten marriage.” The same mea­
sure, then—the unwritten nature of the father's marriage—is used to curtail both the
Greek authority over the daughter and the Egyptian authority over the son. This is hardly
coincidental. Whoever curtailed the two capacities did so when they were joined in the
same textual framework. This framework was, I suggest, “the law of the Egyptians”'. Ac­
cordingly, “the law of the Egyptians” was not just a list of rules. It was a legal (p. 552)
treatise that was subject to adjustments that were made regardless of the different “na­
tional” roots of the provisions.

In fact, we may witness one such adjustment in the making in the aforementioned re­
sponse of Ulpius Dionysodoros. A judge directs a question to the legal expert about a
father's authority to dissolve his daughter's marriage. The legal expert, consulting the
“law of the Egyptians”', learns that this power has already been curtailed whenever the
father is united with his wife in an unwritten marriage. Yet the expert goes further. He ap­
plies the curtailment to a circumstance not originally considered by the law—that of the
father giving the daughter in marriage—and then issues his response according to his
own new interpretation. It was perhaps through such responses that the law was devel­
oped and updated in the course of its existence. The response also indicates who was en­
trusted with that development—a close group of legal experts versed in local law, who
acted as legal advisors in courts presided over by Roman officials. Their position was sim­
ilar perhaps to that held, in the same period, by some Roman jurisprudents by virtue of
the ius respondendi.18

“The law of the Egyptians” was created, then, in order to make provincial practices acces­
sible to Roman judges. Yet once it was created, the public was not oblivious to its exis­
tence and did not hesitate to use its provisions regardless of the origin, as well as the eth­
nic, civic, and cultural affiliation, of the users. Accordingly, in the case of Dionysia, P.Oxy.
II 237 reports the practice of Greek and Egyptian institutions alike. That of the latter is
still more striking in view of her father's social standing: an ex-gymnasiarch of
Oxyrhynchus and thus a representative of the Greek municipal elite (Mélèze-Modrzejews­
ki 1970, 332). The precedents brought forward by Dionysia exhibit a mixture of Greek and
Egyptian names. In the hearing before Titianus, the names are Greek and Roman, while
in that before Paconius Felix they are markedly Egyptian. Finally, SPP XX 4 bears Greek
or Hellenized names, although the institution itself seems Egyptian.

In conclusion, the source material dealing with “the law of the Egyptians” is indeed
scarce. Yet what has come down to us reveals an institution of the highest importance for
the legal history of Egypt as a Roman province. The law was a manual composed for the
use of Roman judges. It was meant to give them some knowledge of non-Roman practices
in the province. As such, it recorded practices of both Greek and Egyptian origin. The law
was an object of ongoing study, interpretation, and development by a group of local legal

Page 11 of 20
Law in Graeco-Roman Egypt: Hellenization, Fusion, Romanization

experts who acted as consultants in provincial courts of law. The contents of the law were
employed by all non-Romans. It was thus perhaps the first code of law that was applicable
to all Egyptians regardless of their origin. Accordingly, just as the creation of a network of
Greek notary offices brought about a formulaic unification of Egypt under Greek aus­
pices, so did “the law of the Egyptians” bring about a unity of its own: a universal obser­
vance of precepts of various origin that were assembled in a manual prepared for use in
the provincial courts of law.19

(p. 553) Romanization

In 30 BCE Egypt became a part of the Roman Empire and remained under Roman rule un­
til the Arab occupation of 641. How did the Roman presence influence the legal land­
scape of Egypt in the course of these seven centuries? To what extent did it cause non-Ro­
mans to adopt the institutions of the ius civile?

In Rome, most enforceable claims had to be expressed in the language of the praetor's
edict, which utilized highly professional terminology. To employ these terms themselves,
litigants needed the help of Roman legal experts. They also had to cloak every suit in a
distinctly Roman mantle. This need to Romanize suits was a major impulse for Romaniza­
tion wherever actions were founded on an edict of a Roman magistrate, usually that of
the praetor urbanus or on similar edicts of a provincial governor. This was not the case in
Egypt, however. There, plaintiffs were at complete liberty to present their claims in court
in any form they wished. The terms did not have to be particularly Roman, so legal lan­
guage was not Romanized through the terminology of the suit. Nor was there room for
special experts in Roman law who would explain how to use that language. Certainly, Ro­
man officials could introduce Roman legal institutions and terms through their verdicts,
and they occasionally did. In general, however, no predilection was shown towards the ius
civile in comparison to other, non-Roman sources either with regard to peregrines or even
in relation to the cives Romani (Wolff 1966, 38).

In theory, Roman law could have penetrated Egypt through the propagation of Roman
schemes in legal documents. This is, we recall, the way Greek law came to the fore in the
late Ptolemaic and early Roman periods. This, however, has never been the case with
Latin. In most spheres of private law Latin documentation was at best marginal. Both Ro­
mans and non-Romans wrote their contracts in Greek and shaped them according to the
same “schemes” that existed and were developed by scribes long before the Roman con­
quest (Wolff 2002, 163–165).

There was, however, one exception: the law of persons and family and the law of succes­
sion in particular. At least until 235 CE, wills of Roman citizens had to be written in Latin
in order to be valid (ibid., 134). The rule was enforced in Egypt, and so, in first- and sec­
ond-century Egypt, wills of Roman citizens are written in Latin. Romans put into use oth­
er elements of the law of succession as well (e.g., the acquisition of inheritances through
the formal acts of the cretio and the agnitio possessionis bonorum), Roman legacy cate­
gories (legatum and fideicommissum), and the opening of wills before a public official. Ro­
Page 12 of 20
Law in Graeco-Roman Egypt: Hellenization, Fusion, Romanization

man citizens seem, then, to employ in Egypt, even if in an unrefined form, much of the
hereditary system of the ius civile and the ius honorarium.

The law of succession is also a sphere in which we can establish a reception of Roman le­
gal institutions by non-Romans in the early days of the empire. In 6 ce the lex Julia vicesi­
maria imposed a 5 percent tax on estates inherited by Roman citizens. (p. 554) To enforce
the collection of the new tax, all Roman wills were required to be opened at a special bu­
reau set up at the emperor's shrine, the Caesareum, or at the statio vicesimae heredi­
tatium. The opening of a will had to be attended by the witnesses who had been present
at its composition. Non-Romans were not subject to this inheritance tax, but they deliber­
ately adopted the new procedure for their own wills. In the second century CE Greek wills
were subject to an official opening as frequently as Roman wills (Yiftach-Firanko 2002,
160–165). The case of the wills brings out, then, an important point. Roman legal institu­
tions were rarely deployed by Romans in Egypt, but when they were, they influenced the
practices of non-Romans as well.

In the third century, most inhabitants of the imperium Romanum became Roman citizens
as an outcome of the Antonine Constitution of 212 CE. The constitution had one immedi­
ate effect. In Rome, one way of creating a contractual bond was through the act of stipu­
latio. The party who was to become the creditor asked the prospective debtor orally
whether he were willing to undertake an obligation. The obligation itself was created by a
positive answer and was enforceable in a Roman court through the actio ex stipulatu if
the stipulatio was about a fixed sum of money and through a condictio if it was not (Kaser
1971, 542). Since the prospective creditor could incorporate in his question any lawful
contents, the stipulatio was used for the creation of virtually any obligation. It could thus
be used after 212 to make obligations of a non-Roman nature undertaken by the new Ro­
mans actionable in a Roman court; a clause recording the stipulatio was simply incorpo­
rated in the Greek legal documents. Greek contracts were thus adapted to the Roman
contractual system.

Yet the insertion of the stipulatio clause into the Greek contracts did not really Romanize
them. Many transactions recorded on Greek papyri from Egypt were enforceable in a Ro­
man court even without the stipulatio. This was, for example, the case with sale and
lease; these were consensual contracts, and the consensus could easily be extracted from
the Greek contract as it stood. There were, however, some contracts for which the nudus
consensus was not enough, nor could the Greek contract be reinterpreted in any other
way that would make it enforceable according to Roman law (Pringsheim 1961, 251). The
Greek scribes of the early third century may not have known the difference and probably
did not care. They simply inserted the clause in every contract regardless of whether it
was really necessary.20 For them, if the insertion did not help, it certainly would not hurt.
It certainly spared them the need of engaging the subtleties of the Roman contractual
system. From this perspective, the appearance of the stipulatio clause may have checked,
rather than promoted the Romanization of Greek contracts in Egypt.21

Page 13 of 20
Law in Graeco-Roman Egypt: Hellenization, Fusion, Romanization

As in the earlier period, the law of family, status, and succession constitutes a different
story. In the third and fourth centuries the patria potestas, the Roman dotal system, mar­
riage, adoption, and manumission are mentioned directly or at least leave some traces in
documents composed by the new Romans (Wolff 2002, 137–148; Arjava 1998, 155–159).
In some cases, there is no real reception of Roman concepts. In 331 CE a constitution of
Constantine the Great inflicted harsh punishments on spouses who (p. 555) divorced for
any reason besides those listed in its text (C. Th. 3.16.1). The constitution was in all prob­
ability never enforced in Egypt, yet the news that the emperor in Constantinople did not
approve of divorces did not escape its inhabitants. And so, in fourth-century petitions re­
lating to domestic crises spouses accuse each other of an unlawful breakup. They never
dispute their partners' right to end the marriage nor ask that the sanctions of C.Th. 3.16.1
be imposed on them. This is merely a rhetorical bid for the benevolence of the addressed
official (Yiftach-Firanko 2001, 1338).

Other Roman regulations struck deeper roots in third-century Egypt. Greek and Roman
adult women alike were required to have a formal guardian present for certain legal acts.
Yet there was one difference: Unlike their Greek counterparts, Roman women were ex­
empt from this tutelage as soon as they gave birth to three children (ius trium liberorum).
Since this was the case with most women before the age of thirty, the tutelage became
unnecessary for most. After 212 the Greek and Roman systems of guardianship fused. As
a consequence, the ius liberorum affected all female inhabitants. The fact that it was en­
forced is clearly shown by a rapid decline of the tutelage in the third century CE (Arjava
1997, 29). In the case of tutelage and perhaps also in that of patria potestas, the Antonine
Constitution meant an actual introduction of a Roman legal institution. Even in these cas­
es, however, the introduced institution was adapted to the social mentality of the new Ro­
mans. Here, too, Romanization probably did not involve profound changes in Egypt's le­
gal landscape.

Conclusion
The Ptolemaic kings of Egypt ruled a variety of ethnic groups that were diverse in lan­
guage, culture, religion, and legal practices. In the Ptolemaic period no attempt was
made to fuse these different elements into one “nation”, nor was a single code of law ever
created that was binding on the entire population of Egypt. The main themes, rather,
were tolerance and even the protection of particular legal traditions. Among other things,
both the Greek immigrants and the Egyptian population were allowed to maintain their
own scribal traditions and to be judged in autonomous courts of law in accordance with
their ancestral laws.

By the beginning of the Roman period, changes were under way. The autonomous courts
of law had by then ceased to exist, and so had the earlier national codes on which they
based their sentences. The second century CE witnessed the abandonment of demotic
script in legal documents and the emergence of a new law, “the law of the Egyptians”,
which was applied by the entire population and consisted of Greek and Egyptian elements

Page 14 of 20
Law in Graeco-Roman Egypt: Hellenization, Fusion, Romanization

alike. The present chapter has sought to (p. 556) shed some light on that process by focus­
ing first on the developments that brought about the triumph of Greek as the exclusive
language of legal documents.

The Ptolemies never aimed at eliminating demotic as a language in legal documents, but
a process set in motion in the third century BCE by a Ptolemaic king eventually led to that
outcome. In the late third century bce, agoranomeia were established in the capitals of
the nomes throughout Egypt to allow the state to monitor foreclosure on assets placed as
security for debts. Soon, however, the agoranomoi assumed new tasks.

In the third century BCE, most Greek documents were composed privately, that is, with­
out the involvement of a public notary. The security provided by these documents was
sometimes limited. Consequently, once the agoranomeia were established, they were also
used for the registration of particularly valuable transactions: sale of real property and
delivery of dowries at marriage. Because the record was written in Greek, whenever the
parties were Egyptian, the act of registration was frequently an occasion on which they
became exposed to Greek language and schemes. The state soon regulated the registra­
tion procedure and utilized it for its own purposes. An early piece of evidence of state in­
tervention is provided by P.Köln V 219.

Soon the agoranomoi took on new tasks. In the second century, Egyptian documents were
presented as evidence to Greek judges, who needed a reliable Greek account of their con­
tents. The agoranomoi were charged with creating those accounts, as well as registering
the contracts in their files. The new procedure further exposed Egyptians to Greek
schemes used in the agoranomeion for the translation of their demotic acts. It also fur­
thered the employment in the agoranomeion of Egyptian scribes who could perform those
translations.

Around the same time, the agoranomoi also started recording legal documents of their
own, and the fact that the document was composed by a public notary accorded it en­
hanced security. The agoranomic document was also drawn up in Greek, a fact that facili­
tated its use as evidence before Greek judges. In Pathyris in Upper Egypt, the outcome
was a shift toward the agoranomic Greek instrument, especially in high-value transac­
tions, particularly sales. This was yet another impulse for deploying Greek rather than
Egyptian documentation methods.

A seemingly “anti-demotic” policy was manifested after the Roman conquest. The Romans
imposed the deposition of demotic documents in Alexandria. They also created grapheia
in the villages that made Greek instruments widely available in the nomes' hinterlands.
The grapheion notary could also compose demotic documents, yet the imposition of the
addition of a Greek hypographê to every demotic document made its composition unre­
warding and brought about its demise in the following generations.

What gave raise to the new policy? Did it derive from a more general anti-Egyptian atti­
tude on the part of the Romans? I think not. This is evident from the discussion of “the
law of the Egyptians.” This manual, the use of which in court is attested in the (p. 557)

Page 15 of 20
Law in Graeco-Roman Egypt: Hellenization, Fusion, Romanization

second century CE, consisted of Greek and Egyptian elements alike and related, as far as
we know, to family law and the law of succession. There is no indication that regulations
of Greek origin were treated more favorably by the law than Egyptian ones. In fact, even
in the field of written contracts, Roman policy was neither nationally nor culturally moti­
vated. For a Roman judge and administrator, having to deal with evidence in one foreign
language, Greek, was challenging enough. An additional language would have been too
much. Even so, demotic documents were never banned. Their composition was simply
made unrewarding until it was abandoned altogether.

Finally, we come to Romanization. Egypt was under Roman rule for nearly seven cen­
turies, but in the period discussed in this chapter the Roman presence did not influence
people's legal practices as Greek presence had. In Egypt, the language of the courts was
not particularly Roman, nor did Roman judges apply Roman legal concepts or institutions
more than non-Roman ones in formulating their rulings. The Latin language and formulae
were also not very common in legal documents. In general, Romans and non-Romans in
Egypt issued the same Greek legal documents, applied local institutions and mechanisms,
and usually administered justice according to the same non-Roman guidelines. Yet Roman
law did leave its marks in certain areas: As elsewhere in the Roman empire, Roman citi­
zens in Egypt followed major elements of the Roman law of succession, family, and per­
sonal status. In these areas, Roman concepts also influenced, probably as early as the
first century CE, the practices of non-Romans.

The picture was not altered considerably even after 212, when the Antonine Constitution
granted most inhabitants of Egypt the status of Roman citizen. Roman legal institutions
continued to be more commonly applied in the law of succession, family, and status. In
other areas such as contract law, the adjustment of local practices to Roman concepts
was superficial and crude. The stream of Roman concepts into Egypt may have picked up
the pace in the fifth century, a period that is beyond the scope of this discussion.

Bibliography
Arjava, A. 1997. “The Guardianship of Women in Roman Egypt.” Akten des 21. Interna­
tionalen Papyrologenkongresses, Berlin 1995. APF (Beiheft 3): 25–30.

——— 1998. “Patria Potestas in Late Antiquity.” JRS 88: 147–165.

Depauw, M. (1997). A Companion to Demotic Studies. Brussels: Fondation Égyptologique

Reine Elisabeth.

———. 2003. “Autograph Confirmation in Demotic Private Contracts.” CdÉ 78: 66–111.

Harrison, A. R. W. (1968). The Law of Athens, vol. 1. Oxford: Clarendon.

Hobson, D. 1981. “Greeks and Romans at Socnopaiou Nesos.” In Proceedings of

(p. 559)

the Sixteenth International Congress of Papyrology, ed. R. S. Bagnall, G. M. Browne, A. E.

Hanson, L. Koenen, 389–403. Chico, Calif.: Scholars Press.

Page 16 of 20
Law in Graeco-Roman Egypt: Hellenization, Fusion, Romanization

Jördens, A. 2005. “Griechische Papyri in Soknopaiu Nesos.” In Tebtynis und Soknopaiu

Nesos: Leben im römischen Fajum: Akten des Internationalen Symposions vom 11. bis 13.
Dezember 2003 in Sommerhausen bei Würzburg, ed. S. Lippert and M. Schentuleit, 41–
56. Wiesbaden: Harrassowitz.

Kaser, M. (1971). Das römische Privatrecht, 2d ed., vol. 1. Munich: C. H. Beck.

Kreller, H. (1919). Erbrechtliche Untersuchungen auf Grund der gräko-ägyptischen Pa­

pyrusurkunden. Leipzig: Teubner.

Mélèze-Modrzejewski. J. 1964. “Réflexions sur ‘le droit ptolémaïque.’” Iura 15: 32–56.

———. 1970. “La règle de droit dans lʼÉgypte romaine.” In Proceedings of the Twelfth In­
ternational Congress of Papyrology, ed. D. H. Samuel, 317–377. Toronto: Hakkert.

———. 1975. “Chrématistes et laocrites.” In Le monde grec: Pensée, littérature, histoire,

documents: Hommages à Claire Préaux, ed. J. Bingen, G. Cambier, and G. Nachtergael,
699–708. Brussels: Editions de lʼUniversité de Bruxelles.

Muhs, B. 2005. “The Grapheion and the Disappearance of Demotic Contracts in Early Ro­
man Tebtynis and Soknopaiou Nesos.” In Tebtynis und Soknopaiu Nesos: Leben im römis­
chen Fajum: Akten des Internationalen Symposions vom 11.bis 13.Dezember 2003 in
Sommerhausen bei Würzburg, ed. S. Lippert and M. Schentuleit, 93–104. Wiesbaden:
Harrassowitz.

Pestman, P. W. 1978. “Lʼagoranomie: Un avant-poste de lʼadministration grecque enlevé

par les Égyptiens?” In Das ptolemäische Ägypten: Akten d. internat. Symposions, 27.–29.
September 1976 in Berlin, ed. H. Maehler and V. M. Strocka, 203–210. Mainz am Rhein:
von Zabern.

———. 1981. “Nahomsesis, una donna dʼaffari di Pathyris: Lʼarchivio bilingue di Pelaias,
figlio di Eunus.” In Scritti in onore di Orsolina Montevecchi, ed. E. Bresciani, G. Geraci, S.
Pernigotti, and G. Susini, 295–315. Bologna: CLUEB.

———. 1985a. “Registration of Demotic Contracts in Egypt: P.Par. 65; 2nd cent. B.C.” In
Satura Roberto Feenstra: Sexagesimum quintum annum aetatis complenti ab alumnis col­
legis amicis oblata, ed. J. A. Ankum, J. E. Spruit, and F. B. J. Wubbe, 17–25. Fribourg: Uni­
versity Press.

———. 1985b. “Ventes provisoires de biens pour sûreté de dettes: “Onai en pistei” à Pa­
thyris et à Krokodilopolis.” In Textes et études de papyrologie grecque, démotique et
copte, ed. P. W. Pestman, 45–59. Leiden: Brill.

———. 1995. “Appearance and Reality in Written Contracts: Evidence from Bilingual Fam­
ily Archives.” In Legal Documents of the Hellenistic World, ed. M. J. Geller and H.
Maehler, 79–87. London: Warburg Institute, University of London.

Page 17 of 20
Law in Graeco-Roman Egypt: Hellenization, Fusion, Romanization

Préaux, Cl. 1965. “La preuve à lʼépoque hellénistique, principalement dans lʼÉgypte
grecque.” La Preuve: Première Partie = Recueils de la Société Jean Bodin, vol. 16, 161–
222. Brussels: Editions de la Librairie encyclopédique.

Pringsheim, F. 1961. “Stipulationsklausel.” In Gesammelte Abhandlungen, vol. 2, 194–

256. Heidelberg: Universitätsverlag Winter.

Rupprecht, H.-A. 1995a. “Die dinglichen Sicherungsrechte nach der Praxis der Papyri:
Eine Übersicht über den urkundlichen Befund.” In Collatio iuris romani: Études dediées á
Hans Ankum à lʼoccasion de son 65e anniversaire, ed. R. Feenstra, A. S. Hartkamp, J. E.
Spruit, P. J. Sijpestein, and L. C. Winkel, vol. 2, 425–436. Amsterdam: J. C. Gieben.

———. 1995b. “Sechs-Zeugenurkunde und Registrierung.” Aegyptus 75: 37–53.

———. 1997. “Zwangsvollstreckung und Dingliche Sicherheiten in den Papyri der

(p. 560)

ptolemäischen und römischen Zeit.” In Symposion 1995, Vorträge zur griechischen und
hellenistischen Rechtsgeschichte (Korfu, 1.-5. September 1995), 291–302. Cologne: Böh­
lau Verlag.

———. 2003. “Die Streitigkeit zwischen Satabous und Nestnephis.” In Symposion 1999,
Vorträge zur griechischen und hellenistischen Rechtsgeschichte (Pazo de Mariñán, La
Coruña, 6.-9. September 1999), G, Thür, F. Javier (eds.), 481–492. Cologne: Böhlau Verlag.

Wolff, H.J. 1956. “Zur Romanisierung des Vertragsrechts der Papyri.” ZSav 73: 1–28.

———. 1966. “Organization der Rechtspflege und Rechtskontrolle der Verwaltung im

ptolemäisch-römischen Ägypten bis Diokletian.” TR 34: 1–40.

———. (1970). Das Justizwesen der Ptolemäer, 2d ed. Munich: Beck.

———. (1978). Das Recht der griechischen Papyri Ägyptens in der Zeit der Ptolemaeer
und des Prinzipats. Vol. 2, Organisation und Kontrolle des privaten Rechtsverkehrs. Mu­
nich: Beck.

———. 2002. Das Recht der griechischen Papyri Agyptens in der Zeit der Ptolemaeer und
des Prinzipats. Vol. 1:, Bedingungen und Triebkräfte der Rechtsentwicklung, ed. H.-A.
Rupprecht. Munich: Beck.

Yiftach-Firanko, U. 2001. “Was There a ‘Divorce Procedure’ among Greeks in Early Ro­
man Egypt?” In Atti del XXII Congresso Internazionale di Papirologia, Firenze 1998, ed. I.
Andorlini, G. Bastianini, M. Manfredi, and G. Menci, 1331–1339. Florence: Istituto Papiro­
logico G. Vitelli.

———. 2002. “Deeds of Last Will in Graeco-Roman Egypt: A Case Study in Regionalism.”
BASP 38: 149–164.

———. 2003. Marriage and Marital Arrangements: A History of the Greek Marriage Docu­
ment in Egypt, 4th Century BCE—4th Century CE. Munich: Beck.

Page 18 of 20
Law in Graeco-Roman Egypt: Hellenization, Fusion, Romanization

———. 2007. “The Rise of the Hypomnêma as a Lease Contract.” In Proceedings of the
24th International Congress of Papyrology, 1–7 August 2004 Helsinki, ed. J. Frösen, T.
Purola, and E. Salmenkivi, 1051–1062. Helsinki: Societas Scientiarum Fennica.

Notes:

(1.) P.Enteux. 61.11, epilysis; P.Enteux. 14.4 and P.Tebt. III.1 817 recto. 19–20, epikatabolê.

(2.) Cf., in particular, P.Enteux. 14, 15 (222 and 218 bce, Magdôla), 61 (246–240 BCE,
Ghoran), P.Köln V 219 (209/192 BCE, Arsinoites), P.Tebt. III.1 814 (227 BCE, Tebtynis) and
817 (182 BCE, Krokodilopolis), MChr. 233 recto (111 BCE, Pathyris).

(3.) Tholthis: BGU XIV 2398, 2399; SB XIV 11375 (all from 213–211 BCE); 11376 (239
BCE). Theogenis: CPR XVIII 25, 27 (both 232 BCE), and Wolff (1978, 195).

(4.) In the third century, Alexandrian landed property transactions were also registered in
a special archive. Cf. P.Hal. 1.242–259, BGU VI 1213.9 (both III BCE), and Wolff (1978,
188–190).

(5.) See P.Tebt. III.1 814, col. I (239 BCE, Tebtynis).

(6.) For example, BGU III 999 (99 BCE, Pathyris).

(7.) In most late-Ptolemaic transactions in Upper Egypt the actual payment of the tax fol­
lowed rather than preceded the registration. In BGU III 999, for example, the tax was
paid seven months after the registration. The opposite case is reported in P.Stras. II 81 +
82 (115 BCE, Diospolis Magna). Cf. also Pestman (1985b, 56–57), who discusses diverg­
ing practices in the case of provisory sale.

(8.) In P.Adler, from the archive of Horos, son of Nechoutes, nearly all twenty-one Greek
documents are agoranomic; fourteen are registration certificates.

(9.) This is shown in particular by the short-sighted and short-lived effort to reinstate the
authority of the laokritai: P.Tebt. I 5.207–220 of 118 BCE (e.g., Wolff 1970, 87, 204;
Mélèze-Modrzejewski 1975, 705–706).

(10.) Cf., for example, P.Giss. 37 (134 BCE, Pathyris).

(11.) Comp. Muhs (2005, 100–104), on Wrst-century CE Tebtynis and Soknopaiou Nesos.

(12.) Cf., for example, P.Mich. II 182 (182 BCE, Krokodilopolis, Arsinoites).

(13.) Witnesses are still reported in agoranomic wills. Cf. Kreller (1919, 315).

(14.) Particularly in the case of sale documents. Loans and especially marriage docu­
ments and wills are still largely demotic in this period. Cf., for example, Pestman (1981,
299–304).

Page 19 of 20
Law in Graeco-Roman Egypt: Hellenization, Fusion, Romanization

(15.) That is, M. Mettius Rufus (89–91 CE) and Ser. Sulpicius Similis (107–112 CE). Cf.
P.Mert. III 101; P.Oxy. II 237.8.21–27 (109 CE).

(16.) In the second century, the edict P.Oxy. I 34 verso col. iii = M.Chr 188 (127 CE) im­
posed additional requirements. Cf. also BGU V 1210.221–228 (after 149, Theadelphia).

(17.) See P.Oxy. II 237.7.29 (186 CE); IV 706.7 = M.Chr 81 (73 CE?); XLII 3015.3 (after
117 CE, all from Oxyrhynchus); P.Tebt. II 488.21–27 (after 121/122 CE, Tebtynis); SPP XX
4.16 = CPR 118 = M.Chr 84 (124 CE, Ptolemais Euergetis?).

(18.) Cf. Digest 1.2.2.49.

(19.) On the question of the fusion of the two legal traditions, cf. Mélèze-Modrzejewski
(1964).

(20.) Cf., for example, the insertion in petitions or wills in which no obligation was creat­
ed; in these, therefore, the clause was not required.

(21.) On the infiltration of Roman concepts into Greek contracts in the later Roman Em­
pire, cf. Wolff (1956, 20–27).

Uri Yiftach-Firanko

Uri Yiftach-Firanko is Assistant Professor of Classics, Hebrew University of

Jerusalem.

Page 20 of 20
Egyptian Religion and Magic in the Papyri

Egyptian Religion and Magic in the Papyri

Willy Clarysse
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Egyptology, Classical Religions and
Mythologies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0024

Abstract and Keywords

The first section of this article discusses traditional religion, looking at the Ancient Egypt­
ian worldview, mummification and afterlife, and the role of the temples in economy and
administration. The second section considers new developments in Egyptian religion such
as listening gods, animal cults, Egyptian “saints”, oracles, dreams, and katochê. The third
section describes the growing state intervention, examining the administration of tem­
ples, priestly privileges, temple asylum, and dynastic and imperial cults. The fourth sec­
tion looks at the impact of the Greek, describing interpreatio graeca, the Hellenization of
the gods, and astrology. The last section describes the end of Egyptian religion, looking at
polytheism, religion without temples, and Egyptian religion within Christianity.

Keywords: traditional religion, Egyptian religion, mummification, Hellenization, interpreatio graeca, state inter­
vention, Greek gods, polytheism, temple administration

Traditional Religion: Unity in Diversity

The Ancient Egyptian Worldview

Egyptian religion was based on a single worldview, which was realized in different ways
in different localities. The ordered world was created by the gods out of the chaos at the
beginning of time, but chaos remained present all around the world. It became visible in
the ocean surrounding the inhabited lands, in the foreign countries ever ready to invade
Egypt, in the desert on both sides of the fertile Nile valley, in death threatening all living
things, and in social upheaval and revolt. The cosmic order (M3c.t), reflected in the social
order, was maintained by the gods who had created it, with the help of mankind. The gods
had not completely withdrawn into the heavenly realm but remained available through
their cult statues in the temples.

Page 1 of 27
Egyptian Religion and Magic in the Papyri

The temple was the house of the god (ḥw.t-ntr), where the god lived with his fami­
(p. 562)

ly (wife and child) and was served by the priests as representatives of the pharaoh. The
often imposing building was a cosmos in miniature, with a star-studded roof supported by
huge stone papyrus columns. The sanctuary with the divine statue was carefully protect­
ed from the forces of chaos all around: only pure priests (wcb) were allowed to enter; on
the entrance towers (pylons), the pharaoh is represented killing foreign enemies, symbol­
izing the forces of chaos (figure 24.1), and inside the entrance gates armed guards stand
ready to destroy all demons who dare to enter the sacred realm. Just as Egypt is a cosmos
—an ordered space—compared to the barbarian countries, so the temple is the center of
order and cleanliness as opposed to the dirty streets of the surrounding town. In return
for the daily offerings that the pharaoh presents on the walls and his representatives, the
priests, lay before the shrine, the god keeps the whole world moving by making the sun
come up and the Nile rise at the right moment.

Figure 24.1. Pylon of the Edfu temple, showing King

Ptolemy XII smiting the enemies. Photo by W.
Clarysse.

Page 2 of 27
Egyptian Religion and Magic in the Papyri

Figure 24.2a–b. Two oracle questions, one positive

and one negative. Photo courtesy of the Ashmolean
Museum, Oxford.

The Egyptian temple has been compared, with some reason, to our nuclear power sta­
tions: Cosmic powers, both dangerous and life giving, are hidden inside (p. 563) and need
to be kept under control. The statue is hidden in the temple's dark interior, and the com­
mon people are not allowed to pass beyond the sacred pylons. The statue of the god is
washed, dressed, and offered food and drink by the higher priests, who have to be per­
fectly clean (circumcised, washed, and shaven). Only on festive occasions does the god,
usually carried in a bark on a stretcher, make an appearance outside the temple. On such
occasions he can be consulted by the believers, who lay down oracle questions along the
processional street (figure 24.2a–b). Since the holy of holies is in fact close to the back
side of the temple, the rear wall is sometimes decorated with a cult relief, and people
come here to approach the god, whom they know to be present on the other side of the
wall. Unlike churches or mosques, Egyptian temples are not places where people gather
(p. 564) inside: Instead they watch at the front door and along the processional way (dro­

mos); perhaps on special occasions some were allowed in the first court.

Page 3 of 27
Egyptian Religion and Magic in the Papyri

Figure 24.3. Bust of Sarapis. Graeco-Roman Muse­

um, Alexandria (Bagnall and Rathbone, Egypt: From
Alexander to the Copts [2004], 34). Reproduced by
permission of the editors.

Side by side with the huge stone temples numerous mud-brick shrines were served by
priests of lesser rank, often for lesser gods, saints, and sacred animals. They were less
grand but more accessible than the major gods in their stone fortresses. In the Fayyum
village of Kerkeosiris (ca. 1,500 inhabitants), for instance, no fewer than thirteen such
shrines are attested, owning 6 percent of agricultural land (Crawford 1971, 86–90). Isis
priests (isionomoi, demotic in-wwy) (Depauw 1998) and ibis raisers (ibioboskoi) were a
typical feature of many villages. The two cults left an imprint on the landscape, as can be
seen from the many toponyms including the words isieion and ibion (demotic cḫy) (Vandor­
pe 1991).

(p. 565) Local Variants

Against this broadly uniform background, each region has its own gods and adds its own
accents. Thus, in Memphis the world is created by Ptah, in Elephantine by the ram-head­
ed Chnum, and in Hermopolis by Thoth. In Edfu the sun god Horus, with the head of a fal­
con, fights the forces of chaos by killing a hippopotamus or a crocodile, whereas in the
Fayyum the crocodile god Souchos is identified with the sun god and stands for the cos­
mic world order in the form of the yearly Nile flood. Whereas Amon in Thebes and Sara­
pis in Alexandria (figure 24.3) live with wife and child as a holy family (Amon, Mut, and
Chonsou; Sarapis, Isis, and Harpochrates), Thoth is a bachelor, and Horus of Edfu lives in
a long-distance relationship with Hathor, who once a year makes the trip of about fifty
kilometers by boat from her own temple in Dendera to that of Horus.

Page 4 of 27
Egyptian Religion and Magic in the Papyri

Mummification and Afterlife

Mummification, burial, and mortuary cults are linked with the cult of Osiris, the god who
was killed and resurrected (see Riggs 2005). The undertakers' business is the best-known
branch of the temple activities, partly because mortuary priests often kept their papers in
a tomb. Thus, several interesting family archives in both demotic and Greek have been
found in tombs. In principle, the mummification priests, called taricheutai (“picklers” in
Greek, ry- ḥb in demotic), who read from the sacred books during the funeral and those
who looked after the tomb (w3ḥ-mw, translated as choachytes, “water pourers,” in Greek)
—are specialists each in their own field, but in practice the three tasks were usually in
the hands of a single group of people. The funerary priests jealously guarded their territo­
ries, and in their title deeds tombs are treated as valuable assets on a par with houses
and land. Mummification was not only for Egyptians, but, as was already the case with
Joseph's father, Jacob, in Genesis, foreigners living in Egypt quickly adopted the custom.
This is evident in the lists of the deceased kept by the choachytes and in the Greek-style
mummy portraits of the Roman period.

Not only humans were mummified but also sacred animals, which were often ceremonial­
ly buried in huge subterranean catacombs. The theagoi (tˍ3y-ntr) had the honor of carry­
ing the sacred animals to their tombs (Dils 1995)(figure 24.4), and the government con­
tributed to the expenses with lavish gifts.

The Role of the Temples in Economy and Administration

Figure 24.4. Theagoi carrying the mummified croco­

dile (Theadelpheia; now in the Alexandria Museum);
plate from E. Breccia, Monuments de lʼEgypte gréco-
romaine, I.2, pl. 64 [1926]). Reproduced by permis­
sion of the Graeco-Roman Museum, Alexandria.

The temples were also important landowners, sometimes in several nomes (Evans 1961).
Grain, wine and oil were in theory produced for the cult of the gods, but the priests also
profited. The temple factories also produced oil, fine linen (the (p. 566) so-called byssus),
Page 5 of 27
Egyptian Religion and Magic in the Papyri

and papyrus. Temple notaries, another source of income, painted their elegant cursive hi­
eroglyphs (demotic) on large papyrus rolls with lots of blank space on all sides, whereas
their Greek colleagues, who had to buy their papyrus from outside, filled their much
smaller sheets with tiny, tightly packed letters. Native high culture was preserved in the
temples, where the priests learned to read and write hieroglyphic, hieratic, and demotic
scripts in the “house of life” and also practiced traditional sciences, such as mathematics,
astronomy and astrology, medicine, dream interpretation, architecture, and so forth. In a
recent survey (Ryholt 2005), it has been calculated that half of the Tebtynis temple li­
brary, with the remains of more than two hundred books, consisted of cultic texts, a quar­
ter were narratives and a quarter scientific works.

In the pharaonic period the temples also functioned as wheels in the royal administration.
They nominally owned large tracts of land, but in fact royal taxes on land were partly col­
lected for the king through the temples. There was no opposition between temple and
state: The temples were very much integrated into (p. 567) the royal administration. In the
Graeco-Roman period this system was gradually replaced by a separate civil administra­
tion, and this contributed to the slow decline of the Egyptian temples.

Figure 24.5. Semtheus identified with Greek Herak­

les. Photo courtesy of the Ashmolean Museum, Ox­
ford.

At the pylon or temple gate (rw.t-di-m3c.t) justice was rendered both by native priests
(laokritai) and by Greek officials. According to the hieroglyphic inscriptions, the god him­
self was available here for those who were seeking justice. This principle is illustrated by
the numerous demotic temple oaths on ostraca: When the accusing party could not prove
his case, the accused was allowed to deny the accusation by an oath (cnḫ) invoking the lo­
cal god. If the accused took the oath, the case was dismissed; otherwise he was consid­
ered guilty. In such cases there is no (p. 568) real distinction between priestly and divine
justice: The priestly judges used the god to render justice. In Krokodilopolis, the capital of

Page 6 of 27
Egyptian Religion and Magic in the Papyri

the Fayyum, this gate of justice is called “the northern gate” (p3 r3 mḥty), transliterated
in Greek as (P)remitieion. Not only the Egyptian priests but even the Greek strategos ren­
dered justice here, and the tribunal before the temple gate is even mentioned in some
acts of Christian martyrs under Diocletian. Again temple and civil authorities work hand
in hand (Quaegebeur 1993).

New Developments in Egyptian Religion

With the exception of the Theban temples in Karnak, Luxor, and the Theban west bank,
tourists usually admire temples that were constructed in Graeco-Roman times. They were
built according to the detailed guidelines provided by the “Book of the Temple” (Quack
2005), which is preserved in multiple copies, and at first sight it seems as if nothing has
changed since time immemorial. But changes did take place, most of them so gradually
that they escape attention, whereas some, like the growing state intervention and use of
Greek, are easier to grasp. These developments become apparent when documents of dif­
ferent periods are compared.

Listening Gods

Figure 24.6. Ground plan of the Edfu temple (Bag­

nall and Rathbone, Egypt: From Alexander to the
Copts [2004], 229). Reproduced by permission of the
editors.

Whereas the traditional gods in their temple were out of reach for the ordinary people,
popular forms of these very same gods were accessible at the back of the temple, at the
entrance gate, or in small local shrines. People consulted them in case of illness (healing
gods), when they had to make important decisions (oracle gods), or when a conflict had to
be settled (divine justice). These gods turned their benevolent face to the visitor, and
their ability to listen to the prayers was often expressed by multiple ears accompanying
the divine picture. These same qualities are expressed by their typical names, such as
Nfr-ḥr (Greek Nepheros), that is, Beautiful-of-face, or Ḏhwty-sḏm (Greek Thotsytmis), and
even Msḏr-sḏm (Greek Mestasytmis), that is, Thoth-who-listens and The-ear-that-listens.
The divine names return as personal names: Nepheros (fem. Tnepheros) is common in the
Fayyum, and in the same region Mestasytmis is Hellenized as Mysthas and Akousilaos
(from Greek ἀκούω, “to listen”) (see figure 24.7). Although this kind of personal piety

Page 7 of 27
Egyptian Religion and Magic in the Papyri

goes back to the New Kingdom, it becomes widespread in the Graeco-Roman period,
when the Egyptian listening gods (theoi epêkooi) become popular all over the Mediter­
ranean world. (p. 569)

Animal Cults

A typical development of the later periods is the animal cult. Sacred animals had existed
since at least the New Kingdom, but they rose in prominence in later times. They could be
individual animals like the Apis, Mneuis, or Buchis bulls in Memphis, Heliopolis, and Her­
monthis; the Hesis cow in Aphroditopolis (Atfih); or the twin crocodile gods in the
Fayyum. Often, however, a whole species was considered sacred. Thus, ibises were raised
on special farms (ibiônes), and it was forbidden to hunt the oxyrhynchus and lepidotus
fishes not only in Oxyrhynchus but all over Egypt as well (Heinen 1991). Appian (De ani­
malibus XII.7) describes the lion den in Leontopolis; the plural in the name of the city
(“city of the lions”) suggests that several sacred lions were kept in the local zoo.

Figure 24.7. Votive inscription found in Pisa, Italy.

Dedication by Flavius Pharnutianus Rufus about 100
CE. The name Pharnutianus shows his Egyptian
background. Three pairs of ears are surmounted by
crowns of Egyptian gods: the lotus of Harpochrates,
the kalathos of Sarapis, and the cow horns, sun disc,
and feathers of Isis (SEG XXVI 1143; there is also a
plate in M. Guarducci, Epigrafia greca, vol. 3, Rome
[1974], 69).

Some sacred animals were living individuals in which the god manifested himself in this
world. Bouchis is Montou, the sacred crocodile is Sobek (Souchos), and the Hesis cow is
Isis (as is stated by the priests in P.Zen. Pestman 50.5). Moreover, the same god can be
represented in a human form as Montou or Isis, or as a human with an animal head, in
the case of Souchos. In the Fayyum the local crocodile gods in the villages (e.g., So­
knopaios = Souchos, lord of the island; Soknebtynis = Souchos, lord of Tynis; Soknokon­
neus = Souchos, lord of Bakchias) were as many individualizations of the one god Sou­
chos. But the sacred fishes of Neith in Esna are called eidôla or “images” of the goddess
(Quaegebeur, Clarysse, and Van Maele 1985, 223), whereas Apis is the “herald” of Ptah,
not Ptah himself. Similarly, the ibis birds and baboons were images of Thoth rather than

Page 8 of 27
Egyptian Religion and Magic in the Papyri

Thoth himself. The relationship between sacred animal and god was therefore a complex
one. (p. 570)

After death sacred animals, like humans, were mummified and buried and received a fu­
nerary cult. The burial was a great and costly occasion, with a mourning period of seven­
ty days, as for humans. The expenses for myrrh and byssus needed for the burial of Apis
were partly subsidized by the government and partly paid for by contributions from the
temples all over the country (P.Lund. III 10 = SB 8750; 98 CE). The stone sarcophagus
with the Apis bull was then hauled in procession from the city to the Serapeum, up the
steep desert ridge, over a distance of three kilometers. Those who could attend the burial
proudly left hundreds of commemorative small stelae in the tombs. In Ombos, the main
local civil officials participated in the yearly burial of the ibises and hawks (O.Joachim),
whereas, according to the inscriptions from Hermonthis, Queen Cleopatra herself attend­
ed the burial of Buchis (Goldbrunner 2004, 112–113, 152–155).

Egyptian “Saints”

Divinized individuals similarly functioned as intermediaries between the distant gods and
the believers. They might be pharaohs of the past, like Amenemhat III of the Middle King­
dom (ca. 2000 BCE), who was worshipped in the Fayyum under the name Marres or Pra-
marres (“pharaoh Marres”) (Widmer 2002), or private individuals, such as Imouthes, the
minister of King Djoser (ca. 2600 BCE) and architect of the step pyramid in Saqqara, and
Amenothes, son of Hapu, the architect of King (p. 571) Amenophis III (ca. 1375 BCE), both
sages and healing gods. The cult of Amenothes in the famous temple of Deir el-Bahari in
Thebes lasted until the mid-fourth century CE. Greek and demotic ostraca tell of us of his
miraculous healings, and in the second century BCE the rules of his cult association were
written down on a Greek papyrus (Quaegebeur 1977). Most local “saints,” however, are
far less conspicuous: In the list of the deceased buried in the necropolis and cared for by
the choachytes, they are marked as “masters” (p3 ḥry in demotic; φρι in Greek) and
“praised” (hsy in demotic; Άσιης, Έσιης in Greek) and no doubt received special treat­
ment in death (ibid.).

Oracles

Oracles had also been known for many centuries in Egypt, but shortly before the Ptolema­
ic period a new type of oracle questions developed, by which the query was written down
on two chits of papyrus, usually one positive and one negative. The structure of these
questions remained unaltered for nearly a millennium, notwithstanding the changes in
language (from demotic to Greek to Coptic) and religion (from Egyptian gods to Graeco-
Egyptian gods to Christian saints and Christ himself). First comes an invocation of the de­
ity, then the actual question in the form of a conditional sentence, positive or negative,
then an imperative, asking for the decision of the divinity. A typical example is the demot­
ic P.Oxf.Griffith 1 (Bresciani 1975; figure 24.2a–b): “(1) The servant Tesenouphis, son of
Marres, says before his lord Soknopaios, the great god, and before Isis with the beautiful

Page 9 of 27
Egyptian Religion and Magic in the Papyri

throne: (2) If it is good for me to plow the shore of the lake this year, year 33, (3) may this
(piece of) writing be taken out for me.”

The believer took the answer home with him, whereas the other chit remained in the tem­
ple and was eventually thrown away. Hundreds of these chits, often still sealed and un­
opened, were recently found in a rubbish dump near the temple of Tebtynis (P.Zauzich
dem. 10, introduction). The popularity of this type of document is illustrated not only by
the numerous papyri found all over Egypt but also by an edict in which the prefect for­
bade oracle consultation during the visit of the Emperor Septimius Severus: “Let no one
pretend to know matters beyond human knowledge and profess (to know) the obscurity of
future things either by means of oracular responses, i.e., written documents allegedly is­
sued in the presence of the divinity, or by means of processions of statues or suchlike
charlatanry” (SB XIV 12144).

Dreams

The gods could also give advice through dreams. Dream books are attested since the New
Kingdom and occur also in demotic; they explain the meaning of dreams, (p. 572) taking
account of the character or social position of the dreamer, as was to happen also in the
later Greek dream book of Artemidoros. Often the believer spent the night in the temple
expecting the god to send a dream that would provide a cure for illness or trouble (Malin­
ine 1962). The most famous example of this phenomenon is the katochos Ptolemaios, who
lived in the great temple of Sarapis in Saqqara for fifteen years. Ptolemaios and his
younger brother Apollonios carefully listed their dreams day by day in Greek, in demotic,
and even in Egyptian by means of Greek characters. But at a certain point Apollonios lost
faith and wrote his brother a desperate letter that ended with this statement: “Never
again can I hold up my head in Trikomia for shame that we have given ourselves away
and been deluded, misled by the gods and trusting in dreams” (Select Papyri I 100). Oth­
ers were more successful, and the Egyptian priest Hor from the Delta city of Sebennytos
even managed to obtain a royal audience in Alexandria on August 29, 168, after he had
predicted the retreat of the Syrian king Antiochos and the salvation of Egypt. Hor wrote
his dreams in demotic on potsherds, which were found a few hundred metres from where
his contemporaries Ptolemaios and Apollonios lived.

Katochê

Ptolemaios was a katochos, that is, he was possessed by the god and could not leave the
temple precinct. He made a living by performing minor religious tasks, selling clothes and
even soup, and perhaps even begging. The phenomenon of the katochê is still imperfectly
understood, but it was certainly motivated by religion and not simply a form of temple
asylum. Ptolemaios was in the power of the god, who communicated with him through his
dreams. In these dreams he clearly longed to retrieve his liberty, but this apparently did
not happen, as his archive was found in Saqqara, where he lived in or near the shrine of
the Syrian goddess Astarte.

Page 10 of 27
Egyptian Religion and Magic in the Papyri

It cannot be proven that the katochoi were identical with those persons who dedicated
themselves to a god, promising to pay a small monthly sum in exchange for protection
against all kinds of demons and spirits (Clarysse 1988). Texts of this type have been found
in Tebtynis and Philadelphia and may be linked to the temple's medicinal practices. Some
Egyptian priests, especially those of Sachmet, were experts in traditional medicine, com­
bining detailed knowledge (demotic and Greek manuals were found in the temples) with
magic. Holy water was poured over statuettes of Horus, the master of the dangerous ani­
mals, and then drunk by the patients (figure 24.11); magical imprecations were sung in
combination with effective drugs to drive away the malignant spirits. In order to have ac­
cess to this alternative medicine, people had to pay a kind of insurance to the temple,
making themselves “slaves” of the god for the rest of their lives.

(p. 573) Growing State Intervention

The Administration of the Temples

The internal organization of the temples was in the hands of the priests. They did not
function throughout the year but were divided into at first four, then (from 237 BCE on)
five groups (s3.w in Egyptian, φύλαι in Greek), who alternated in their duties to the gods.
The leaders (sḥn.w, Greek ἡγούμενοι) of these groups were in charge of the temple and
yearly choose a director (mr-šn, which in Greek could be transliterated as λεσώνης or
translated as ἀρχιερεύς). In each temple the government installed a financial administra­
tor, called an epistates. Originally he came from outside the temple, but quite soon the
epistates also belonged to one of the priestly families. Priestly functions were usually
passed down from father to son, but the newly installed priests had not only to conform to
ritual prescriptions (they had, for instance, to be circumcised) but also to pay an entrance
fee to the government.

Egyptians who entered the royal administration usually came from the native upper class­
es linked with the priests. The same families provided not only the prophets, stolistai and
hiereis, who together constituted the higher clergy, but also the Egyptian notaries, schol­
ars, doctors, and even civil servants. In many cases local grandees combined priestly and
administrative titles and even military functions. It is often unclear whether these were
priestly families entering the service of the Ptolemies or government officials gaining ac­
cess to the temples. In the later period civil officials gradually took over some of the most
important temple functions (Gorre 2009).

An interesting feature of the Roman period, when the temples became more self-cen­
tered, is the opposition between the higher and lower clergy, which resulted in several
lawsuits in which pastophoroi (lower-rank priests) clashed with the hiereis (SB X 10564).
The Gnomon of the Idios Logos also forbids priests to hold other positions (BGUV 1210),
and government jobs were now clearly separated from temple functions.

Page 11 of 27
Egyptian Religion and Magic in the Papyri

The Temple in the Administration

As stated earlier, state and temple did not stand in opposition, but the temples were an in­
tegral part of the system. Since in pre-Ptolemaic times the reading and writing of the
complicated hieroglyphic script were a near monopoly of the temples, the priests were at
the same time the obvious organizers of the administration.

This ancient land-tenure regime under the administration of the local temple continued
for some time in the Ptolemaic period, especially in Upper Egypt. According to the Edfu
donation text, written in hieroglyphs on the outside wall of the temple, the temple owned
no less than 3640 hectares, three quarters of which were in the nome of Edfu, the rest in
other southern provinces. Within the Edfu (p. 574) nome about 20 percent of the land
seems to have been owned by the god Horus (Manning 2003, 77). But in the demotic pa­
pyri of the Hauswaldt archive (third century BCE), temple land was sold by individuals
and even auctioned off by the administration; thus, some temple land was in fact treated
as private property and was only nominally owned by the temple. There were therefore
two types of temple land, real temple land (sacred land), owned by the god and adminis­
tered by his priests, and private land that belonged to the god's estate (ḥtp-ntr). The sa­
cred land gradually came under state control and was confiscated in the Roman period.
The private land (in fact, land passed down in hereditary lease) yielded no rent, but a
small harvest tax had to be paid. Originally this tax was paid through the temple adminis­
tration, but from 223 BCE on and especially after the great revolt (206–186 BCE) the tax
went directly to the king. He partially compensated the temple for this loss of income by a
subvention (syntaxis), from which the offerings to the gods could be paid. Since the tax
was now levied by government officials and not by the temple administration, it was con­
tinually threatened with being used for other, more urgent secular purposes, notwith­
standing promises in the royal decrees to keep “the revenues of the temples in place.”
Only for a tiny part of these lands did the temples continue to levy the harvest tax them­
selves, under strict government control, for specific purposes. This limited system of fi­
nancial independence was finally abolished by the Emperor Hadrian in 123 CE (Vandorpe
2005).

In some places the temple continued to function as an intermediary between taxpayers

and state far into the Roman period. Thus, in Soknopaiou Nesos taxes on oil production
and weaving were irst paid to the temple, which then passed them on to the fisc (Lippert
and Schentuleit 2005b). Part of the produce was destined for internal use in the temple.

Priestly Privileges

In the Ptolemaic census lists, priests of all categories are listed separately. They make up
between 5 and 10 percent of the village population in a strongly Hellenized area such as
the Fayyum (Clarysse and Thompson 2006, 177–186). For most of these, the priesthood
was an honorary status rather than a full-time profession, as appears from some bilingual
surety contracts. Thus, a god bearer (ṯ3y ntr, Greek theagos) of Souchos (figure 24.4) was
a simple donkey driver (P.Lille dem. II 49) (Dils 1995, 155 n5), or an astronomer of Her­

Page 12 of 27
Egyptian Religion and Magic in the Papyri

ishef becomes a cobbler in the Greek abstract (P.Lille dem. II 51). Nonetheless, being
counted as a priest clearly involved fiscal privileges if only because this particular donkey
driver or cobbler was not on the list when donkey drivers or cobblers were called upon
for corvée works.

In the Roman period a limited number of priests were fully or partially exempted from
works on the dikes and from the heavy Roman poll tax, the laographia, which normally
amounted to at least 16 and even 32 drachmas a year, the equivalent of two to four
weeks' wages. For each temple a quota was fixed. In (p. 575) 112 CE, for instance, Tebty­
nis was allowed 40 priests and 40 pastophoroi, Bakchias 64 priests (BGU XIII 2215, 113/4
CE), and Soknopaiou Nesos had more than 100 priests.

Priestly functions were clearly coveted. They were usually passed down from father to
son but could also be sold off. When no heirs were available, the functions were sold by
auction by the idios logos for thousands of drachmas (one drachma was equivalent to a
worker's wage for one day) (P.Lund. III 10).

Temple Asylum

As sacred spaces, temples, like churches, might provide asylum for people who risked
punishment by officialdom. No doubt this was the case for Egyptian temples in earlier pe­
riods also. Some even believe that the katochoi (or at least some of them) were in fact
seeking asylum in the temple. In contracts of debt and surety, the debtor or person guar­
anteed often had to declare that he would not seek refuge in a temple.

The asylum inscriptions, in which the king guarantees the rights of the temple against en­
croachments by officials, are a typical phenomenon of the late Ptolemaic temples. Though
they have been interpreted as signs of the increasing power of the temples, it is far more
likely that they show that the temples needed royal help to guarantee their traditional
rights against local grandees (Bingen 1989, 26–31).

Dynastic and Imperial Cults

When the Ptolemies organized the cult of the royal family around the tomb of Alexander
the Great in Alexandria, the Egyptian temples took the hint and organized their own cults
for the living royal couple, for some queens, and for the Ptolemaic ancestors (figure 24.8).
This cult is a totally new feature in the temple decoration: Whereas the pharaoh normally
confronts the gods as the representative of humanity, here his own ancestors stand on the
side of the gods.

The cult is described in some detail in the priestly decrees, which were proclaimed dur­
ing yearly gatherings of representatives of the whole Egyptian priesthood in Alexandria
or Memphis. In the Canopus decree a fifth phyle is added to the traditional four, in honor
of the reigning kings, and in the Rosetta Stone new divine honors are added to those that
already exist. The elaborate crowns of the divinized queen Arsinoe or princess Philotera/
Berenike, described in those texts, are also pictured on the temple walls and the statues.

Page 13 of 27
Egyptian Religion and Magic in the Papyri

Here the priests as a group function more or less in the same way as the Greek poleis
elsewhere in the Hellenistic world, voting divine honors for the king, a procedure that is
definitely un-Egyptian. In pharaonic Egypt, the pharaoh did not pretend to be a god, and
honors were not voted by his subjects (Clarysse 1999).

With Augustus, the queens disappear from the temple walls, but so also do the scenes of
the royal cult. Though Augustus himself is called “son of the god,” the Roman pharaoh is
again a representative of humanity facing the gods and no more (p. 576) than that. The im­
perial cult took place in special temples (Sebasteion, Hadrianeion) in the metropoleis and
was Greek in style. It was part of the daily routine of local officials (see, e.g., BGU II 362).

Collaboration

The privileged position of the priests in the tax system, their integration in the Ptolemaic
administration, the royal subvention of the temples, and the elaboration of an Egyptian
royal cult all led to a close cooperation of the Egyptian priesthood with the Ptolemaic
regime. In their petitions the priests often insist on their task of bringing food and drink
offerings to the divine royal family. The trilingual decrees stress again and again the legit­
imacy of the Macedonian dynasty, and insurgents are not only enemies of the gods, impi­
ous (asebeis) and ignorant, but are also accused of profaning and demolishing temples
and divine statues. Probably these accusations were true, and many temples suffered dur­
ing the revolts: Valuable assets were to be found there, and the priests had compromised
themselves by collaborating with the foreign occupants (Véïsse 2004, 197–242).

Figure 24.8. Ptolemy IV making an offering to the

gods of Edfu, followed by Ptolemy III, II, and I with
their queens, as synnaoi theoi. Drawing from E.
Chassinat, Le temple dʼEdfou, vol. 1, 527, pl. XXXVIa,
Paris (1897).

On the other hand, part of the clergy may well have sympathized with the insurgents. The
rebel kings Haronnophris and Chaonnophris were titled “loved by Amonrasonther, loved
by Isis,” and one can hardly doubt that they were proclaimed in the temple of Thebes. In
163, shortly after the revolt of Dionysios Petosorapis, the police searched for arms in the
great Memphite Serapeum (UPZ I 5.10). And the prophetic literature, such as the potter's
oracle and the oracle of the lamb, apparently anticipated a savior and a new golden age,
which would bring an end to the “city on the sea” (Blasius 2002). Whereas the priestly
princes of Memphis fraternized with the Ptolemies and officially crowned them in Mem­

Page 14 of 27
Egyptian Religion and Magic in the Papyri

phis (Thompson 1988, 138–54), the lower clergy may have been more ambivalent in their
loyalty.

(p. 577) The Impact of Greek

Interpretatio Graeca

When Herodotus gave Greek names to the Egyptian gods, calling Amon Zeus, Osiris
Dionysos, and Neith Athena, this interpretatio graeca was just a way to explain to his
readers the nature of these gods by an often superficial equivalence. Neith, for instance,
was identified with Athena because of the martial characteristics of the two goddesses
and the close link between the cities of Athens and Sais. In the Roman period, however,
Neith is really represented as a Greek Athena, the red crown becomes a Greek helmet,
and the shield with two crossed arrows develops into a wheel of fortune (figure 24.9).
When Athena is represented on terracotta lanterns, the Greek goddess apparently inher­
its the celebrated lamp festival of Neith in Sais (Quaegebeur 1983, 318–321; Quaegebeur,
Clarysse, and Van Maele, 1985, 218–220). In a wonder story about Imhotep, the son of
Hephaistos (Ptah) (P.Oxy. XI 1381), the cult of the saint is rightly given a long Egyptian
history, one that reaches back to the Old Kingdom and the reign of Mycerinos. But at the
same time Imhotep is Asklepios, and the author of the story is miraculously healed
through a dream, as in the temples of the Greek god of medicine. The author wants
“every Greek tongue to tell the story of the god and every Greek man to honor Imouthes
the son of Ptah.” Greek and Egyptian god have become inextricably intertwined.

Figure 24.9. Egyptian Neith as Greek Athena, with

her son, the sphinx god Tutu. Vienna 5077; drawing
from J. Quaegebeur, Studia Hellenistica 27 (1983):
309. Reproduced by permission of Studia Hellenisti­
ca.

The interpretatio graeca of Egyptian deities often becomes apparent in personal names.
Many people in antiquity had double names, which they could use alternately, depending

Page 15 of 27
Egyptian Religion and Magic in the Papyri

on the circumstances. The most famous of these is perhaps Simon-Petros; Simon is his
Jewish name, while Petros (“the rock”) promotes him to the leading figure among the
apostles, the first pope. When the two names of a person belong to different languages,
one can be simply the translation of the other, as in Didymos (Thomas, “the twin”), or,
much later, Schwartzert (Melanchton, “black earth”). In Graeco-Roman Egypt, this kind of
double name often yields information about the Greek interpretation of Egyptian religion.
In their simplest form, double names can be simple translations, such as Hatres/Phatres
and Didymos “(the) twin”; Pbekis and Hierax, “the falcon”; Mersis and Pyrrhos, “the red
one”; Mysthes and Akousilaos, “(the god) who listens”; or Thaesis and Isidora, “the one
who belongs to Isis/the one given by Isis.” They can also convey the equivalence of Greek
and Egyptian gods, as in Dionysios and Petosorapis (Dionysos = Osorapis); Apollonia and
Senmonthis (Apollo = Montou); Paniskos and Psemminis (Pan = Min); or Imouthes and
Asklepiades (Imhotep = Asklepios). Sometimes theological speculation is involved, as in
Semtheus alias Herakleodoros (in terracottas the Egyptian child-god Sm3-t3wy or
Semtheus, a form of Harpochrates, is often depicted with a club, like Greek Herakles, fig.
24.5), or Herais alias Tiesris (p. 578) (Hera corresponds to Egyptian Mout, whose temple
has a characteristic moon-shaped sacred lake called išrw = esris in Egyptian).

Hellenization of the Gods

The Hellenization of the gods is evident not only in their names but can also be seen in
their dress and their mythology. Whereas on the temple walls the gods and the king are
represented in the traditional fashion with kilts, staffs, wigs, and crowns, they can also
take on a Greek face. The Sarapis cult may have functioned as a model here: Sarapis is
shown as a Greek Zeus or Hades (with the three-headed Kerberos at his feet), with a clas­
sical hairstyle and beard, a Greek coat, and Greek sandals. The only Egyptian element is
the kalathos on his head, a vase-shaped container with vegetable motifs, showing his fer­
tility function (see figure 24.3). A wooden board found at Tebtynis shows us a divine triad
that includes Soknebtynis and Min (figure 24.10). Soknebtynis is an enthroned Zeus-like
figure, while Min is a standing young man (with no beard). Soknebtynis can be identified
as the crocodile god because he holds a little crocodile in his lap; Min is recognizable be­
cause he holds his penis in his hand. They look like Greek gods, except for (p. 579) some
Egyptian attributes (crowns, scepters, crocodile). In the thousands of terra cotta statues
found in houses and tombs, Sarapis, Isis, and Harpochrates are usually shown in Greek
form. The child Harpochrates, nude or with a Greek cloak, often wears an Egyptian crown
or the lock of youth. When he wears a club he is identified with Greek Herakles, who, like
Horos, demonstrated his power as a baby by killing or chasing the snakes sent by hostile
gods (Seth and Hera, respectively). Here Greek and Egyptian myths are becoming inter­
twined. Harpochrates holds his finger to his mouth, the traditional Egyptian way of repre­
senting a child (figure 24.5). But this attitude was reinterpreted by later writers as a sym­
bol of silence and wisdom.

The Greek Dioskouroi, Kastor and Polydeukes, are well attested through personal names,
and their cult has often been considered a rare example of a purely Greek cult in Egypt.
An oracle question to the Dioskouroi, however, found in the Egyptian temple of the croco­
Page 16 of 27
Egyptian Religion and Magic in the Papyri

dile god Souchos in Bakchias, and the double name Dioskourides alias Psansnos, “the one
of the Dioskouroi, alias the two brothers” show that here, too, Greek and Egyptian cannot
be separated. The twin crocodiles, which are represented suckling the goddess Neith in
the traditional imagery, have become grecized as Kastor and Polydeukes, or the sons of
Zeus have been Egyptianized as crocodile gods (Quaegebeur 1983, 312–316).

The Oracles

The Hellenization of the gods can be seen quite clearly in the oracle questions mentioned
earlier (Valbelle and Husson 1998). The earliest demotic oracle questions are all ad­
dressed to Egyptian gods. The oldest example known to us comes from Memphis and
probably even precedes the arrival of Alexander. Most Ptolemaic chits come from the
Fayyum and are addressed to Sobek (Souchos). In Soknopaiou Nesos, Souchos is closely
linked (married) to the goddess Isis Nepherses, “Isis with the beautiful throne.” In other
parts of Egypt other gods come to the fore, e.g., the hippopotamus goddess Thoeris in
Oxyrhynchus or the ibis of Thoth in Hermopolis. The earliest of these papyri are written
in demotic, but already in the third century BCE many questions in Tebtynis are written in
Greek. Often the Greek text is written with a brush, not with a reed pen, which shows
that the writers, no doubt local priests, were Egyptians. They wrote in Greek because this
was the language of the better-educated classes, and the god was of course part of this
upper-class world.

Figure 24.10. Soknebtynis-Kronos (with a small

crocodile on his lap) and ithyphallic Min in Tebtynis.
From Egyptian Religion, the Last Thousand Years:
Studies Dedicated to the Memory of Jan Quaegebeur
(OLA 84, Leuven, 1998), 244. Reproduced by permis­
sion of the editor.

Page 17 of 27
Egyptian Religion and Magic in the Papyri

In the Roman period demotic disappears, and the Egyptian gods are now always ad­
dressed in Greek. Moreover, Soknopaios is now identified with Ammon, an oracle god
who enjoyed wide fame with the Greeks for many centuries (Alexander even went to visit
him in Siwa). In Oxyrhynchus, Sarapis reigns supreme, a Greek Dionysos-Hades grafted
upon the cult of Osiris-Apis. Sarapis is now (p. 580) identified with Zeus and Helios, the
Greek chief god, and the Greek sun god (and oracle god) Apollo. The syncretistic view al­
lows gods from different backgrounds to merge their functions and to become each oth­
ers' equivalents. Egyptian gods receive a Greek name and a Greek iconography and even
take over bits and pieces of Greek mythology, all the while retaining their Egyptian
essence.

Astrology in the Greek Fashion

The ceiling of one room in the temples of Esna (200 BCE), Dendera (60 BCE), and Shen­
hur (Roman period) is decorated with a representation of the sky, depicting the constella­
tions of the Greek zodiac, the planets, and the zodiacal signs (fishes, twins, lion, scales,
etc.). Images in traditional Egyptian style render a Greek astronomical view of the cos­
mos based on Babylonian astronomy, to which are added the thirty-six decans of Egyptian
origin. In the Roman period the zodiac (p. 581) sometimes replaced the sky goddess Nut
on the inner side of the lid of sarcophagi. Astronomy and astrology were widely practiced
by the learned Egyptian scribes, but it is only in the Ptolemaic period that the first astro­
nomical papyri are found in temple surroundings. Horoscopes in Demotic (38 BCE–171
CE) and in Greek (10 BCE–508 CE) register the positions of heavenly bodies in the zodiac
for the birthdate of individuals. Only exceptionally is this followed by a prediction, as in P.
Oxy. IV 804: “It contains dangers. Be careful for 40 days because of Ares.” The horo­
scopes found in the temple of Medinet Madi (Narmouthis) show that the priests were in­
volved in this enterprise (Jones 1999).

Greek Texts in and around Egyptian Temples

The walls of the Egyptian temples are covered with sacred hieroglyphs, but Greek texts
gradually appear at the gates and even inside the sacred enclosure of the temples. Thus,
the great trilingual decrees were, as they say, set up by the priests “in the most conspicu­
ous place in the temples of first, second, and third rank.” As early as the late Ptolemaic
period Greek stelae in front of the gates or around the temple indicated the area of asy­
lum.

In the late Ptolemaic and early Roman period, private individuals who contributed to the
building of gates, walls, and chapels inside the temple often advertised their benefactions
by way of Greek inscriptions. The most conspicuous example is perhaps Parthenios of
Coptos, of whom no fewer than twenty-five stelae have been found—in hieroglyphs, de­
motic, Greek, and a mixture of these (Farid 1988). Greek building inscriptions were even
engraved on the architectural elements of the temple, for example, over the doorway
(I.Fay. II 105, 107; second century BCE). In the late Ptolemaic period the hymns to Isis-
Thermouthis by Isidoros were inscribed on the vestibule pillars in the forecourt of the

Page 18 of 27
Egyptian Religion and Magic in the Papyri

Narmouthis temple (Davoli 1998, pl. 109; I.Métriques 175). The Greek verses in hexame­
ters and elegiac distychs celebrate Isis as a goddess for the entire world, but they also re­
mind the reader of the Middle Kingdom origin of the temple and identify Pharaoh Se­
soosis (Sesostris) with the divinized Poremanres (Amenemhat III). In 96 BCE
Herakleodoros, son of Sostratos, and his family dedicate the vestibule and the sphinxes in
front of it with two other Greek inscriptions on the same pillar, right above the hymns
(I.Fay. III 158–159). In Narmouthis as elsewhere, temple constructions are no longer
namelessly attributed to the reigning pharaoh; rather, sponsors want to see their names
advertised. This is a Greek way of self-promotion, and most of the inscriptions are also in
Greek.

In the Ptolemaic temple of Philae in the deep south, Greek dedicatory inscriptions by offi­
cials (especially the local garrison commander), dating mostly from the second century
BCE, decorated the wall leading from the quay to the temple, but the priests themselves
also publicized the privileges they received from the kings in Greek form. In the first cen­
tury BCE and in the Roman period, most of the (p. 582) inscriptions are private proskyne­
mata to Isis on the great pylon left by pilgrims, both in Greek and demotic (Dietze 1994).
Greek “acts of worship” are common in all temples; even Stotoetis, a cutter of hiero­
glyphs, scratched his name in Greek (I.Fay. III 168).

Pilgrims approaching the great Serapeum of Memphis and walking along the alley of
sphinxes would perhaps not notice the hundreds of Greek inscriptions scribbled on the
back of the sphinxes (Rogge-Harrauer 1999), but they could not miss the Greek chapel
with Corinthian capitals (the lychnaption) at the end of the dromos and two groups of
Greek statues, one including Dionysos, the other representing Greek poets and philoso­
phers, of whom Pindar, Protagoras, and Plato can be identified thanks to inscriptions on
their bases (Lauer and Picard 1955). Inside the temple wall the visitors would pass the
shrine of Astarte, in which the Greek katochos Ptolemaios kept his little archive, including
several works of Greek literature. Near the entrance of the temple they could also see the
colorful advertisement of a Cretan dream interpreter, written in Greek iambics under an
Apis bull. The nameless Cretan was accredited “by an order of the god” (Cairo stela
27567; Thompson 1988, pl. VII).

Papyri found inside the temple similarly show the increasing presence of Greek. The
priests of Tebtynis copied their literary works in demotic on the back of Greek administra­
tive documents. In their houses were found Greek documents, as well as fragments of
Homer and Euripides, medical and astronomical texts, and even philosophical works (van
Minnen 1998; Ryholt 2002). Two jars full of ostraca, found in a room inside the temple
wall of Narmouthis, are quite puzzling. Some are Greek, others demotic, but demotic
signs are found in the Greek texts and vice versa, which is odd because the two scripts go
in different directions. The texts were probably written in a school context; in this temple
school of the second century CE the two languages were inextricably intermixed (O.
Narm. I and II).

Page 19 of 27
Egyptian Religion and Magic in the Papyri

The End of Egyptian Religion

The One God and the Many

In traditional polytheism the major gods had clearly defined fields of activity, both geo­
graphical (Souchos was, for instance, worshipped in the Fayyum and in Kom Ombo, Ptah
in Memphis, Khnum in Elephantine and Esna, Amon in Thebes) and functional (Isis is the
mother goddess, Hathor specializes in sex, Sakhmet sends and cures illnesses; Thoth, the
scribe of the gods, is also the god of wisdom; Osiris (p. 583) takes care of the afterworld;
Horos is both the sun and the young king, etc.). But many gods functioned as creator
gods, many could merge with the sun-god Re, most goddesses were spouses and mothers,
and so on. In the Late Period, moreover, visiting gods could share a temple, as synnaoi
theoi, with the main deity and then assume some of his characteristics. In wisdom litera­
ture the abstract idea of “the gods” or even “God” coexisted with individual gods. Major
gods could take over the names and roles of minor gods, as Osiris did with Sokar in Mem­
phis or with Chenti-Amenti in Abydos.

This phenomenon became ever more prominent. Thus, in the litanies in honor of Isis,
nearly all female goddesses, Egyptian as well as Greek, become manifestations of a uni­
versal goddess, “Isis with the ten thousand names” (myrionymos). Similarly, Sarapis takes
over the roles of Osiris, Hades, Dionysos, Zeus-Ammon, and Apollo, and he is addressed
in the oracle questions as “Zeus Helios Sarapis.” In the terracottas the child-god Har­
pochrates not only wears the royal double crown but also bears the solar nimbus or the
club of Herakles (figure 24.5). All Egyptian gods are great, twice great, three times great
(trismegistos), and their universal power is expressed by all kinds of symbols: animal
heads, knives, staves, crowns, wings, snakes, and uraei.

Thus, new gods who had not played any important role in the traditional temple cult
could come to the fore. Harpochrates, who functions as a child of the major gods in the
temples, now becomes a “great god” in his own right, as is clear from the hundreds of
magical stone stelae in which he is portrayed trampling crocodiles and holding snakes
and scorpions in his hand. Water poured over the stela and collected in a small basin had
a healing power (figure 24.11) (Sternberg-El Hotabi 1999). Bes, who kept the evil powers
away from the mammisi, the god-king's birth chamber in the temple, now receives a cult
of his own. A typical example of such a new god, who gained access to a temple cult only
in the Roman period, is Tutu (figure 24.9): He is represented as a striding sphinx with a
serpent tail, turning his head to the viewer; animal heads in his neck and on his chest
(crocodile, ram, lion) show the god's demonic powers. Sometimes he is coupled with his
mother, Neith (Athena), and the wheel of fortune; he also controls fate. However, he is al­
so linked to Amon-Re and the solar cult. His popularity is attested by numerous votive ste­
lae that were set up in the temples and by the frequent personal names Totoes and
Tithoes, but only one temple dedicated to him is known, that at Kellis in the Dakhleh Oa­
sis (Kaper 2003).

Page 20 of 27
Egyptian Religion and Magic in the Papyri

Religion without Temples

The near absence of temples and cult personnel equates Tutu with the demons of whom
he is the master. Demons were widely feared and propitiated in Greco-Roman times. We
meet them in the demotic self-dedications, in magical papyri, and especially in onomas­
tics. This is illustrated by personal names such as Herieus, (p. 584) Hry.w, “the benevolent
(ones),” which can be translated in Greek as Euphron (Clarysse 1991); Panetbeus, Pa-n3-
ḏb3.w, “the one of the avenging gods,” or Panechates, Pa-n3-ḫty.w, “the one of the ḫty
demons.” When named after these dangerous demons, a child was placed under their pro­
tection. Because of the high infant mortality rate, children were considered to be under
constant threat of the “evil eye.” On mummy portraits they sometimes wear a collar with
an amulet in a small container, and in private letters their names are often preceded by
the word abaskantos, “may jealousy, i.e., the evil eye, not touch them” (Bonneau 1982).

Oracles, dreams, and astrology were originally linked to the temple (or rather to the fore­
court of the temple, the only place accessible to the general public). But in some in­
stances the link with the temple may have weakened: Thus, the Cretan dream interpreter
was not a priest of Sarapis, and the katochos Ptolemaios also seems to have functioned as
a private interpreter of his own dreams and those of his friends. Oracle questions could
be written by priestly scribes, but the believers themselves could also interpret the move­
ments of the sacred bark in the procession or of the holy Apis bull in his court. Thus, the
so-called oracle of Astrampsychus, which first appeared in the third century CE, was a do-
it-yourself oracle, containing a list of ninety-one questions and 910 answers. Originally
written in Greek, it was very popular through translations (and Christianization) in Latin
and most European languages (Stewart 2001).

The most prominent aspect of religion outside the temples is no doubt found in magical
papyri. Although magic (i.e., the scientific exploitation of the divine powers) existed for
centuries both in Egypt and in Greece, magical papyri are a typical feature of the late Ro­
man and Byzantine periods. On the one hand, there are magical handbooks, such as the
magician's library found at Thebes, which comprises both Greek and demotic papyri (Tait
1995); on the other hand, hundreds of texts on papyri, ostraca, and lead tablets contain
practical applications of the handbooks to individual situations: amulets against illnesses
and the evil eye, as well as love charms and spells to make a horse lose a race. In some
cases the written charm is coupled with a doll on which the magical rites had been per­
formed (du Bourguet 1980).

Page 21 of 27
Egyptian Religion and Magic in the Papyri

Egyptian Religion within Christianity

Figure 24.11. Horos cippus, Brussels, Musées Roy­

aux dʼArt et dʼHistoire, E. 4995. Photo copyrighted
by the IRPA-KIK, Brussels.

Christianity is distinguished from pagan religion by its ideology of aggressive monothe­

ism, its abolition of sacrifices, and its scientific approach based on a sacred book. As we
have seen, the strict polytheism of traditional Egyptian religion had already given way to
different forms of henotheism, by which different gods took over each other's roles to
such an extent that they tended to merge. Instead of absorbing the traditional gods,
Christianity demonized them. It did not actually deny their existence but turned them into
malicious demons. In addition, Decius (p. 585) (p. 586) (250 CE), who ordered every citizen
of the empire to sacrifice food and drink offerings (θύειν καί σπένδειν), had already per­
ceived that sacrifices made a real difference: Whereas ancient religious festivals implied
an abundant meal for the whole community, Christians offered the body and blood of
Christ only in a symbolic way. The poor were cared for in a different manner—Christian
charity. Egyptian religion did use written texts (e.g., the Book of the Dead, the Book of
Breathing, and the Book of the Temple), but there was not one fundamental charter of be­
lief like the Bible or the Qurʼān.

Not everything in the new religion, however, was new to the Egyptians: Pharaoh was the
son of god and of his earthly father, like Jesus; the creation of the world by the divine lo­
gos is also found in the theology of Ptah; the resurrection of the dead was foreshadowed
by the cult of Osiris; the Eucharistic meal was similar to the sacred meals in the temples
of Sarapis and Thoeris (Youtie 1948); the Holy Trinity could be compared with the divine
families in the temple; even the expectation of a messianic king was not unknown, as ap­
pears from the names of the rebel pharaohs Haronnophris and Chaonnophris (Véïsse
2004, 99, 247).

Page 22 of 27
Egyptian Religion and Magic in the Papyri

In one or two generations' time, between 300 and 360, Egypt shed its traditional religions
and went over massively to Christianity. In the first half of the fourth century the old and
the new religions coexisted for a while, as can be seen in Panopolis, in Philae, or in the
Khargeh Oasis. But the drastic change in onomastics from pagan names based on the
Egyptian gods to Christian names, mentioning “god” (Theo- and -noute names), abstract
virtues (e.g., Eusebios, Eulogios), biblical persons (e.g., Elizabeth, Johannes), and saints
(e.g., Petros and Paulos, Kosmas and Damianos) illustrates the pervasiveness of the trans­
formation (Bagnall 1993, 280–281). The last inscriptions recording the rise of the Nile in
a traditional religious context date from 295 CE, in 320 CE the son of a priest is still cir­
cumcised, the last Buchis bull is buried in the reign of Constantius (340 CE), the last
bloody sacrifice is attested in Deir el-Bahari in the second quarter of the fourth century,
and in 336 CE a priest of Zeus and Hera is still appointed in Oxyrynchus (ibid., 261–273).
According to Ammianus Marcellinus, the Emperor Julian was present at the enthrone­
ment of an Apis bull in 362 CE. In any case, by the end of the fourth century pagan reli­
gion had disappeared from public view, and Egypt had become a Christian country fa­
mous for its hermits and monasteries.

The old religion lived on in the new to a limited extent. Thus, the cult of the Virgin Mary
with the little child, Jesus, took over the iconography from the earlier Isis cult. Mary be­
came the new theotokos, “mother of the god.” Of course, she has her hair decently cov­
ered, in opposition to the sexy hairstyle of Isis. Christian knights like Georgios, Victor, or
Apa Claudius, sitting on horseback and piercing a demonic figure with their lance, take
over the role of Horos and Heron. Mummification continues for a while against the oppo­
sition of the ecclesiastical authorities. The rising of the Nile is now celebrated in the
church, which also takes over the age-old custom of delivering oracles. Magic and astrol­
ogy still flourish, and here the gods (p. 587) live on as demons that have to be manipulated
or overpowered. But these, like the survival of the old Egyptian ankh symbol for the life-
giving cross (Cramer 1955) are poor remnants of the old religion in a new world (Frank­
furter 1998).

Bibliography
Bagnall, R. S. (1993). Egypt in Late Antiquity. Princeton: Princeton University Press.

Bingen, J. 1989. “Normalité et spécificité de lʼépigraphie grecque et romaine de

lʼEgypte.” In Egitto e storia antica dallʼellenismo allʼetà araba: Bilancio di un confronto,
ed. L. Criscuolo and G. Geraci, 15–35. Bologna: CLUEB.

Blasius, A., ed. (2002). Apokalyptik und Ägypten: Eine kritische Analyse der relevanten
Texte aus dem griechisch-römischen Ägypten. Leuven: Peeters.

Bonneau, D. 1982. “Lʼapotropaïque ‘Abaskantos’ en Égypte.” RHR 199: 23–36.

Brashear, W. M. 1995. “The Greek Magical Papyri: An Introduction and Survey. Annotated
Bibliography 1928–1994.” In Aufstieg und Niedergang der Römischen Welt: Geschichte

Page 23 of 27
Egyptian Religion and Magic in the Papyri

und Kultur Roms im Spiegel der neueren Forschung, vol. 2, 18.5, 3380–3684. Berlin: de
Gruyter.

Bresciani, E. (1975). Lʼarchivio demotico del tempio di Soknopaiu Nesos nel Griffith Insti­
tute di Oxford, I. Testi e documenti per lo studio dellʼAntichità. Milan: Istituto editoriale
Cisalpino.

Clarysse, W. 1988. “A Demotic Self-dedication to Anubis.” Enchoria 16: 7–10.

———. 1991. “Hakoris, an Egyptian Nobleman and His Family.” Ancient Society 22: 235–
243.

———. 1999. “Ptolémées et temples.” In Le décret de Memphis, ed. D. Valbelle and J.

Leclant, 41–62. Paris: Fondation Singer-Polignac.

Clarysse, W., and D. J. Thompson. (2006). Counting the People in Ptolemaic Egypt. Cam­
bridge: Cambridge University Press.

Clarysse, W., and H. Willems, eds. (1998). Egyptian Religion: The Last Thousand Years,
OLA 84–85. Leuven: Peeters.

Cramer, M. (1955). Das altägyptische Lebenszeichen im christlichen (koptischen) Ae­

gypten. Wiesbaden: Harrassowitz.

Crawford, D. J. (1971). Kerkeosiris: An Egyptian Village in the Ptolemaic Period. Cam­

bridge: Cambridge University Press.

Davoli, P. (1998). Lʼarcheologia urbana nel Fayyum di età ellenistica e romana. Naples:
Procaccini.

Depauw, M. 1998. “The isionomos or in-wwy” In Egyptian Religion: The Last Thousand
Years, ed. W. Clarysse and H. Willems, 1131–1153. Leuven: Peeters.

Dietze, G. 1994. “Philae und die Dodekaschoinos in ptolemäischer Zeit.” Ancient Society
25: 63–110.

Dils, P. 1995. “Les t3j (n3) ntr.w ou θεαγοί. Fonction religieuse et place dans la vie civile.”
BIFAO 95: 153–171.

Du Bourguet, P. 1980. “Une ancêtre des figurines dʼenvoûtement percées dʼaiguilles, avec
ses compléments magiques, au Musée du Louvre.” Livre du Centenaire de lʼI.F.A.O. 1880–
1980, 225–238. Cairo: Institut Français dʼArchéologie Orientale.

Dunand, F., and F. Zivie-Coche. (1991). Dieux et hommes en Egypte, 3000 av. J.-C.-395 apr.
J.-C.: Anthropologie religieuse. Paris: Armand Colin.

Evans, J. A. S. 1961. “A Social and Economic History of an Egyptian Temple in the

(p. 588)

Greco-Roman Period.” Yale Classical Studies 17: 143–283.

Page 24 of 27
Egyptian Religion and Magic in the Papyri

Farid, A. 1988. “Die Denkmäler des Parthenios, des Verwalters der Isis von Koptos.”
MDAIK 44: 13–65.

Frankfurter, D. (1998). Religion in Roman Egypt: Assimilation and Resistance. Princeton:

Princeton University Press.

Goldbrunner, L. (2004). Buchis: Untersuchung zur Theologie des heiligen Stieres in

Theben zur griechisch-römischer Zeit. Monographies Reine Elisabeth 11. Brussels: Fon­
dation Egyptologique Reine Elisabeth.

Gorre, G. (2009). Les relations du clergé égyptien et des Lagides dʼaprès la documenta­
tion privée. Stud. Hell. 45. Leuven: Peeters.

Heinen, H. 1991. “Thoeris und heilige Fische: Eine neue griechische Inschrift für Ptole­
maios X. Alexander I.” In Hellenistische Studien: Gedenkschrift für Hermann Bengtson,
ed. J. Seibert, 41–53. Münchener Arbeiten zur Alten Geschichte 5. Munich: Editio Maris.

Jones, A., ed. (1999). Astronomical Papyri from Oxyrhynchus (P.Oxy. 4i33–4300a). Mem­
oirs of the American Philosophical Society 233. Philadelphia: American Philosophical So­
ciety.

Kaper, O. E. (2003). The Egyptian God Tutu: A Study of the Sphinx-god and Master of
Demons with a Corpus of Monuments. Orientalia Lovaniensia Analecta 119. Leuven:
Peeters.

Lauer, J.-Ph., and C. Picard. (1955). Les statues ptolémaïques du Sarapeion de Memphis.
Publications de lʼInstitut dʼArt et dʼArchéologie de lʼUniversité de Paris 3. Paris: Presses
universitaires de France.

Lippert, S., and M. Schentuleit, eds. (2005a). Tebtynis und Soknopaiu Nesos: Leben im
römerzeitlichen Fajum. Wiesbaden: Harrassowitz.

———. 2005b. “Die Tempelökonomie nach den demotischen Texten aus Soknopaiu Ne­
sos.” In Tebtynis und Soknopaiu Nesos: Leben im römerzeitlichen Fajum, ed. S. Lippert
and M. Schentuleit, 71–78. Wiesbaden: Harrassowitz.

Malinine, M. 1962. “Une lettre démotique à Aménothes fils de Hapou.” Revue

dʼEgyptologie 40: 37–42.

Manning, J. G. (2003). Land and Power in Ptolemaic Egypt: The Structure of Land Tenure.
Cambridge: Cambridge University Press.

Quack, J. 2005. “Die Ueberlieferungsstruktur des Buches vom Tempel.” In Tebtynis und
Soknopaiu Nesos: Leben im römerzeitlichen Fajum, ed. S. Lippert and M. Schentuleit,
105–115. Wiesbaden: Harrassowitz.

Quaegebeur, J. 1977. “Les saints égyptiens préchrétiens.” OLA 8: 129–143.

Page 25 of 27
Egyptian Religion and Magic in the Papyri

———. 1983. “Cultes égyptiens et grecs en Egypte hellénistique: Lʼexploitation des

sources.” In Egypt and the Hellenistic World: Proceedings of the International Colloquium
Leuven 24–26 May 1982, ed. E. Van't Dack, P. Van Dessel, and W. Van Gucht, 303–324.
Studia Hellenistica 27. Leuven: Peeters.

———. 1993. “La justice à la porte des temples et le toponyme Premit.” In Individu, so­
ciété et spiritualité dans lʼEgypte pharaonique et copte: Mélanges offertes au Professeur
Aristide Théodoridès, ed. C. Cannuyer and J.-M. Kruchten, 201–220. Ath—Brussels—
Mons: Association montoise dʼEgyptologie.

Quaegebeur, J., W. Clarysse, and B. Van Maele. 1985. “Athêna, Nêith and Thoêris in Greek
Documents.” ZPE 60: 217–223.

Riggs, C. (2005). The Beautiful Burial in Roman Egypt: Art, Identity, and Funerary Reli­
gion. Oxford: Oxford University Press.

Rogge, E., and H. Harrauer. (1999). Kunsthistorisches Museum Wien. Aegyptis­

(p. 589)

che-Orientalische Sammlung. 11 Statuen der 30. Dynastie und der ptolemäisches-römis­

chen Epoche. Mainz am Rhein: von Zabern.

Ryholt, K. 2005. “On the Contents and Nature of the Tebtynis Temple Library.” In Tebtynis
und Soknopaiu Nesos: Leben im römerzeitlichen Fajum, ed. S. Lippert and M. Schen­
tuleit, 141–170. Wiesbaden: Harrassowitz.

Sauneron, S. (1988). Les prêtres de lʼancienne Egypte. Paris: Édition Perséa.

Sternberg-El Hotabi, H. (1999). Untersuchungen zur Ueberlieferungsgeschichte der Ho­

russtelen: Ein Beitrag zur Religionsgeschichte Aegyptens im 1. Jahrtausend v.Chr. Aegyp­
tologische Abhandlungen 62. Wiesbaden: Harrassowitz.

Stewart, R. (2001). Sortes Astrampsychi. Vol. 2, Ecdosis altera. Munich: Saur.

Tait, W. J. 1995. “Theban Magic.” Papyrologica Lugduno-Batava 27: 169–182. Leiden: Brill.

Thompson, D. J. (1988). Memphis under the Ptolemies. Princeton: Princeton University

Press.

Valbelle, D., and G. Husson. 1998. “Les questions oraculaires dʼEgypte: Histoire de la
recherche, nouveautés, et perspectives.” In Egyptian Religion: The Last Thousand Years,
ed. W. Clarysse and H. Willems, 1055–1071. Leuven: Peeters.

Van Minnen, P. 1998. “Boorish or Bookish? Literature in Egyptian Villages in the Fayum in
the Graeco-Roman Period.” Journal of Juristic Papyrology 28: 99–184.

Vandorpe, K. 1991. “Les villages des ibis dans la toponymie tardive.” Enchoria 18: 115–
122.

———. 2005. “Agriculture, Temples, and Tax Law in Ptolemaic Egypt.” CRIPEL 25: 165–
171.
Page 26 of 27
Egyptian Religion and Magic in the Papyri

Véïsse, A.-E. 2004. Les “révoltes égyptiennes”: Recherches sur les troubles intérieurs en
Egypte du règne de Ptolémee III à la conquête romaine. Studia Hellenistica 41. Leuven:
Peeters.

Widmer, G. 2002. “Pharaoh Maâ-Rê, Pharaoh Amenemhat, and Sesostris: Three Figures
from Egypt's Past as Seen in Sources of the Graeco-Roman Period.” In Acts of the Seventh
International Conference of Demotic Studies, Copenhagen, 23–27 August 1999, ed. K. Ry­
holt, 377–393. Copenhagen: Carsten Niebuhr Institute of Near Eastern Studies.

Youtie, H. C. 1948. “The Kline of Sarapis.” Harvard Theological Review 41: 9–29 = Scrip­
tiunculae, vol. 1, 487–509. Amsterdam: Hakkert, 1973.

Willy Clarysse

Willy Clarysse is Professor of Ancient History, Katholieke Universiteit, Leuven.

Page 27 of 27
The Papyri and Early Christianity

The Papyri and Early Christianity

David G. Martinez
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0025

Abstract and Keywords

Two events, one occurring during Greek rule in Egypt and the other during Roman rule,
brought about a new status for the great world language. The first was the translation of
the Septuagint, the Greek version of the Hebrew Bible, produced in Egypt, starting early
in the Ptolemaic dynasty. The second was the advent of Christianity and the vigorous pro­
duction and dissemination of Greek Christian scriptures and other sacred material in
Egypt and throughout the Mediterranean world. This article reflects Greek in its capacity
as the language of the sacred, as well as of daily life in early Christianity. It examines
Christian literary papyri and Christian documentary papyri.

Keywords: Mediterranean world, Greek Christian scriptures, Christianity, Septuagint, Hebrew Bible, Christian lit­
erary papyri

AFTER Alexander the Great brought Egypt into the fold of the Macedonian empire in 332
BCE, Greek began to take root in that country as the language of business, commerce,
and government as it did in other parts of his realm. This situation did not change when
Egypt came under Roman control after 30 BCE. Two events, one occurring during Greek
rule in Egypt and the other during Roman rule, brought about a new status for the great
world language. The first was the translation of the Septuagint, the Greek version of the
Hebrew Bible, produced in Egypt, starting early in the Ptolemaic dynasty (Jellicoe 1968,
29–73). The second was the advent of Christianity and the vigorous production and dis­
semination of Greek Christian scriptures and other sacred material in Egypt and through­
out the Mediterranean world.

For Diaspora Jewish communities and Christians in Egypt and elsewhere, the Septuagint
became far more than a Greek rendition of a sacred Hebrew text. It assumed a life and
momentum of its own as an inspired document, and Philo characterized the legendary
seventy-two Jewish scholars who produced it “not as translators, but as prophets and hi­
erophants” (de vita Mosis II 40). Moreover, Augustine demanded that Jerome base his
Latin version of the Old Testament on the holy translation rather than the original He­
brew.1 Early Christian texts completed what the Septuagint had started. The world lan­

Page 1 of 35
The Papyri and Early Christianity

guage thus became the (p. 591) language of sacred scripture,2 and by and large, the same
koinê Greek idiom (tempered at times by bilingual influence) in which the earliest Chris­
tians of Egypt filed their petitions with the government, jotted down their receipts of sale,
and composed their private letters was the language in which they read their Bible.3 The
two parts of this chapter reflect Greek in its capacity as the language of the sacred, as
well as of daily life in early Christianity.

Christian Literary Papyri

The earliest Christian documents on papyrus are biblical and literary. In examining this
material, let us first consider the most frequently read texts of this sort in Egypt, ranked
here by the number of separate copies.4 A list such as this has limited and largely symbol­
ic value.5 We of course do not know how many copies of such works remain unidentified
and unpublished. In addition, it may seem strange that an elegant codex that contains al­
most all of Matthew's gospel and a scrap of papyrus used as an amulet with fragmentary
remains of Matthew's paternoster each counts as one copy. The latter state of affairs is le­
gitimate, however, since two such documents provide distinctive and different kinds of ev­
idence for the use of that biblical book. With regard to the former problem, comparison of
this list with a similar one using van Haelst's 1976 (p. 592) catalogue shows that, although
the number of copies of individual works has fluctuated over thirty years, the general dis­
tribution of what were the most popular biblical and Christian documents in Egypt has
not changed very much.

Psalms: 209

Matthew: 50

John: 45

Genesis: 36

Exodus: 29

Acts: 25

Shepherd of Hernias: 23

Daniel: 22

Romans: 20

Luke: 19

Page 2 of 35
The Papyri and Early Christianity

Isaiah: 18

Hebrews: 15

I Corinthians: 14

Origen: 14

Ecclesiasticus: 13

Revelation: 13

Jeremiah: 12

Mark: 12

Criteria for Determining Christian Ascription

By a vast margin the Psalter emerges as the most frequently read text from the Greek
Bible in Egypt. This and other Old Testament books were of course read by Hellenized
Jews living in Egypt as well as Christians. Sometimes it is impossible to ascribe with any
certainty a papyrus to one tradition or the other. Various factors, however, such as the
presence of contracted sacred names (nomina sacra), codex format, types of script, and
various conventions of writing numbers, make it likely to a greater or lesser degree that
most of the copies of works from the Greek Old Testament from Egypt come from Chris­
tians rather than Jews.6 The first two of these criteria merit more than brief mention.

The codex,7 made of either papyrus or parchment, is the form of presentation that resem­
bles the modern book in contrast to the older format of the roll (chapter 11), the latter
consisting of sheets glued together side by side and wound around a dowel of wood,
bone, or metal,8 with the writing only on the internal side or “recto” (the side with the
fibers running horizontally). The codex was produced by stacking pages on top of each
other and folding them down the middle, with writing continuing from front to back.
Many earlier codices are of the “single quire” type, that is, one stack of folded sheets. As
the format grew more sophisticated, the refiners of this style assembled a number of
smaller folded stacks, or quires, and sewed them together at the back. Whereas Chris­
tians greatly preferred the codex format to the roll, at least for the scriptures, from the
earliest times their literature appears in Egypt (i.e., the second century CE), we know of
very few fragments of codices datable before the fourth century whose Jewish ascription
is verifiable or at least made likely on other grounds.9 Until more evidence on early Jew­
ish codex copies emerges, these observations allow us to affirm that, for the second and

Page 3 of 35
The Papyri and Early Christianity

third centuries, a Greek copy of the Old Testament in codex format, absent other indica­
tors for or against, is likely to be of Christian origin.

The presence of nomina sacra in papyrus documents affords surer evidence of Christian
authorship.10 In the early centuries CE Christian scribes began writing certain sacred
words (or words to which sacred meaning was extended) in abbreviated form with an
overstrike. They practiced this most frequently on about fifteen words, of which the fol­
lowing four were the most frequently and consistently abbreviated from early times: θεός
(θ̅c,̅ θ̅υ)̅ , κύριος (κ̅c̅, κ̅υ)̅ , Ἰησοṽς (ι̅c,̅ ι̅υ̅), Χριστός (χ̅c̅, χ̅υ)̅ .11

The most vexing issues connected with this convention continue to be its ultimate origins
and the related problem of why words that belong to a fairly set corpus are abbreviated
and others just as theologically “charged” (such as logos)are not. The early investigator L.
Traube (1907) sought to explain the practice as a (p. 593) Hellenistic Jewish corollary to
the abbreviated writing of the Tetragrammaton. Scholars of documentary Greek texts
have understood it in terms of conventions of abbreviation in inscriptions and nonliterary
papyri. Others have traced it to Christian scribes who began with either κύριος or θεóς as
a surrogate for the Tetragrammaton, extending the technique from those to other
words.12 Contesting Traube's thesis of Jewish origins, C. H. Roberts notes that among
Greek papyrus and parchment copies of the scriptures, which are demonstrably Jewish,
the presence of these abbreviations is negligible.13 He describes the roster of abbreviated
words, particularly with regard to the four primary terms, as embodying an “embryonic
creed of the first church” (1979, 46), originating from the Christian community in
Jerusalem and representing a common stock of Christian beliefs. This formulation sug­
gests the possibility that one or more authoritative texts, which had a creedal flavor, set
(or at least helped to set) the roster.14

Septuagint/Old Testament

Psalms is striking not only for number of copies but also for the antiquity of many of
them. It is widely held that the earliest Christian biblical papyrus is the Rylands Gospel of
John fragment, P52 (discussed later). One of our earliest papyri of Psalms, however, cer­
tainly rivals and possibly usurps that title: Bodl. MS. Gr. bibl. g. 5 (P), 15 a codex fragment
that preserves parts of Ps. 48.20–49.3; 49.17–21 (figure 25.1a–b). The editors, Barns and
Kilpatrick (1957), date the script from the end of the first century to the end of the sec­
ond. The codex format at this early date makes Christian ascription likely.16

Various factors drive the overwhelming popularity of this book among Christian readers
in Egypt. Liturgical papyri and other materials reveal the abundant use of the Psalms in
corporate worship and private devotion, as well as their impact on liturgical language
(Schermann 1912, 196–201). Monks in the Pachomian community were expected to learn
the Psalms;17 recitation and/or reading of them formed a vital part of the daily office of
monastic groups.18 On a theological level, the authors of the New Testament (Hurtado
2006, 28), as well as the early fathers, including those from Egypt, found in the Psalms a
treasure trove of Christological prophecy, typology, and allegory.19 Origen authored a

Page 4 of 35
The Papyri and Early Christianity

number of exegetical works on the book, including probably his earliest commentary
(about 222 CE), which reached Psalm 15 (16).20 Didymus produced an exhaustive exposi­
tion, some of which survives in the Tura papyri (discussed later). Under Athanasius's
name we have fragments of scholia to the Psalter.21

Figure 25.1a–b. Bodl. MS. Gr. bibl. g. 5 (P), recto

and verso: fragment of a codex that presents parts of
Ps. 48.20–49.3; 49.17–21. Photo courtesy of the
Bodleian Library, University of Oxford.

In addition, as Christians sought a voice for their own literature alongside (but by no
means replacing) the classical standards for schools,22 the Greek Psalter served as a wor­
thy parallel to Homer in its antiquity, poetic power, and didactic value.23 Of the approxi­
mately thirteen documents that use biblical material for instruction, collected by Cribiore
(1996) in her catalogue of school exercises, eight come from the Psalms. For example,
P.Lond.Lit. 207 preserves on its recto side (p. 594) (p. 595) Psalms 11.7–14.4, written in a
large and elaborate capital script, which the editors date to the third/fourth century, with
dots added by another hand above the letters for syllable division to the end of Ps. 13.24
On the verso is written Isocrates' ad Demonicum 26–28, in a different hand of the same
general date, described by Cribiore as that of a teacher, also with dots but between the
letters and probably added by the same hand that wrote it.25 The syllabic dots in both
passages point to an educational context, and the fact that the Psalms text contracts
κύριος, θεóς, and ἄνθρωπος as nomina sacra strongly argues for Christian provenance.
Texts such as this lend a measure of documentary confirmation to the views of the earlier
Greek fathers, particularly Origen and the fourth-century Cappadocians, who encouraged
Christian students to study Greek literature alongside biblical texts.26

In yet another respect, for Christians the Psalter held a position similar to that of Homer,
that is, as an ancient poetic text of magical power. This to some extent applies to the book
as a whole; however, certain Psalms, especially LXX 90 (91), had particular prestige.
There are now at least twenty-five texts from Egypt that preserve all or parts of 90 (91)
Page 5 of 35
The Papyri and Early Christianity

and have been identified either certainly or possibly as magical amulets dating from the
fourth to the tenth century.27

Of the Greek Old Testament after Psalms, Genesis and Exodus claim the greatest number
of separate copies. For example, P.Yale I 1 (van Haelst 1976, 12) preserves Gen. 14.5–8 on
the recto and 12–15 on the verso. Its codex format and convention of writing numerals as
symbols, which differs from the normal Jewish practice, makes it likely that it is from
Christian rather than Jewish hands (Roberts 1979, 78). The original editor, C. Bradford
Welles, dated the script to the last quarter of the first century CE and regarded it as the
oldest evidence for the codex and the earliest Christian document. Most palaeographers,
however, have assigned the text to the late second or early third century.28 P.Amh. I 3, in
addition to preserving on the recto side one of the earliest Christian private letters (dis­
cussed later), on the verso gives us possibly the oldest Greek text of Genesis 1.1–5 (Gren­
fell and Hunt assign the script to the late third/ early fourth century), copied in both the
LXX and in the version of Aquila (van Haelst 1976, #3). Musurillo (1956, 126) identifies
this text as an amulet, but reusing a letter for two dif erent versions of exactly the same
passage rather suggests an educational function (Aland 1976, #Var 35 p. 360; see figure
25.3 below).29

New Testament

We are immediately struck by the extent to which the survival rate of separate copies of
Matthew, John, and Acts exceeds that of Pauline and other epistolary literature. The
gospels, with and without Acts, were sometimes grouped in a single codex separate from
the epistolary section of the NT, which was also, as we shall see, at times so grouped. The
facility of the format for clustering related genres of literature surely fostered its rise to
popularity.30 One of the earliest such gospel/Acts (p. 596) codices, P.Chester Beatty I or
P45 (incorporating a separately published fragment, P.Vindob.inv.G. 31974), comprises
fragments of thirty leaves of a codex that originally contained around fifty–five leaves,
dating to the third century (van Haelst 1976, #371). This manuscript probably, as another
Egyptian gospel codex, Codex Washingtonianus (ibid., #331), definitely did, placed the
gospels in the so-called Western order: Matthew, John, Luke, and Mark.31 The evidence of
magic further reinforces the conception of the four gospels as an authoritative and pow­
erful textual unit. Their opening verses were considered particularly potent formulae and
were employed as amulets, interlaced with Ps. 90 (discussed earlier) and other texts of
power.32

Papyrological evidence has favored the gospel of John with early survivals, the most fa­
mous of which is P52 (Jn. 18.31–33, 37–38).33 It was C. H. Roberts (1935) who first dated
this fragment to the early part of the second century, judging it the oldest copy of any
part of the NT and the earliest Christian papyrus. Although the roster of papyrologists
and other scholars who agree or mostly agree with Roberts's assessment is impressive,34
it is not unanimous. Notably E. G. Turner (1977, 100) cautioned that the late second cen­
tury also presents viable palaeographic parallels, and more recently B. Nongbri (2005), in
an exhaustive examination of previously offered, as well as newly adduced evidence, has

Page 6 of 35
The Papyri and Early Christianity

made the same argument, extending the possibilities into the early third century. Indeed,
early dating for P52 and implications drawn from it have sometimes proven
overzealous.35 Small and fragmentary documents make it difficult for us to form a palaeo­
graphic “big picture,” always more important for dating than individual letter shapes. For
texts such as this it is safer to posit a broader spread, such as “early second to early third
century.” 36

Figure 25.2. Chester Beatty Biblical Papyrus II (P46;

third century); leaf 41, which comprises the last
words of the epistle to the Romans (from 16.23; this
manuscript transposes the doxology of 16.25–27 to
the end of chapter 15) and the opening of the epistle
to the Hebrews. After the end of Romans, the nota­
tion roughly calculates the length of the book as
“1,000 lines” for the purpose of paying the scribe.
Image reproduced by permission of the Papyrology
Collection, Graduate Library, University of Michigan.

Although the Pauline epistles lag behind the two best-attested gospels and Acts and even
the three best-represented OT books in number of copies, they boast one of the earliest
and fullest papyrus codices, P.Chester Beatty II, supplement (P46; van Haelst 1976, #497)
(figure 25.2). The surviving eighty-six leaves, which are divided between the University of
Michigan and the Chester Beatty Library (Dublin), 37 comprise nine epistles ascribed to
Paul in order by size: Romans (beginning at 5.17), Hebrews,381 and 2 Corinthians, Eph­
esians, Galatians, Philippians, Colossians, and 1 Thessalonians. The codex is a single-
quire39 type, and judging from what we know of that format, we may determine that the
original (probably of about 104 leaves) lacked all or part of 1 and 2 Timothy and Titus
since the missing leaves at the end could have accommodated 2 Thessalonians but not the
pastorals (Epp 2002, 495–502). Majority scholarly opinion favors an early third-century
date of the script.40 This splendid document may not, however, be the earliest papyrologi­
cal evidence for literature under Paul's name. That distinction should likely go to a small
fragment of a codex, P.Ryl.I 5 (van Haelst 1976, #534), which preserves a few verses of a
Page 7 of 35
The Papyri and Early Christianity

book that P46 likely omits, Titus 1.11–15, 2.3–8, which some scholars date to the second
century.41 P46's incorporation of Hebrews (p. 597) (p. 598) as Pauline, as well as its likely
omission of the pastoral books, and this early fragment of one of those books, which may
have been part of a larger collection, provide evidence for emergent attitudes toward
canonicity and authorship.42

Biblical Papyri: Some Observations

In the two previous sections we have observed that the best-attested books of the Greek
Bible among Egyptian Christians are Psalms, Matthew, John, Genesis, and Exodus. The
penchant for the Fourth Gospel has generated considerable discussion on a number of
fronts, especially with regard to the alleged “Gnostic” coloring of the early Egyptian
church. For example, K. Aland (1967, 99–103) has argued that the popularity of John in
Egypt evinces a dominant strain of Gnosis since John was the preferred gospel of Gnos­
tics. However we may assess the influence of Gnosticism in Egypt on other grounds, the
papyrological data on the gospels do not support it.43 Even if we limit our survey to the
earliest papyrus copies (between approximately the second to early fourth centuries), the
difference between the number of copies of John (seventeen) and Matthew (fourteen) is
not great. When our investigation extends into later antiquity and other media besides pa­
pyrus, reflected in the list that introduced this section, Matthew slightly trumps John.
Thus, we cannot in any real sense speak of a favoritism toward John either in number or
antiquity of copies, regardless of whether this gospel was any more popular among Gnos­
tics than in catholic/orthodox communities.44

What is more striking is the preponderance of Matthew and John over Luke and Mark.45
The perception may have been that the content of the other Synoptics was already ac­
commodated in Matthew, and that Matthew and John, the two eyewitnesses, fully articu­
lated the polarity of “somatic” and “spiritual” respectively, in their portrait of Jesus, to
use the terms of Clement of Alexandria.46 We may reasonably see the somatic/spiritual
pattern emerging in the three dominating Old Testament books as well, with Genesis and
Exodus revealing the great narratives of creation, the patriarchs, and Moses and with the
Psalms providing spiritualizing and doxological reflection on these events as Heils­
geschichte.

Apocryphal Literature

Of the substantive remains of Apocryphal materials on papyrus (van Haelst 1976, #568–
620), the following documents are among the earliest and most significant. The first pub­
lished Oxyrhynchus papyrus (van Haelst 1976, #594), which preserves several logia of Je­
sus and dates palaeographically to the early third century, was later identified as a Greek
fragment of the Gospel of Thomas, based on the 1945 discovery of a full Coptic transla­
tion among the Nag Hammadi finds. A few years later, two other third-century papyri
from Oxyrhynchus were published and subsequently (p. 599) identified as belonging to the
same work.47 That the text of these fragments differs at points from that of the Coptic

Page 8 of 35
The Papyri and Early Christianity

probably indicates that more than one Greek edition of the Gospel of Thomas circulated
in antiquity (Cameron 1992, 535).

Of a similar date as P.Oxy.I1 are the Jesus logia of P.Lond.Christ. 1 (P.Egerton 2), 48 which
consist of two codex leaves and fragmentary remains of a third, preserving about ninety
lines of text with an additional fragment of one of the leaves published by M. Gronewald
(P.Köln VI 255), adding four new lines of text, and supplementing six others. Bell and
Skeat, the original editors of P.Lond.Christ. 1, dated the hand to the mid-second century.
Gronewald notes that the new Cologne fragment exhibits apostrophe between conso­
nants, a palaeographic trait not common until the third century.49 In content, the papyrus,
in its versions of Jesus' “Search the scriptures” dialogue with the Pharisees, the cleansing
of the leper, and the answer to the question about paying taxes, displays an interweaving
of Johannine and synoptic accounts (Jn. 5.39, 45; Jn. 9.29 et al.; Mk. 1.40 ff and parallels;
Mk. 12.13 ff and parallels; see Porter 2006, 326 f). In addition, in a lamentably fragmen­
tary section, it preserves a miracle story “on the bank of the Jordan” involving sowing and
reaping, found in none of the canonical gospels. If, as seems likely, the author knew the
New Testament gospels, he probably had none before him as a written text but repro­
duced tracts of them from memory (Jeremias 1963).

Patristic Literature

Of the rich patristic literature that flourished from the second to the eighth century, the
papyri present us with wealth in some cases and bewildering dearth in others.50 The po­
larities are well illustrated by the prominent and prolific early authors who lived in Egypt.
Of the five great Alexandrian fathers of the first five centuries, Clement (ca. 150–ca. 215),
Origen (ca. 185–ca. 254), Athanasius (ca. 296–373), Didymus (ca. 313–398), and Cyril (d.
444), Origen and Didymus survive in appreciable quantities, mainly due to one spectacu­
lar find in 1941. A huge cache of papyri in a cave in Tura, about twelve kilometers south
of Cairo, yielded more than two thousand papyrus pages, including eight codices, palaeo­
graphically dated to the sixth or seventh century and probably belonging to the nearby
monastery of Saint Arsenios.51 The Origen material preserved among the Tura codices
(182 pages) includes large sections of the theological treatises Contra Celsum, de Pascha,
and Dialogus cum Heraclide, as well as his homily on the “witch of Endor” episode in I
Sam. (LXX I Reg.) and his commentary on Romans.52

The majority of the Tura papyri, however (1,118 pages), preserve works of Didymus the
Blind, namely extensive portions of his exegetical treatises on Genesis, Job, Psalms, Eccle­
siastes, and Zacharias, as well as smaller fragments, including a dialogos with a heretic,
the only example of nonexegetical writing we have in this group, and a small section of
his commentary on John (covering 6.3–6.33), the only piece of New Testament exegesis
we have from Didymus (p. 600) any where.53 The Psalms and Ecclesiastes commentaries
are actual lectures to advanced students and thus offer us an invaluable glimpse into
what might be called a university classroom of fourth-century Alexandria.54 These com­
mentaries richly illustrate what some Christian school texts considered earlier suggest:

Page 9 of 35
The Papyri and Early Christianity

the integration of biblical and classical texts in the Christian school curriculum in
Egypt.55

Aside from the Tura finds, the most notable survival of the Alexandrian theologians is
fifty-two well-preserved pages of a codex from the sixth-seventh century found in El Deir
(north of Hawara in the Fayyum), containing a large section (books 6–9) of Cyril's treatise
de adoratione et cultu in spiritu et veritate, which is now spread among collections in
Dublin, London, Paris, and Vienna (van Haelst 1976, #638). No papyri have yet surfaced
of the works of the great Athanasius, a formidable presence in some documentary texts.56
A few lines from Clement's Stromata VI 8 survive on a fragmentary papyrus in Cologne.57

Some of the earliest Christian authors, including those known as the Apostolic Fathers,
are the most poorly represented. The papyri give us not a single fragment of Clement of
Rome, Polycarp, or Papias; moreover, no copies of the two early examples of apologetic
literature, the works of Justin Martyr and the Epistle to Diognetus, survive. The Didache,
Ignatius, and Barnabas each rate a single Greek papyrus.

Conversely, one of the documents among the “Apostolic Fathers” canon boasts more
copies than any other non-biblical text and more than many biblical ones: The Shepherd
of Hermas. With the recently published Oxyrhynchus fragments (P.Oxy. LXIX 4705–4707)
we now have twenty-three different papyrus copies of this enigmatic author from a fairly
wide geographic distribution.58 As C. H. Roberts (1979, 22) remarks, “[The Shepherd of
Hermas'] popularity in provincial Egypt may give us a better insight into the character of
the churches than anything else.” He says this with regard to the intense Jewish coloring
of the work, which, according to him, looms large also in early Egyptian Christianity (in
contrast with the Gnostic strain that some have seen as predominant). More specifically,
the treatise, with its emphasis on purity of heart, which induces purity of action, and its
correlating of goodness and evil with angelic and demonic beings, is reminiscent of the
thought of the monastic leader Pachomius. The Shepherd may have furnished, if not a
textbook, at least inspiration for monastic ideals (Rousseau 1985,136–38).

The practical focus of The Shepherd also lends itself to an educational context; both Euse­
bius (hist. eccl. III 3.6) and Athanasius (ep. fest. 39) attest that The Shepherd was used
for the instruction of catechumens (Staats 1986, 107). It would not be difficult to view the
work as a kind of Christian corollary of Isocrates' ad Demonicum.59 Also, whether or not
one judges the work as an apocalypse, The Shepherd breathes the spirit of the apocalyp­
tic throughout (Osiek 1999,10–12). We see the affinity of Egyptian Christians for apoca­
lyptic literature from the stature of the books of Daniel and Revelation in the papyri.60 In
addition, the first literary evidence of Christianity in the Arsinoite nome is Eusebius's ac­
count (hist. eccles. (p. 601) VII 24) of Dionysius, bishop of Alexandria (elected 248), paying
a pastoral visit to the Christians in the region in order to teach them the true meaning of
Revelation and refute the false interpretation of a local bishop, that the book promised an
earthly millennium of carnal pleasure.61

Page 10 of 35
The Papyri and Early Christianity

Christian Documentary Papyri

When we come to documentary evidence, a disappointment immediately confronts us:
Whereas on the literary side we have several papyri whose scripts editors and other
scholars have dated with a degree of confidence to the second century, there are no docu­
mentary Christian texts indisputably assignable to that early period. In fact, even for the
first part of the third century, the volume of identifiably Christian nonliterary papyri is
meager and at times controversial. The fact that certain “monotheistic” expressions and
other types of religious language appear in both Christian and non-Christian private let­
ters and other documents complicates the task of identification considerably.62

This difficulty finds ample illustration in the case of P.Harris 107. The editor (with reser­
vation) and other scholars have dated the hand to the early third century, and some have
judged it one of the earliest, if not the earliest, Christian private letters.63 In it a certain
Besas addresses his mother, “Heartiest greetings in God. Above all I pray to God the Fa­
ther of the truth and to the Comforter, the Spirit, that they protect (διαϕυλάξωσιν) you in
soul, body, and spirit; for your body, health (ὑγείαν); for your spirit, happiness (εὐθυμίαν);
and for your soul, eternal life (ζωὴν αἰώνιον).” Besas then instructs that his garment be
sent “for the Pascha festival” (εἰ<ς> τὴν έορτὴν τοṽ Πάσχα). In addition to other aspects
of the letter, the fact that the prayer apparently invokes the dyad of Father and Spirit (but
omits any reference to the Son) has raised the possibility of Gnostic or Manichaean iden­
tity.64 Earlier defenders of a Christian provenance, however, have argued for τῆς
ἀληθείας as a Christological title based on John 14.6 (“I am the way, the truth, and the
life”).65 The entire invocation could thus amount to a Trinitarian formula that corresponds
to the human psychological triad of “soul, body, and spirit,” which itself, despite the dif­
ferent order of the terms, recalls the Pauline blessing of I Thes. 5.23 (“May your spirit,
soul, and body be kept sound and blameless”), as in other Christian papyri.66

More recently, Gardner, Nobbs, and Choat (2000) have reasserted the Manichaean identi­
ication of the letter, dating it late third or early fourth century and arguing from cogent
parallels to its phraseology found in the recently published Manichaean Greek and Coptic
papyri of the Kellis collection and in other Manichaean literature. The authors mount a
strong (but to my mind not decisive) case (p. 602) that the letter is Manichaean rather
than Christian. The date is crucial, since Manichaean provenance requires that it be late
third century at the earliest. Although, as I have already mentioned, palaeographic dating
must be allowed a generous spread, the most convincing parallels to the papyrus' script
are documentary hands of the earlier third century rather than the third/fourth-century
school hands adduced by Gardner, Nobbs, and Choat (2000).67

Manichaeans at Kellis and elsewhere indeed seem to have adopted “God of truth” as a
kind of watchword. The phrase, however, also occurs in a well-known Psalm (LXX 30(31).
6), frequently commented on and alluded to by Alexandrian exegetes and other church fa­
thers, sometimes with specific reference via John's gospel to “Truth” as a Christological
title.68 The fact that it occurs ten times in the fourth-century liturgical compilation known
as the “Prayers of Serapion” suggests that it could have had currency in other Egyptian

Page 11 of 35
The Papyri and Early Christianity

liturgical traditions both then and earlier.69 With regard to the letter's elaboration of the
Pauline trichotomy (“body, soul, spirit,” here in the order common in Egyptian Christian
liturgies: “soul, body, spirit”), for which Gardner, Nobbs, and Choat (ibid.) adduce paral­
lels from the Coptic Kellis Manichaean texts, it is also the case that similar language ap­
pears in the Serapion compilation (1 Johnson [p. 50]; 13.19 Funk). In addition, colloca­
tions such as “body, soul, and spirit,” as well as (δια)ϕνλάσσαν (“protect”), also occur in
magical texts (the latter especially in the imperative as a common amuletic formula).70

This modest rebuttal to the evidence of Gardner, Nobbs, and Choat (ibid.) does not refute
their claim, but it does at least suggest for P.Harris 107 and the Kellis Manichean texts
that the phrases common to both could have their ultimate source in the language of
liturgy and protective magic. The reference to Easter in this letter could accommodate ei­
ther Christian or Manichaean provenance.71

Figure 25.3a. P.Amh.I 3 (Pierpont Morgan Library

Pap. G. 3) recto. A private letter dated to 264/265–
282/283 CE, describing business transactions involv­
ing clergy. Above the second column of the letter in
the margin is written in a late third/early fourth-cen­
tury hand Hebrews 1:1. Photo courtesy of the Pier­
pont Morgan Library, New York.

Another letter, PAmh.I 3 (264/282 CE; Naldini 1998, #6; figure 25.3a–b), portrays a series
of complicated commercial maneuvers in which church officials, at the highest and lowest
levels, play a role. The fragmentary nature, however, of the papyrus, due in part to its be­
ing trimmed down when reused for biblical passages (discussed earlier), frustrates our
best attempts to discern the exact nature of their participation. The lacunae deprive us of
the names of the sender or recipient. The writer (now in Rome) addresses certain parties
who work with (or for) him in the Arsinoite nome and advises them on business matters
that involve, among other things, buying dry goods in the Arsinoite and selling them in
the capital. The moneys from these transactions are to be paid to church officials. The pa­
pyrus specifically mentions Maximus, the current bishop of Alexandria, and a lector
whose name is lost. The two may constitute an ecclesiastical polarity: the bishop as the
Page 12 of 35
The Papyri and Early Christianity

highest church official and the lector the lowest, the first stage of the clerical cursus hon­
orum.72 Involved in the transactions are also Theonas, probably Maximus's successor, and
Preimeiteinos, an associate of the bishop whose clerical status (or lack thereof) is not as
certain. (p. 603) To my knowledge, this papyrus provides our earliest attestation of Christ­
ian clergy in a dated Christian text. In its mutilated state it does not reveal whether the
bishop and his associates function as mere depositories for the funds or play a more ac­
tive and engaged role in the business at hand. The latter scenario, if true, should cause
no surprise. We know from later papyri that bishops frequently had their roots in and
moved among the affluent classes. According to land registers from the Hermopolite
nome (dated to 350–375 CE), some of them owned considerable tracts of property.73

Figure 25.3b. P.Amh.I 3 (Pierpont Morgan Library

Pap. G. 3) verso. Genesis 1:1–5 in the versions of the
Septuagint (lines 1–9) and Aquila (lines 10–16), early
fourth century. The scribe who produced this pas­
sage may be the same one who produced the He­
brews passage on the recto. Photo courtesy of the
Pierpont Morgan Library, New York.

Moving from the realm of private to public documents, we come to two of the earliest
nonliterary papyri that mention the word “Christian.” P.Oxy. XLII 3035, an “order to
arrest” (or summons; see Gagos and Sijpesteijn 1996), dated to February 28, 256, from
the president of the city council, describes the wanted man as (p. 604) “Petosorapis, son of
Horos, a Christian” (Χρησιανόν, that is, Χρησ <τ>ιανόν; the editor, P. Parsons, notes an­
other example of this spelling74). It is not certain whether his faith is the reason for the
summons or merely incidental. If the former, the papyrus could pertain to the persecution
under Valerian75 this text, however, dates a year before that emperor's actions against be­
lievers. This is not to say there may not have been smaller persecutions by individual au­
thorities or that Petosorapis's faith was not somehow involved. The designation, however,
may be no more than an individualizing description, such as those that specify profession,
which occur in directives such as these.76 The papyrus shows incidentally that some
Egyptian Christians had no problems with traditional theophoric names (Petosorapis

Page 13 of 35
The Papyri and Early Christianity

means “given by Sarapis”; Pestman 1994, 32, 46). (p. 605) Probably the earliest public
document that mentions a Christian in Egypt, P.Vindob. inv. G 32016,77 which dates to the
early third century CE, comprises a list of nominees for liturgical office in Arsinoe, the
capital of the Arsinoite nome in Middle Egypt. Each entry typically provides the name of
the nominee, a second name by which he is known, his father's and grandfather's names,
the district of the metropolis where he lives, and the net worth of his property (one had to
own a certain amount of land to be considered for this office). Following these standard
items, a different clerk adds a further descriptive phrase that one could call “other” or
“miscellaneous,” to give the superior officer in charge of ranking the nominees further in­
formation on which to base his decisions. About one of them, a certain Antonius
Dioscorus, the first clerk records, “Antonius Dioscorus, son of Origen, Alexandrian” (the
district and property designations are lost); then the second clerk adds, ἔστ (ι) Διόσκορος
Χρηστιανός, “Dioscorus is a Christian.” It is likely that the clerk intends this characteriza­
tion as a liability rather than an advantage (van Minnen 1994, 76). Nevertheless, the eval­
uator of the nominees ranks Dioscorus second in qualifications for the office.78

At a minimum this text tells us that Christians, although periodically suffering persecu­
tion under the Roman authorities, participated in public life and local government in
Egypt from an early date. Through their economic means and social position, they were
competitive for public oice despite a skeptical and at times hostile political atmosphere.
With regard to social status, Dioscorus possesses distinct advantages. He has a Roman-
style double name, with a Roman gentilicium and a Greek cognomen (a distinction pos­
sessed by no other nominee for this particular liturgy), as well as Alexandrian citizenship.
These factors may have merited him (or perhaps imposed upon him, since many wished to
avoid such public service) a high place in the ranking, despite a likely attempt by one of
the clerks to scuttle his candidacy by marking him as a Christian (ibid.).

The four pieces here considered, some of our earliest documentary papyrological evi­
dence for Christianity in Egypt, present a microcosm of the issues that such texts raise:
the relationship of the church with the world (especially with the civil establishment), the
emerging Christian clergy and spiritual elites, the increasing influence of Christian writ­
ings on daily life, and the obscure shadings between orthodoxy and heresy. The following
topics explore these issues further.

Persecutions of Christians
Although earlier regimes conducted localized cases of persecution (Frend 1965, 389–
398), no general edict against the faith emerged until the reign of Decius in 249.79 In that
year the emperor ordered all subjects to sacrifice to the gods and document their act with
petitions or requests for state certiication. Refusal could bring death, as the first attested
martyrdom from the edict, that of Pope Fabian (p. 606) (January 250), shows.80 Forty-five
applications for such certificates, called libelli, have survived on papyrus, of which thirty-
four are from the town of Theadelphia. A very typical example is P.Mich. III 157, dated to
June 17, 250 CE:81

Page 14 of 35
The Papyri and Early Christianity

[first hand] To the officials in charge of the sacrifices, from Aurelius Sakis of the
village Theoxenis, with his children Aion and Heras, temporarily residents in the
village Theadelphia. We have always been constant in sacriicing to the gods, and
now too, in your presence, in accordance with the regulations, we have sacrificed
and poured libations and tasted the offerings, and we ask you to certify this for us
below. May you continue to prosper.

[second hand] We, Aurelius Serenus and Aurelius Hermas, saw you sacriicing.

[first hand] The first year of the Emperor Caesar Gaius Messius Quintus Traianus
Decius Pius Felix Augustus, Pauni 23.

The fact that we do not have Decius's actual edict and cannot be sure of its precise word­
ing has led some historians to question whether it was actually a directive against Chris­
tians or a traditionalist emperor's attempt to shore up the piety of a beleaguered state
(Rives 1999). Christian authors certainly believed the former,82 and the structure of the li­
belli lends some support to that view. The formulaic phrase in these documents, “we have
always been constant in sacrificing to the gods” (vel sim.), seeks to avoid any suspicion of
the forbidden faith. Also, the declarations are quite meticulous in affirming that the par­
ticipant performed the slaughter, libations, and sacrificial meal, the three-tiered ritual of
ancient pagan sacrifice, which one may observe as early as the Iliad and Odyssey. It is al­
so likely that similar mandates by later emperors were directed against Christians, as we
are about to see.

The “order to arrest” (discussed earlier) does not provide conclusive information about
hostilities against Christians under Valerian (253–260), and the same can be said for oth­
er papyri from that period (for a possible exception see Whitehorne 1977). We are on
firmer footing with regard to the persecutions advanced by Diocletian and his fellow
tetrarchs, sometimes called the “Great Persecution” (303–305 in the western part of the
Roman Empire; 303–312 in the eastern part; Frend 1992, 673). Diocletian conducted
these actions by a series of edicts,83 the first of which apparently provided for the de­
struction of Christian churches and scriptures.84 On a more social level, the first edict or­
dered that Christians of the upper classes lose their station, and those holding civil jobs
be reduced to slavery. The “infrastructure” part of that edict finds ready illustration in
P.Oxy. XXXIII 2673, dated to February 5, 304. In this document a lector affirms on oath to
the president of the town council and his associates that the church possesses no goods
whatsoever, “neither gold nor silver nor money nor clothes [vestments?] nor beasts nor
slaves nor lands nor property either from grants or bequests.” Thus, not only was the
church itself demolished and disbanded (he calls it ἥ ποτε ἐκκλησία, “the former
church”), but it had also been forced to yield all of its possessions, of which the lector
claims the church has none, “excepting the bronze implements85 which (p. 607) were
found and delivered to the logistes to be carried down to the most glorious Alexandria in
accordance with what was written by our most illustrious prefect Clodius Culcianus.”

Page 15 of 35
The Papyri and Early Christianity

We learn from the third/fourth-century Christian apologist Lactantius (de mortibus perse­
cutorum 15.586) that Diocletian's first edict also involved a command to sacrifice. In this
case one had to sacrifice to the gods in order to gain access to the courts, a measure de­
signed to bar Christians from the legal establishment. An early fourth-century private let­
ter, P.Oxy. XXXI 2601,87 demonstrates how an enterprising believer might evade this re­
quirement: “Kopres to his sister Sarapias, very many greetings. Before all else, I pray for
the good health of you all before the Lord God. I want you to know that we arrived on the
11th. It became known to us that those who present themselves in court are being made
to sacrifice. I made power of attorney in favor of my brother.” After several other lines he
concludes on the verso side: “Deliver to my sister from Kopres. Amen.” The word “amen”
is written in cryptogram form, Ϟθ that is, the numeral 99, the numerical value of the let­
ters of the word spelled in full (α = 1 + μ = 40 + η = 8 + ν = 50). This isosephism, like the
obscure χμγ, becomes part of a stock of cryptograms and symbols that increasingly ap­
pear in fourth-century or later Christian documents, frequently at the beginning and
end.88 Kopres also shows his faith by casting the words “Lord God” in abbreviated form
with the overstrike, κυρί(ῳ) θ [(ε)ῳ̃, but he does so in a way that reveals either careless­
ness or ignorance of the accepted standards in literary texts.89

Kopres evades the sacrificial test imposed by Diocletian by composing a “deed of repre­
sentation” or “power of attorney” for one he calls “my brother” (άποσυστατικόν ἐποίησα
τῳ̃ άδελϕῳ̃ μου), that is, he authorizes him to appear in court in his place.90 Is this “broth­
er” a pagan friend, as the editor P. Parsons suggests, and, if so, why the designation
“brother”? We must bear in mind that “brother” had a broad currency of usage, and it
may have been possible to apply it to a pagan associate without being misunderstood.91 It
is also possible that “brother” could be taken literally, a real blood relative who was not a
Christian and would not have had difficulty performing the ritual on Kopres's behalf. Re­
gardless of his identity, it is clear that Kopres suffers no crisis of conscience in employing
him in this manner (Judge and Pickering 1977, 53). He also seems to portray the whole af­
fair as an annoyance rather than a major obstacle, and the fact that later in the letter he
sends for his family indicates that he does not consider this present manifestation of
Diocletian's hostilities as having reached the deadly proportions that came later (Parsons
ad loc.).

Monasticism

Monasticism originated in Egypt.92 Early Egyptian monastic life reveals both the ferment
of prayerful contemplation and the turmoil of controversy. This polarity finds full expres­
sion in three dossiers of letters from the mid-fourth century (p. 608) addressed to influen­
tial monks: the archives of Paieous (330s), Papnouthios (probably early 340s), and
Nepheros (mainly 350s),93 all likely associated with the same monastic community, which
had emerged from the Melitian Schism.

Melitios, bishop of Lykopolis, took issue with Peter, bishop of Alexandria, for adopting
what he considered overly lenient measures for the return of those who had lapsed from
the faith during a lull in the Diocletianic persecution (around 306). Mounting acrimony

Page 16 of 35
The Papyri and Early Christianity

led to Peter's excommunicating Melitios. The official persecution increased momentum,

issuing in Peter's martyrdom and Melitios's exile. When the latter eventually returned to
Egypt, he founded a schismatic church and ordained clergy. Eventually the Melitian bish­
ops could compete with their Catholic counterparts in numbers and influence, and monas­
teries arose that were connected with the Melitian separatists. The council of Nicaea
(325) saw a reconciliation brokered between the schismatics and Peter's successor,
Alexander, which recognized the legitimacy of the former's clergy; but hostilities erupted
again when the more stringent Athanasius assumed the see (328). In time, because of op­
position to Athanasius, the Melitians allied themselves politically with an Arian party,
thereby enabling him and their other enemies to brand them, for the most part unfairly,
as Arians themselves.94

The first of these dossiers mostly originates from or is addressed to Apa Paieous (Apa,
“father”), the head (or one of the heads) of a large Melitian community, the monastery of
Hathor in the Upper Cynopolite nome. In one of the letters, P.Lond. VI 1914, dated to 335
CE, we meet the formidable figure of Saint Athanasius and, as we would suspect, in no
positive light. The writer, Kallistos, a Melitian monk or cleric, recounts his and his col­
leagues’ sufferings at the hands of Athanasius and his circle. He begins his narrative with
Isaac, the bishop of Letous, an opponent of Athanasius and apparently a Melitian himself
or at least a sympathizer, who was dining at an army camp in Alexandria with Heraiscus,
who was likely the Melitian bishop of Herakleopolis Magna.95 When drunken partisans of
Athanasius try to seize Isaac, he escapes with the help of sympathetic soldiers, but
Athanasius's hit squad manages to assault and beat to the point of death four other
“brothers” who were coming into the camp. They continue their reign of terror by intimi­
dating and roughing up any in the area who were offering housing and refuge to the
Melitian monks, including Heracleides, manager of a local “lodge” (μονή).

The harassment of the Melitian brothers was getting on successfully, but, as Kallistos con­
tinues to report, things start to unravel. First comes the heartfelt repentance of the offi­
cer in charge, the praepositus castri, who aided Athanasius's gang, expressed in “a report
to the bishop: ‘I sinned and was drunken in the night, in that I maltreated the brethren’
And he also on that day put on a love feast [άγάπην],96 although he was a pagan [Ἕλλην],
because of the sin that he committed.” In addition, Athanasius receives word that a plot
to rescue one of his well-known supporters, a monk named Makarios, whom the emperor
himself had arrested, has failed, resulting in the further arrest of another ally, Archelaos.
One (p. 609) is tempted to hear a note of relish, as Kallistos remarks twice during his de­
scription, “Athanasius is getting very downhearted.”97 Any elation which he and other
Melitians may feel over Athanasius’ hard times is tempered by persisting difficulties of
their own. Despite the praepositus’ repentance, the brothers continue to be denied access
to Heraiscus (Hauben 1981, 453).

Other letters reveal the monks and monasteries in the more characteristic light as agents
and centers of intercession and healing. In P.Lond. VI 1926 (= Horsley 1987, 245–250),
from the second of the three archives, a woman, “Valeria in Christ,” addresses the monk
Papnouthios: “I beg and entreat you, most valued father, to ask for me [help?] from Christ

Page 17 of 35
The Papyri and Early Christianity

and that I may obtain healing; for by ascetics and the devout revelations are shown. For I
am beset with a great disease consisting of terrible shortness of breath. For so I have be­
lieved and do believe, that if you pray for me, I will receive healing. I beseech God, I be­
seech also you, remember me in your holy prayer. Granted that I have not come before
your feet in body, I have come to your feet in spirit.”

The need for healing from spiritual ills also finds its locus in petitionary letters to monks.
In P.Lond. VI 1917 a repentant sinner writes in Greek so fractured as to be almost incom­
prehensible: “To you then I write—Apa Paieou, that you may lift up your hands to our
Master God in the semblance of a cross … that God may [erase] the bond of my sins by
your most secure, most holy prayers.” He continues, “A transgression from the devil has
come upon me,” apparently in the form of violating a covenant made between him and
several monks. The writer seems further to describe his lapse as follows: “I abode, being
tempted, in the vegetable garden.”

Just what the sin, the covenant, and the garden were, the lacunae of the papyrus at this
point and the obscurity of the language do not reveal. We may think of the writer as mak­
ing a commitment (the διαθήκη?) to enter monastic life but being detained by the entice­
ments of the world (the garden?); this, however, is only a guess. At any rate he addresses
the great leader of the monastery:

By all means therefore beloved … night and day entreat the Lord of all—all those
who are in the Son, being in the Father, and he that is in the Father is in the Son—
that he may restore me into your hands … And not only did I write this, but I
wrote also … [he mentions other monks, including those with whom he had made
the covenant], that they may lift up most holy hands to God with all their hearts, in
the semblance of a cross, and may not cut me off … but … be compassionate and
merciful, being zealous on my behalf to God. By all means then beloved, write
from cell to cell [κατὰ μονήν] and to Apa Sourous and to Apa Pebe, that they also
may be merciful for me and may call upon God with zeal.

Despite his poor command of Greek, he reveals some knowledge of the scriptures and sensitivity
to “insider” buzz phrases and practices of the monks. In his penitent prayer, “that God may
[erase] the bond of my sins,” he invokes the dramatic Pauline metaphor of the cheirographon, to­
tally erased (not just crossed out) by divine mercy (Col. 2.14) and monastic intercession. His un­
grammatical description of the people of the Lord as “those who being in the Son are in the Fa­
ther, and the (p. 610) one who is in the Father is in the Son” recalls Johannine passages (e.g., Je­
sus’ description of his relationship with the Father in John 10.38 and 17.21) and may reflect the
manner in which the community spoke of itself. Similarly, the convention of praying with lifted
hands “in the semblance of a cross” may reveal insider knowledge of devotional practice.98 At
any rate, he feels that he has offended the entire community, and so his petition is community
wide. He asks Paieou not only for his prayers, but also to circulate his petition for forgiveness
and prayer “from cell to cell,” invoking the combined and concentrated intercession of the entire
monastic guild not only in Paieou's monastery but also in Melitian groups beyond it.99
The writers of these letters view the monks as the religious corollaries of the officials
with whom they were accustomed to file their petitions in epistolary form, imploring them

Page 18 of 35
The Papyri and Early Christianity

to redress their grievances. The phrase so familiar from those documents, “If you do this,
I will have received justice,” finds striking parallel in the words quoted earlier by the
asthmatic Valeria: “If you pray for me, I will receive healing.”

And so, as I hope the two parts of this chapter have demonstrated, the written word on
papyrus provided many early Christians of Egypt a vital conduit for both sides of the di­
vine-human dialogue. It bore the oracles of God in the form of the New Testament and the
inspired Septuagint; it also conveyed the far more fragile human word to God, as believ­
ers addressed their petitions to the monks as the nome- and village-level brokers of divine
mercy and spiritual power.

Acknowledgments
I am grateful to Peter van Minnen for reading an earlier draft of this article and offering
many helpful suggestions. I also thank J. Chapa and S. Charlesworth for allowing me ac­
cess to their unpublished articles.

Bibliography
Aland, K. 1967. Studien zur Überlieferung des Neuen Testaments und seines Textes. Ar­
beiten zur neutestamentlichen Textforschung. Vol. 2. Berlin: de Gruyter.

———. 1976. Repertorium der griechischen christlichen Papyri. Vol. 1, Biblische Papyri.
Berlin: de Gruyter.

———, and H.-U. Rosenbaum. 1995. Repertorium der griechischen christlichen Papyri.
Vol. 2, Kirchenväter-Papyri (Teil 1: Beschreibungen). Berlin: de Gruyter.

Artz-Grabner, P. 2002. “‘Brothers’ and ‘Sisters’ in Documentary Papyri and in Ear­

(p. 617)

ly Christianity.” Rivista Biblica 50: 185–204.

Bagnall, R. S. 1993. Egypt in Late Antiquity. Princeton: Princeton University Press.

Barns, J. W. B., and G. D. Kilpatrick, eds. 1957. “A New Psalms Fragment.” Proceedings of
the British Academy 43: 230–232.

Bell, H. I., and T. C. Skeat, eds. 1935. Fragments of an Unknown Gospel and Other Early
Papyri. London: Trustees of the British Museum.

Black, M. 1967. An Aramaic Approach to the Gospels and Acts, 3d ed. Oxford: Clarendon
Press.

Blanchard, A., ed. 1989. Les débuts du codex: Actes de lajournée dʼétude. Turnhout: Bre­
pols.

Bonner, C. 1934. A Papyrus Codex of the Shepherd ofHermas. Ann Arbor: University of
Michigan Press.

Page 19 of 35
The Papyri and Early Christianity

Boon, D. A. 1932. Pacomiana Latina. Louvain: Bibliothèque de la Revue dʼHistoire.

Boyaval, B. 1975. “Le cahier scolaire dʼAurelios Papnouthion.” ZPE 17: 225–235.

Bruce, F. F. 1990. The Acts of the Apostles: The Greek Text with Introduction and Com­
mentary, 3d ed. Grand Rapids, Mich.: Eerdmans.

Burkert, W. 1985. Greek Religion. Cambridge, Mass.: Harvard University Press.

Cameron, R. 1992. “Thomas, Gospel of.” In Anchor Bible Dictionary, ed. D. N. Freedman,
vol. 6, 535–540. New York: Doubleday.

Capasso, M., R. Savorelli, R. Pintaudi, and M. Gigante. 1990. Miscellanea Papyrologica in

occasione del bicentenario dellʼedizione della Charta Borgiana. Papyrologica Florentina
19. Florence: Gonnelli.

Cavallo, G. 1967. Richerche sulla Maiuscola Biblica. Florence: Edizioni Gonnelli.

Chadwick, O. 1968. John Cassian, 2d ed. London: Cambridge University Press.

Charlesworth, S. 2006. “Consensus Standardization in the Systematic Approach to Nomi­

na Sacra in Second- and Third-century Gospel Manuscripts.” Aegyptus 86: 37–68.

Comfort, P. W., and D. P. Barrett, eds. 2001. The Text of the Earliest New Testament Greek
Manuscripts, 2d ed. Wheaton, Ill.: Tyndale House.

Creed, J. L. (ed.). 1984. Lactantius: De mortibus persecutorum. Oxford: Clarendon Press.

Cribiore, R. 1996. Writing, Students, and Teachers in Graeco-Roman Egypt. American

Studies in Papyrology 36. Atlanta: Scholars Press.

———. 1999. “Greek and Coptic Education in Late Antique Egypt.” In Ägypten und Nubi­
en in Spätantiker und Christlicher Zeit, vol. 2, 279–286. Wiesbaden: Reichert.

———. 2001. Gymnastics of the Mind. Princeton: Princeton University Press.

Cross, F. L., and E. A. Livingstone. 1997. The Oxford Dictionary of the Christian Church,
3d ed. Oxford: Oxford University Press.

Crouzel, H. 1969. “La lettre du P.Harr. 107 et la théologie dʼOrigène.” Aegyptus 49: 138–
143.

Daniel, R. W., and F. Maltomini. 1990–1992. Supplementum Magicum, 2 vols. Papyrologi­

ca Coloniensia XVI.1–2. Opladen: Westdeutscher Verlag.

Delatte, A., and P. Derchain. 1964. Les intailles magiques Gréco-Éegyptiennes. Paris: Bib­
liothèque nationale.

Page 20 of 35
The Papyri and Early Christianity

Delgado, L. M. 2001. Léxico de magia y religión en los papiros máegicos griegos. Dic­
cionario Griego Español, Anejo 5. Madrid: Consejo Superior de Investigaciones Cientifi­
cas.

Devresse, R. 1970. Les anciens commentateurs grecs des Psaumes. Vatican City: Bibliote­
ca apostolica vaticana.

Di Berardino, A. 1992. “Fabian.” In Encyclopedia of the Early Church, ed. A. Di

(p. 618)

Berardino and A. Walford, vol. 1, 315. New York: Oxford University Press.

———, ed., and A. Walford, trans. 1992. Encyclopedia of the Early Church. 2 vols. New
York: Oxford University Press.

Emmel, S. 1996. “Greek Papyri in the Beinecke Library.” ZPE 112: 289–291.

Epp, E. 2002. “Issues in the Interrelation of New Testament Textual Criticism and
Canon.” In The Canon Debate, ed. L. M. McDonald and J. A. Sanders. Peabody, Mass. Hen­
drickson.

———. 2005. Perspectives on New Testament Textual Criticism. Leiden: Brill.

Freedman, D. N., ed. 1992. The Anchor Bible Dictionary. New York: Doubleday.

Frend, W. H. C. 1965. Martyrdom and Persecution in the Early Church: A Study of a Con­
flict from the Maccabees to Donatus. Oxford: Blackwell.

———. 1992. “Persecutions.” In Encyclopedia of the Early Church, ed. A. Di Berardino and
A. Walford, vol. 2, 671–674. New York: Oxford University Press.

Gagos, T., and R. Bagnall, eds. 2001. Essays and Texts in Honor of J. David Thomas. Amer­
ican Studies in Papyrology 42. n.p.: American Society of Papyrologists.

Gagos, T., and P. J. Sijpesteijn. 1996. “So-called ‘Orders to Arrest.’” BASP 33: 77–97.

Gamble, H. Y. 1992. “Codex.” In The Anchor Bible Dictionary, ed. D. N. Freedman. Vol. 1,
1067–1068. New York: Doubleday.

———. 1995. Books and Readers in the Early Church: A History of Early Christian Texts.
New Haven, Conn.: Yale University Press.

Gardner, I., A. Nobbs, and M. Choat. 2000. “P. Harr. 107: Is This Another Greek
Manichaean Letter?” ZPE 131: 118–124.

Ghedini, G. 1940. “La lettera PHar. 107.” Aegyptus 20: 209–211.

Gignac F. 1976, 1981. A Grammar of the Greek Papyri of the Roman and Byzantine Peri­
ods. Vols. 1 and 2. Testi e documenti per lo studio dellʼantichità 55.1 and 55.2. Milan: Isti­
tuto editoriale Cisalpino—La Goliardica.

Page 21 of 35
The Papyri and Early Christianity

Gribomont, J. 1992. “Psalms, Book of.” In Encyclopedia of the Early Church, ed. A. Di Be­
rardino and A. Walford, vol. 2, 722–723. New York: Oxford University Press.

Hagedorn, D. 1992. “P. IFAO 31: Johannesapokalypse 1.13–20.” ZPE 92: 243–247.

Harmless, W. 2004. Desert Christians: An Introduction to the Literature of Early Monasti­

cism. Oxford: Oxford University Press.

Hatch, W. H. P. 1936. “The Position of Hebrews in the Canon of the New Testament.” Har­
vard Theological Review 29: 133–151.

Hauben, H. 1981. “On the Melitians of P. London VI (P.Jews) 1914: The Problem of Papas
Heraiscus.” Proceedings of the XVI International Congress of Papyrology, ed. R. S. Bag­
nall, G. M. Browne, A. E. Hanson, and L. Koenen, 447–456. Chico, Calif.: Scholars Press.

———. 1998. “The Melitian Church of the Martyrs.” In Ancient History in a Modern Uni­
versity, ed. T. W. Hillard, R. A. Kearsley, C. E. V. Nixon, and A. M. Nobbs. Vol. 2, 329–349.
Grand Rapids, Mich.: Eerdmans.

———. 2004. “Heraiscus as Melitian Bishop of Heracleopolis Magna and the Alexandrian
See.” JJP 34: 51–70.

Head, P. 2005. “Is P4, P64 and P67 the Oldest Manuscript of the Four Gospels? A Re­
sponse to T. C. Skeat.” New Testament Studies 51: 450–457.

Healy, J. 1905. The Valerian Persecution: A Study of the Relations between Church and
State in the Third Century A.D. Boston: Houghton, Mifflin and Company.

Henner, J., H. Förster, and U. Horak. 1999. Christliches mit Feder und Faden:
(p. 619)

Christliches in Texten, Textilien, und Alltagsgegenständen aus Ägypten. Nilus Bd. 3. Vien­
na: Österreichische V.-G.

Hill, C. E. 2004. The Johannine Corpus in the Early Church. Oxford: Oxford University
Press.

Horsley, G. H. R., ed. 1981, 1982, 1983, 1987, 1989. New Documents Illustrating Early
Christianity, vols. 1–4. North Ryde, NSW: Ancient History Documentary Research Centre,
Macquarie University.

———. 2003. Η ΕΛΛΗΝΙΚΗ ΤΗΣ ΚΑΙΝΗΣ ΔΙΑΘΗΚΗΣ. Thessaloniki: Publication Depart­

ment of Aristotelian University of Thessaloniki.

Hunger, H. 1960. “Zur Datierung des Papyrus Bodmer II (P66).” Anzeiger der österre­
ichischen Akademie der Wissenschaften, phil.-hist. Kl. 4: 12–33.

Hurtado, L. 2006. The Earliest Christian Artifacts: Manuscripts and Christian Origins.
Grand Rapids, Mich.: Eerdmans.

Jellicoe, S. 1968. The Septuagint in Modern Study. Oxford: Clarendon.

Page 22 of 35
The Papyri and Early Christianity

Jeremias, J. 1963. “An Unknown Gospel with Johannine Elements.” In New Testament
Apocrypha, ed. E. Hennecke and W. Schneemelcher and trans. R. Mcl. Wilson, 94–97.
Philadelphia: Westminster.

Johnson, M. E. 1995. The Prayers of Serapion of Thumis. Orientalia Christiana Analecta

249. Rome: Pontificio Istituto Orientale.

Judge, E. A. 1977. “The Earliest Use of Monachos for ‘Monk’ (P. Coll. Youtie 77) and the
Origins of Monasticism.” Jahrbuch für Antike und Christentum 20: 72–89.

Judge, E. A. and S. R. Pickering. 1977. “Papyrus Documentation of Church and Communi­

ty in Egypt.” Jahrbuch für Antike und Christentum 20: 47–71.

Kelly, J. N. D. 1975. Jerome: His Life, Writings, and Controversies. New York: Harper &
Row.

Kenyon, F. C. 1934, 1936, 1937. The Chester Beatty Biblical Papyri, fasc. 3.1, Pauline
Epistles and Revelation, Text; fasc. 3, supp. 3.1, Pauline Epistles, Text; fasc. 3, supp. 3.2,
Pauline Epistles, Plates. London: E. Walker.

Knipfing, J. R. 1923. “The Libelli of the Decian Persecution.” Harvard Theological Review
16: 345–390.

Kyrtatas, D. J. 1987. The Social Structure of Early Christian Communities. New York: Ver­
so.

Laʼda, C. A., and A. Papathomas. 2004. “A Greek Papyrus Amulet from the Duke Collec­
tion with Biblical Excerpts.” BASP 41: 93–113.

Lampe, G. W. H. 1961. A Patristic Greek Lexicon. Oxford: Clarendon.

Lane Fox, R. 1986. Pagans and Christians. London: Penguin.

Layton, R. A. 2004. Didymus the Blind and His Circle in Late-antique Alexandria: Virtue
and Narrative in Biblical Scholarship. Urbana: University of Illinois Press.

Llewelyn, S. R. 1992, 1994, 1997, 2002. New Documents Illustrating Early Christianity,
vols. 6–7, North Ryde, NSW: Ancient History Documentary Research Centre, Macquarie
University; vols. 8–9, Grand Rapids, Mich: Eerdmans.

Lührmann, D. 2000. Fragmente apokryph gewordener Evangelien in griechischer und

lateinischer Sprache. Marburger Theologische Studien 59. Marburg: N. G. Elwert.

Luijendijk, A. 2005. “Fragments from Oxyrhynchus: A Case Study in Early Christian Iden­
tity.” Th.D. diss., Harvard University, Cambridge, Mass.

———. 2009. Greetings from the Lord: Early Christians in the Oxyrhynchus Papyri. Cam­
bridge, Mass.: Harvard University Press (forthcoming).

Page 23 of 35
The Papyri and Early Christianity

Martinez, D. G. 1991. P. Michigan XVI: A Greek Love Charm from Egypt. Atlanta: Scholars
Press.

Martinez, D. G. 1999. P. Michigan XIX: Baptized for Our Sakes: A Leather Trisa­
(p. 620)

gion from Egypt (P. Mich. 799). Stuttgart: Teubner.

Metzger, B. M. 1981. Manuscripts of the Greek Bible: An Introduction to Greek Palaeogra­

phy. New York: Oxford University Press.

———, and B. D. Ehrman. 2005. The Text of the New Testament: Its Transmission, Corrup­
tion, and Restoration, 4th ed. New York: Oxford University Press.

Moulton, J. H., and W. S. Howard. 1929. Grammar of New Testament Greek, vol. 2. Edin­
burgh: T. & T. Clark.

Musurillo, H. 1956. “Early Christian Economy: A Reconsideration of P. Amherst 3 (a).”

CdÉ 31: 124–134.

Naldini, M. 1998. Il Cristianesimo in Egitto: Lettere private nei papiri dei secoli II–IV, 2d
ed. Fiesole, Italy: Nardini.

Nautin, P. 1977. Origène: Sa vie et son œuvre. Paris: Beauchesne.

Navarra, L. 1992. “Valerian.” In Encyclopedia of the Early Church, ed. A. Di Berardino

and A. Walford, vol. 2, 860. New York: Oxford University Press.

Nelson, A. B. 1995. “The Classroom of Didymus the Blind.” Ph.D. diss., University of
Michigan, Ann Arbor.

Nongbri, B. 2005. “The Use and Abuse of P52: Papyrological Pitfalls in the Dating of the
Fourth Gospel.” Harvard Theological Review 98: 23–48.

Osiek, C. 1999. The Shepherd of Hermas. Hermeneia. Minneapolis: Fortress Press.

Paap, A. H. R. E. 1959. Nomina Sacra in the Greek Papyri of the First Five Centuries AD.
Papyrologica Lugduno-Batava VIII. Leiden: Brill.

Parker, D. C. 2000. “A New Oxyrhynchus Papyrus of Revelation: P115 (P. Oxy. 4499)”. New
Testament Studies 46: 159–174.

Parsons, P. J. 1970. “A School-Book from the Sayce Collection.” ZPE 6: 133–149.

Pestman, P. W. 1994. The New Papyrological Primer, 2d rev. ed. Leiden: Brill.

Pintaudi, R., ed. 1980. Miscellanea Papyrologica. Papyrologica Florentina 7. Florence:

Gonnelli.

Pöhlmann, E., and M. West. 2001. Documents of Ancient Greek Music: The Extant
Melodies and Fragments Edited and Transcribed with Commentary. Oxford: Clarendon.

Page 24 of 35
The Papyri and Early Christianity

Pohlsander, H. A. 1986. “The Religious Policy of Decius.” Aufstieg und Niedergang der
römischen Welt: Geschichte und Kultur Roms im Spiegel der neueren Forschung, vol. 2,
16.3: 1826–1842. Berlin: de Gruyter.

Porter, S. E. 2006. “Textual Criticism in the Light of Diverse Textual Evidence for the
Greek New Testament: An Expanded Proposal.” In New Testament Manuscripts, ed. T. J.
Kraus and T. Nicklas, 305–337. Leiden: Brill.

Rea, J. 1979. “P. Oxy. XXXIII 2673.22: Π ΥΛΗ Ν to ʽΥΛΗΝΙ” ZPE 35: 128.

Reynolds, L. D., and N. G. Wilson. 1991. Scribes and Scholars: A Guide to the Transmis­
sion of Greek and Latin Literature, 3rd ed. Oxford: Clarendon Press.

Rives, J. B. 1999. “The Decree of Decius and the Religion of Empire.” JRS 89: 135–154.

Roberts, C. H., ed. 1935. An Unpublished Fragment of the Fourth Gospel in the John Ry­
lands Library. Manchester: Manchester University Press.

———. 1979. Manuscript, Society, and Belief in Early Christian Egypt. Schweich Lectures
of the British Academy, 1977. London.

———, and T. C. Skeat. 1983. The Birth of the Codex. London: Published for the British
Academy by the Oxford University Press.

(p. 621) Roca-Puig, R. 1985. Dos Pergamins Biblics. Barcelona: Graphos.

Rosenbaum, H.-U. 1981. “Patristik und Papyrologie.” In Proceedings of the XVI Interna­
tional Congress of Papyrology, ed. R. S. Bagnall, G. M. Browne, A. E. Hanson, and L. Koe­
nen, 633–642. Chico, Calif.: Scholars Press.

Rousseau, P. 1985. Pachomius: The Making of a Community in Fourth-century Egypt.

Berkeley: University of California Press.

Sanders, H. A. 1935. A Third-century Papyrus Codex ofthe Epistles of St. Paul. Ann Arbor:
University of Michigan Press.

Schermann, T. 1912. Ägyptische Abendmahlsliturgien des erstens Jahrtausends. Studien

zur Geschichte und Kultur des Altertums 6. Paderborn.

Seider, R. 1967–1990. Palaographie der griechischen Papyri, vols. 1–3. Stuttgart: Hierse­
mann.

Sijpesteijn, P. J. 1980. “List of Nominations to Liturgies.” In Miscellanea Papyrologica, ed.

R. Pintaudi, 341–347. Papyrologica Florentina 7. Florence: Gonnelli.

———. 1990. “A New Part of P. Vindob.G 32016: List of Nominations to Liturgies.” In Mis­
cellanea Papyrologica in occasione del bicentenario dellʼedizione della Charta Borgiana,
ed. M. Capasso et al., 503–506. Papyrologica Florentina 19. Florence: Gonnelli.

Page 25 of 35
The Papyri and Early Christianity

Simonetti, M. 1992. “Tyre: Councils.” In Encyclopedia of the Early Church, ed. A. Di Be­
rardino and trans. A. Walford, vol. 2, 855. New York: Oxford University Press.

Skeat, T. C. 1997. “The Oldest Manuscript of the Four Gospels?” New Testament Studies
43: 1–34.

Staats, R. 1986. “Hernias.” Theologische Realenzyklopädie 15: 100–108.

Taubenschlag, R. 1955. The Law of Greco-Roman Egypt in the Light of the Papyri, 332
B.C.–640 A.D., 2d ed. Warsaw: Pañstwowe Wydawnictwo Naukowe.

Tjäder, J.-O. 1970. “Christ, Our Lord, Born of the Virgin Mary.” Eranos 68: 148–190.

Traube, L. 1907. Nomina Sacra. Munich.

Treu, K. 1973. “Die Bedeutung des Griechischen für die Juden im römischen Reich.”
Kairos 15: 123–144.

———. 1974. “Papyri und Patristik.” Kairos 16: 97–114.

Turner, E. G. 1977. The Typology of the Early Codex. Philadelphia: University of Pennsyl­
vania Press.

———. 1987. Greek Manuscripts of the Ancient World, 2d ed., ed. P. J. Parsons. London:
University of London, Institute of Classical Studies, Bulletin Supplement 46.

Van Haelst, J. 1976. Catalogue des papyrus littéraires juifs et chrétiens. Paris: Publica­
tions de la Sorbonne.

Van Minnen, P. 1994. “The Roots of Egyptian Christianity.” APF 40: 71–85.

———. 2001. “P. Harrauer 48 and the Problem of Papas Heraiscus in P. Lond VI1914.” Ty­
che 16: 103–105.

Veilleux, A., trans. 1980–1982. Pachomian Koinonia. 3 vols. Kalamazoo, Mich.: Cistercian
Publications.

Wasserman, T. 2006. “P78 (P. Oxy. XXXIV 2684): The Epistle of Jude on an Amulet?” In
New Testament Manuscripts, ed. T. J. Kraus and Tobias Nicklas. Leiden: Brill.

Whitehorne, J. E. G. 1977. “P. Oxy. XLIII 3119: A Document of Valerian's Persecution?”

ZPE 24: 187–196.

Wifstrand, A. 2005. Epochs and Styles: Selected Writings on the New Testament, Greek
Language, and Greek Culture in the Postclassical Era. Wissenschaftliche Untersuchungen
zum Neuen Testament 179.Tübingen: Mohr Siebeck.

Winter, F. 2000. “Frühes Christentum und Gnosis in Ägypten.” Protokolle zur Bibel
(p. 622)

9: 47–70.

Page 26 of 35
The Papyri and Early Christianity

Wipszycka, E. 1974. “Remarques sur les lettres priveés chrétiennes des iie–ive siècles (à
propos dʼun livre de M. Naldini).” JJP 18: 203–221.

———. 1996. Études sur le christianisme dans lʼÉgypte de lʼantiquité tardive. Studia
Ephemeridis Augustinianum 52. Rome: Institutum Patristicum Augustinianum.

———. 2001. “P.Coll.Youtie 77 = P.Col. VII171 Revisited.” In Essays and Texts in Honor of
J. David Thomas, ed. T. Gagos and R. S. Bagnall, 45–50. American Studies in Papyrology
42. n.p: American Society of Papyrologists.

Notes:

(1.) Ep. 28.2; Kelly (1975, 217 f). As Christians increasingly claimed the LXX as their sa­
cred text, Jews began to turn from it in favor of the Hebrew or other Greek versions (Jelli­
coe 1968, 74–99, esp. 74–76).

(2.) Homer, although a sacred text, never quite reached the level of “scripture”; the Or­
phica may have come closest (Burkert 1985, 297).

(3.) For Aramaic, Hebrew, and Septuagintal influence in NT Greek, see Black (1967);
Moulton and Howard (1929, 413–485); for Coptic influence on Roman and Byzantine doc­
umentary papyri see Gignac (1976, 46–48). Both nonliterary papyrus documents and the
different books of the NT display a broad range of linguistic competence and stylistic so­
phistication. We should avoid two often-made but misleading generalizations: first, that
the language of the NT is predominately “popular” or “vulgar” (see Wifstrand 2005, 71–
77), and second, that it evinces a distinctive dialect, sometimes called “Jewish
Greek” (see Horsley 1989, 5–40; 2003, 1–78).

(4.) The following list was compiled from the Leuven Database of Ancient Books. Included
are all copies from Egypt on papyri and other media (parchment, wood tablets, ostraca,
etc.). The list excludes material later than the ninth century.

(5.) See also the lists just for papyrus texts by Hurtado (2006, 19–24) (limited to the sec­
ond and third centuries) and Llewelyn (1994, 257–259).

(6.) Roberts (1979, 74–78), responding to Treu (1973, 138–144). Treu called these criteria
into question, as have others, notably R. Kraft; see his extensive online discussions and
material at http://ccat.sas.upenn.edu/rs/rak/earlylxx/jewishpap.html.

(7.) Turner (1977); Roberts and Skeat (1983); Blanchard (1989); Gamble (1995, 49–66);
Hurtado (2006, 43–93). Shorter treatments in Gamble (1992); Llewelyn (1994, 249–256);
Metzger and Ehrman (2005, 12–14).

(8.) Separate sheets of papyrus which were not part of rolls, or sheets of various animal
skins were also used for shorter texts, especially for Christian (and non-Christian) docu­
ments of the more “subliterary” type, such as amulets (for example, Daniel and Maltomini
1990, 53–112) and liturgical material (such as Martinez 1999) and for bits of literature for

Page 27 of 35
The Papyri and Early Christianity

educational purposes (see Cribiore 1996). Sometimes such sheets were reused, such as
the late third-century hymn to the Trinity P.Oxy. XV 1786 (van Haelst 1976, #962;
Pöhlmann and West 2001, 190–194), the earliest Christian hymn with musical annotation,
written on the verso of a list of grain deliveries. For a concise overview of the distinction
between the formats (codex, roll, and sheet) see W. Clarysse, Leuven Database of Ancient
Books, Help page, http://www.trismegistos.org/ldab/help.php, s.v. bookform.

(9.) P.Oxy.IV 656 (van Haelst 1976, #13) seems indisputable; more controversial are
P.Oxy. VII 1007 (van Haelst 1976, #5) and P.Berol. 17213 (van Haelst 1976, #15). See
Roberts 33 f., 76–78.

(10.) Traube (1907); Paap (1959); Roberts (1979, 26–48); Hurtado (2006, 95–134); S.
Charlesworth (2006). Shorter treatments in Metzger (1981, 36 f.); Metzger and Ehrman
(2005, 23 f).

(11.) Roberts's first group (1979, 27), with their nominative and genitive forms are given
here (abbreviating was done in all grammatical cases). Roberts has placed the remaining
eleven into two other groups as to the frequency and consistency of their abbreviation,
and it should also be said that in the earliest period both the words abbreviated and the
style of abbreviating were in flux (Bell and Skeat, 1935, 2–4; Roberts 1979, 27–28, 36–37).

(12.) On these views and in general, see Metzger (1981, 36–37).

(13.) Roberts (1979, 28–34). But see also Horsley (1983, 96; 1987, 189); R. Kraft provides
discussion and images at http://ccat.sas.upenn.edu/rs/rak/lxxjewpap/kyrios.jpg.

(14.) For example, consider the last sentence of Peter's famous Pentecost speech, deliv­
ered in Jerusalem, in Acts 2.14–36: “Let all the house of Israel know with certainty that
God has made Lord and Christ this Jesus whom you cruciied.” The sentence embodies in a
kind of creedal form (Bruce 1990, ad loc.) the four most frequently abbreviated words and
includes two others often abbreviated (Ἰσραήλ and σταυρός in the verb ἐσταυρώατε). The
entire Pentecost speech comprises most of the Wfteen significant words.

(15.) See van Haelst (1976, #151); Aland (1976, #AT68).

(16.) Other Psalm papyri that have been dated to the second century or early third, with
their van Haelst (1976) numbers, are as follows: P.Ant. I 7 (#179), P.Bodm. XXIV (#118),
P.Leipz.inv. 170 (#224), PSI VIII 121 (#174). Not in van Haelst: PSICongr.XX 1; P.Barc.
inv. 2 (Roca-Puig 1985, 7–16).

(17.) Praecepta, 139–140 (Boon 1932, 49 f.; cf. Veilleux 1981, vol. 2, 166); Rousseau 1985,
81n20.

(18.) In Pachomian literature see Praecepta 8, 15, 142 (Boon 1932,15 f., 50 f.; Veilleux
1981, vol. 2,146 f., 166). For John Cassian's description (Inst. II 4) of the Egyptian monas­
tic practice of twelve psalms at vespers and twelve at nocturns, see Chadwick (1968, 58).
For the office of the ψάλτης see Wipszycka (1996, 248–251).

Page 28 of 35
The Papyri and Early Christianity

(19.) See in general Devreesse (1970) and Martinez (1999, 6n14).

(20.) On the fragments of this work see Nautin (1977, 262–275). For Origen's later ex­
egetical work on Psalms, see Gribomont (1992, vol. 2, 722).

(21.) For these and the controversy as to their authorship see ibid.

(22.) For Greek education in Egypt in the Hellenistic, Roman, and later periods see
Cribiore (1996, 2001).

(23.) For Homer see Cribiore (1996, 49; 2001, 194–197, 204–205). Neither Homer in par­
ticular nor classical literature in general were ever supplanted, even as Christians began
to introduce biblical material in the curriculum (Reynolds and Wilson 1991, 48–51; Hen­
ner, Förster, and Horak 1999, 51n31).

(24.) Van Haelst (1976, #109); Cribiore (1996, #297).

(25.) P.Lond.Lit. 255 = Cribiore (1996, #298).

(26.) Reynolds and Wilson (1991, 49–50). The Bodmer and Chester Beatty papyri may
have come from a single library at Panopolis, possibly the property of a school where clas­
sical and Christian authors were read (Bagnall 1993, 103–104, and cited lit., esp. in
n382). For similar combinations of the Psalms and classical literature, see the school
notebooks published by Boyaval ( 1975 = Cribiore 1996, #396; IV CE) and Parsons (1970
= Cribiore 1996, #388; late III CE). The last-mentioned text also illustrates a culture of
bilingualism (Greek and Coptic) in Christian education in late-antique Egypt (Cribiore
1999, 281–282; Bagnall 1993). Also in this regard see the notebook MPER NS IV 24 (=
Cribiore 1996 #403; Henner, Förster, and Horak 1999 #42; IV/V CE).

(27.) Laʼda and Papathomas (2004, esp. 107–110); Wasserman (2006, 149 n. 41); Daniel
and Maltomini (1990, 73).

(28.) For the controversy see Emmel (1996, 290). Conversely, Pestman (1994, 31–32) ap­
proves of Welles's dating.

(29.) In the upper margin of the recto, above the second column of the letter, is Hebrews
1.1 (van Haelst 1976, #536). On the reuse of papyri for school texts and the complicated
issue involving the “front” and “back” of these documents, see Cribiore (1996, 60–62).

(30.) Gamble (1992, 1068; 1995, 63); Skeat (1997, 31).

(31.) If Skeat (1997) is correct that the late second-century fragments P4, P64, and P67
constitute our earliest gospel codex, it may also have displayed the Western order (p. 19;
Metzger and Ehrman 2005, 53). Peter Head (2005), however, has called into question
Skeat's understanding of these fragments.

(32.) For example, PGMP 19 and BKT 6.7.1 (= van Haelst 1976, 731; Laʼda and Papath­
omas 2004, #10).

Page 29 of 35
The Papyri and Early Christianity

(33.) P.Ryl. Ill 457 = van Haelst (1976, #462).

(34.) Namely F. G. Kenyon, W. Schubart, H. I. Bell, A. Deissmann, U. Wilcken, and W. H. P.

Hatch (see intro. to P.Ryl. III 457; Metzger and Ehrman 2005, 56).

(35.) Nongbri (2005, passim, esp. 30n22), and cf. Turner (1977, 3 f).

(36.) Other very early papyrus copies of John include P.Oxy.L 3523 (P90), late second cen­
tury (Llewelyn 1994, 242–248), and Papyrus Bodmer II (P66) (= van Haelst 1976, #426;
Metzger and Ehrman 2005, 56–57; Comfort and Barrett 2001, 376–468). For the latter,
the early third-century date assigned by the original editors is confirmed by Turner
(1987, 108, #63). Earlier dates in the second century have been argued by Hunger (1960,
12–33); Cavallo (1967, 23); and Seider (1970, 121).

(37.) The Michigan leaves were published by Sanders (1935); the entire manuscript, both
Michigan and Dublin leaves, was published with an edition of plates by Kenyon (1934,
1936, 1937).

(38.) For the Pauline character of Hebrews and this position in the order of the epistles,
see Metzger and Ehrman (2005, 55); Hatch (1936).

(39.) Bonner (1934, 7–12); Metzger and Ehrman (2005, 54–55).

(40.) For instance, Wilcken (around 200) and Kenyon. Others date it earlier or later (Com­
fort and Barrett 2001, 204–206).

(41.) For instance, Roberts, Bell, and Skeat (Roberts 1979, 13; Roberts and Skeat 1983,
40 f.; cf. Comfort and Barrett, 2001, 135). Others, however, prefer a date in the third cen­
tury (see van Haelst 1976; Aland 1976, #NT 32).

(42.) On the general point see Hurtado (2006, 35–40).

(43.) This point was argued by Prof. Juan Chapa in a paper he presented at the 25th Inter­
national Congress of Papyrology in Ann Arbor, Michigan, in the summer of 2007. I have
depended on Chapa's work for much of the material in this paragraph. See also his pref­
ace to the just published John fragments, P.Oxy. LXXI 4803–4806.

(44.) This was likely not the case either in Egypt or elsewhere; see Hill (2004, esp. 148–
166, 289–293). Earlier C. H. Roberts questioned the extent of Gnostic influence in early
Egypt on a number of grounds, including that pre-fourth-century papyrus texts yield scant
evidence for the movement (1979, esp. 52, 72–73).

(45.) Llewelyn (1994, 260–262) also tabulates evidence that shows that these books were
the two most frequently quoted in patristic writings.

(46.) Apud Eusebius hist. eccles.VI 14.7. I thank Juan Chapa for this reference (see note
43 above).

Page 30 of 35
The Papyri and Early Christianity

(47.) See P.Oxy. IV 654, 655 = van Haelst (1976, #593, 595). For the three P.Oxy. Gospel
of Thomas fragments, see Lührmann (2000,106–129). For their physical properties see
Hurtado (2006, 81–83). For other P.Oxy. Apocryphal fragments, see Epp 2005, 511 f.

(48.) Van Haelst (1976, #586); Llewelyn (2002, 99–101); Lührmann (2000,142–153).

(49.) Turner (1987, 11n50, 108). There are, however, a few earlier examples, and Turner
himself has not called the second-century date of this papyrus into question (1977, 3 and
tables on 144).

(50.) In general see Aland and Rosenbaum (1995); Rosenbaum (1981); Treu (1974).

(51.) In general see Horsley (1987, 196–198); van Haelst (1976, 229 f. on #643).

(52.) Aland and Rosenbaum (1995, 408–413, 452–460, 467–516); van Haelst (1976, #683–
687); Horsley (1987,196).

(53.) Aland and Rosenbaum (1995, 57–162); Nelson (1995, vi–vii, 203–204); cf. Horsley
(1987, 196–197); van Haelst (1976, #643–647).

(54.) For Didymus's pedagogy and the educational context in which he worked, cf. Nelson
(1995, esp. ch. 1); Horsley (1987, 197); Layton (2004, 13–35).

(55.) Among the Greek authors cited in these two commentaries are Aristotle, Democri­
tus, Homer, Isocrates, Plato, Pythagoras, Solon, and Zeno. See Nelson (1995, 86–119,
187–188); Horsley (1987,197).

(56.) There is, however, a graffito from the sixth–seventh century that preserves forty-one
fragmentary lines of his Epistula ad monachos (van Haelst 1976, #625).

(57.) P.Köln VII 297 to be used with Aland and Rosenbaum (1995, 40–46, KV 9a). For an­
other papyrus doubtfully attributed to him, see Aland and Rosenbaum (1995, 39, KV 9);
van Haelst (1976, #636).

(58.) See Gonis's introduction to these texts; Aland and Rosenbaum (1995, 232–310, KV
29–42); van Haelst (1976, ## 655–668). The rich papyrological evidence meshes with the
inclusion of The Shepherd in the Codex Sinaiticus and its esteem by Clement and Origen
in showing how important this book was in Egypt (Staats 1986,107).

(59.) One of the earliest of the Hermas papyri, P.Mich. II 130 = van Haelst 1976, # 657
(ed. late II, approved by Roberts 1979, 14, questioned by Aland and Rosenbaum 1995,
278n3), has features such as the use of lectional signs and the reuse of a documentary
text, which may point to school use (Aland and Rosenbaum 1995, 276, 279n7).

(60.) As to Revelation, among the oldest are P.IFAO II 31 (Hagedorn 1992), late II, early
III; P.Chester Beatty III (van Haelst 1976, #495), early III; P.Oxy. LXVI 4499, late III, early
IV (see Parker 2000).

(61.) Lane Fox (1986, 265–268); Kyrtatas (1987, 175–178); van Minnen (1994, 77).
Page 31 of 35
The Papyri and Early Christianity

(62.) On this point see Epp (2005, 761–763). The earliest Christian letters are from the
third and fourth centuries. Attempts to identify various second-century or second/third-
century letters as Christian (Naldini 1998, #1–4) have run afoul of inconclusive evidence.
On this point and for the difficulties in determining criteria for designating letters as
Christian see Wipszycka (1974); Llewelyn (1992, 169–177, on P.Oxy. XLII 3057).

(63.) See van Haelst (1976, #1194); Naldini (1998, #5).

(64.) For a good survey of scholarly views on the text see Winter (2000, 48–54).

(65.) Crouzel (1969, 138–140), who also connects the idea with Origen, but see Naldini
(1998, 427).

(66.) Namely, another Christian letter, P.Oxy. VIII 1161 (IV) = Naldini (1998, #60), and a
Christian amulet (Daniel and Maltomini 1990–1992, #30). On these (and the document
under discussion) see Horsley (1981, 102 f).

(67.) For example, Seider (1967 #41, Tafel 25).

(68.) Didymus frag. in Pss. (Ps. 30.6) 270, 22. Also LXX Esdra 4.40; cf. Basil adv. Eunomi­
um V (PG 29, 757B): “Blessed be the God of truth, who is the Father of the Truth, Christ.”

(69.) That is, Serapion of Thmuis, a fierce opponent of the Manichaeans. The prayers,
however, likely had multiple authors. For this controversial document see Johnson (1995,
esp. 154 f, 159 f for “God of truth”). Johnson suggests that the phrase “may serve as a
characteristic expression of an editor of the text” (155). For other liturgical and hymnic
examples, see P.Dêr Balizeh, fol. 1, verso 14 (van Haelst 1976, #737); Romanos cant. 25
κβ′ 1; Ghedini (1940, 210).

(70.) On “soul, body, and spirit” see Naldini (1998, 427); Horsley (1981, 102 f; 1987, 38 f).
For magical texts see Martinez (1991, 90 on line 36 f). On (δια)ϕνλάσσειν see Delatte and
Derchain (1964, index s.vv.); Delgado (2001, s.vv.).

(71.) For Πάσχα as “Easter,” cf. Lampe s.v. For the Manichaean practice of the festival, cf.
Gardiner, Nobbs, and Choat (2000, 123 f). The request for the garment maybe routine; it
could be connected with baptismal ritual, which was practiced on Easter in the early
church (see the description of the baptized as “clothed in the garment of salvation, Jesus
Christ,” Cyr. H. catech. 19.10; Lampe 1961 s.v. ἱμάτιον).

(72.) That is, if what we know about lectors elsewhere may be applied to Egypt; see
Lampe (1961, s.v. ἀναγνώστης 2); Wipszycka (1996, 238–248); Naldini (1998, 428 f).

(73.) P.Herm.Landl.; see Horsley (1983, 156 f); Bagnall (1993, 292).

(74.) For a full discussion see Luijendijk (2005, 173 f.; also forthcoming 2009).

(75.) Healy (1905). For shorter accounts see Parsons (ad loc.); Navarra (1992).

Page 32 of 35
The Papyri and Early Christianity

(76.) Such a description, the editor, Parsons, notes, could indicate that Petosorapis “did
not conceal his religion and indeed could be identified by it” by the inhabitants of his vil­
lage. Luijendijk (2005, 225–227; 2009 forthcoming), however, suggests that, since the
word “Christian” occupies the position we would normally expect the name of an occupa­
tion to hold in a list of this kind, it may be that the word is a type of professional designa­
tion, i.e., “Christian clergyman” (see also her discussion of Sotas the Christian [2005,172–
174]).

(77.) This text was published by P. J. Sijpesteijn (1980), as was an additional part of it by
the same editor in 1990. For an important discussion see van Minnen (1994, 73–77).

(78.) In the left margin of the papyrus by each entry a different writer has written a num­
ber by each name.

(79.) Rives (1999); Pohlsander (1986); Frend (1965, ch. 13); more briefly, Horsley (1982,
180–185).

(80.) Horsley (1982, 183); Di Berardino (1992); Cross and Livingstone (1997, 594).

(81.) The translation is that of the editors, A. E. R. Boak and J. G. Winter. Most of the libel­
li have been collected by Knipfing (1923); of these, the present text (with, however, mis­
readings) is no. 35 (p. 385 f). It has also been published with a brief introduction and
commentary by Pestman (1994, 236 f, no. 62).

(82.) For example, Cyprian of Carthage in the Epistulae and de lapsis, and the fragments
of Dionysius, bishop of Alexandria; for specific references and discussion see Horsley
(1982, 183–185).

(83.) For the scheme of the edicts, our sources for them, and further bibliography see
Cross and Livingstone (1997, 483, s.v. Diocletian, and 1258, s.v. persecutions); Frend
(1992, 673).

(84.) The text of this edict has not survived and must be reconstructed from various
sources (Frend 1965, 495 f; 1992, 673; Creed 1984, xxii–xxiii, with comments on Lactant.
de mort. pers. 13, 94).

(85.) Reading at line 22 ὕλην instead of πύλην, which is printed in the original edition; see
Rea (1979).

(86.) On this text see Pestman (1994, 254).

(87.) Ibid., #69; Naldini (1998, #35).

(88.) Neither occurs in papyri before the fourth century. However, Ϟθ has literary attesta­
tion that dates it as early as the late second century (see the editor on this text, line 34),
but it is unusual in Christian letters. For χμγ, which has frustrated attempts at explana­
tion, see especially Tjäder (1970); Horsley (1982, 177–180); and Llewelyn (1997, 156–
168). Other Christian symbols in papyrus documents include the labarum ; staurogram

Page 33 of 35
The Papyri and Early Christianity

( . see Llewelyn 1997, 166n39; Hurtado 2006, 135–154), and various forms of the Chris­
tian cross.

(89.) See the editor on line 5 and Turner's comment, which he cites here: “One must bear
in mind that in private letters the use of a nomen sacrum is surely always imitative, and
the reason for writing it very likely unknown to the writer—unless he is himself a reader
of good written texts (NT, etc.) or a professional scribe of such.” See also Roberts (1979,
27 f); Pestman (1994, 32).

(90.) In general, see P. Parson's introduction to this text, Taubenschlag (1955, 505–508),
and the introduction to P.Oxy. XIV 1642, where the one given power of attorney is also
called “brother.”

(91.) The word can be a near equivalent to “colleague”; Arzt-Grabner (2002, 187).

(92.) See the general treatments and bibliography in Bagnall (1993, 293–303); van Min­
nen (1994, 78–85). See also the essays collected in part III of Wipszycka (1996, 279–403).
The first papyrus to mention a monk is P.Coll. Youtie II 77, republished as P.Col. VII 171
(324 CE), in which the monk Isaac, along with a deacon Antoninos, helps a certain Aure­
lius Isidoros survive an assault; on this papyrus see Judge (1977; idem in Horsley 1981,
124–126, #81); Wipszycka (2001).

(93.) Most of the texts in the first two archives are edited in P.Lond. VI; the third in
P.Neph. For details see Bagnall (1993, 308); van Minnen (1994, 79 f).

(94.) For a fuller account of the Melitian schism and church, see Bell in P.Lond. VI 38–45;
more recent discussion and bibliography in Bagnall (1993, 306–309); Hauben (1998), and
on doctrine and orthodoxy especially 334–346; Harmless (2004, 35, 94–95).

(95.) Van Minnen (2001) made this identification, on which see Hauben (2004).

(96.) For the love feast as possibly distinctively Melitian, see Hauben (1998, 342 f).

(97.) Indeed, his troubles multiplied. The same year Athanasius was summoned to Tyre to
answer for his ruthless conduct against the Melitians and his insubordination, most likely
in part based on his and his followers' activities described in this letter. He was stripped
of his ecclesiastical position and, after an unsuccessful appeal to Constantine, exiled to
Trier (Simonetti 1992, vol. 2, 855).

(98.) The Pachomian community also assumed this stance while reciting the paternoster
(Harmless 2004, 128; I thank Nick Marinides for this reference).

(99.) Some of the monks he asks to pray for him are in the “Upper Country,” possibly the
Thebaid, the editor, Bell, conjectures (81).

David G. Martinez

Page 34 of 35
The Papyri and Early Christianity

David G. Martinez is Associate Professor, Department of Classics and the Divinity

School, University of Chicago.

Page 35 of 35
Manichaeism and Gnosticism in the Papyri

Manichaeism and Gnosticism in the Papyri

Cornelia Römer
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Classical Religions and Mythologies, Classical Philosophy, Greek
and Roman Papyrology
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0026

Abstract and Keywords

The church fathers were appalled in particular by the Gnostics' condemnation of creation.
But the fact that much of their teaching was in many respects not so far from Christian
dogma must have disturbed the advocates of the “real” Christian church. In some of these
Gnostic systems, Christ was the main savior figure; in others, it was the forefathers of the
Old Testament who guaranteed salvation; in Manichaeism, it was the new Messenger of
Light, the apostle Mani, who, coming after Christ, would finally give the right revelation
to the people and excel Christ in doing so. This article deals with religious groups such as
these as they existed in Egypt in the Roman and late antique periods. Papyrology has
played a decisive role in our understanding of the religious movements of the first cen­
turies CE in Egypt and elsewhere in the Mediterranean.

Keywords: Gnostic systems, Christ, papyrology, creation, Mani, church fathers, Old Testament

PAPYROLOGICAL finds have enriched our knowledge not only of early Christianity (chap­
ter 25) but also of some religions outside the Christian church. The rediscovery, editing,
and translation of the holy books of these religions fifteen hundred years or so after they
had to be hidden from the eyes of government and church officials have aroused wide
scholarly interest and even encouraged or generated contemporary religious movements
or currents, both inside and outside Christianity

With Constantine's adoption of Christianity in 312 CE and its acquisition of official status
after he completed his domination of the Roman Empire in 324, the church began a long
process through which imperially sanctioned councils not only determined what constitut­
ed Christian orthodoxy and what was heretical but could use state power against those
deemed heretics. A variety of other religious movements that had flourished in Egypt and
elsewhere around the Mediterranean Sea in the first three centuries CE, many still or
newly popular in the fourth century, came to be considered heretical Their adherents
were persecuted, their books were sometimes burned, and it is only by chance that we
have any texts or artifacts remaining from those groups, whose members would in many
Page 1 of 19
Manichaeism and Gnosticism in the Papyri

cases have called themselves Christians. For a long time, our knowledge of other nonpoly­
theistic religions and religious groups besides Christianity and Judaism derived from the
writings of the church fathers, whose views were naturally partisan Epiphanius, (p. 624)
who was bishop of Salamis on Cyprus in the fourth century, wrote three books titled Pa­
narion—which means “medicine chest”—in which he lists about eighty heretical groups
from Adam down to his own time and warns right-thinking Christians about the immoral
practices and aberrant beliefs of Valentinians, Bardesanians, and Manichaeans, to name
just a few. His “medicine chest” was supposed to contain effective remedies against the
dangers of those heretical beliefs. Some names of the leaders of these groups are still on
the index of the Vatican today even though their beliefs have been consigned to history.

What made these beliefs different from the Christian creed was more than the different
way in which they understood the relationship of humankind to the one God or how the
three parts of the Holy Trinity can be understood in a monotheistic system as a single
God. Questions like these had troubled the early Christian church for quite some time;
the church fathers' answers attempted to explain the secrets of the Holy Trinity by using
the means and thought patterns of Greek philosophy. Fights over the relationship of Fa­
ther and Son culminated in the struggle over the teachings of Arius, who was deacon in
Alexandria in the early fourth century CE. He preached that Christ could not be equal to
God but had to be a creation; otherwise, the One God would have suffered because he
created his son from himself.

Figure 26.1. Mani, drawing by F. C. Burkitt

based on a wall fresco found at Khotscho near
Turfan, from F. C. Burkitt, The Religion of the
Manichees (1925).

The Gnostics and other “heretical” groups whom Epiphanius and other church fathers
talk about are different. Their concepts of God and salvation were less a result of sophisti­
cated philosophical strategies for explaining the world and God than a mixture of various

Page 2 of 19
Manichaeism and Gnosticism in the Papyri

traditions, including Jewish, Egyptian, and Persian ones, which were believed to make sal­
vation possible. These “heretics” may have believed in Christ as the savior of all humani­
ty, but they could not believe in the uniqueness of Christ. Even though many of them
claimed Christ as their savior, they still had a different understanding of what led to sal­
vation. The apostle Paul had stressed in his letter to the Romans (10, 9–13) that it was
faith that allowed people to receive the blessing of salvation. Gnostics believed that it was
gnosis, or recognition of our situation in the world. In their view, the human condition
was not embedded in the wonderful creation of the One God, which is what “real” Chris­
tians believed. For them, creation was both alien and hostile to the human race. They re­
jected the Old Testament and its God. Gnostics saw themselves and the entire world as an
entity of two parts, of which one, the body, was dark and mortal, and the other, which
contained the light, consisted of sparks from the realm of light from which people and all
organic material come and must finally return. Every human being was thus involved in
the final task of the huge machinery to recreate a status in which the dark parts on the
one side and the light parts on the other would one day exist in two completely different
entities and thus recreate the beginning of the world before the deadly mixture came to
be in people and all organic beings. In this scenario, Christ was only one of the messen­
gers sent from the realm of light to save humanity (but not creation) and to teach it about
the final aim of the separation of light and darkness, not only the salvation of each indi­
vidual. (p. 625)

The basic structure of these beliefs, which united the different gnostic groups, led to a
wide variety of practical approaches on how to achieve the final aim. The church fathers,
such as Epiphanius in his Panarion and others, were appalled in particular by the Gnos­
tics' condemnation of creation. But the fact that much of their teaching was in many re­
spects not so far from Christian dogma as the church fathers wanted their followers to
believe must also have disturbed the advocates of the “real” Christian church. In some of
these Gnostic systems, Christ was the main savior figure; in others it was the forefathers
of the Old Testament who guaranteed salvation; in Manichaeism it was the new Messen­
ger of Light, the apostle Mani, who, coming after Christ, would finally give the right reve­
lation to the people and excel Christ in doing so (figure 26.1).

As in Christianity, the ideas of these religious groups, too, show the deep uncertainty
about the human situation in the world, which was characteristic around the turn of the
millennium, and reflect attempts to give the world a new (p. 626) structure so as to envis­
age new strategies for salvation. It seems that the whole Mediterranean basin was at that
time convinced that this world was not the best but that a new, different, and better world
would come, whether on the Last Day, as the Christians believed, or at the end of this
world, when light and darkness would be separated forever. The role of creation as such
was important to both views of the world; for Christians, creation was God's gift to hu­
man beings to use for their benefit during their mortal lives; for the Gnostics, creation
was the machinery that enabled people to overcome the world in its present state be­
cause only through the mixture of light and darkness in the world and in themselves
would they be able to separate the two by making the mind triumph over the body.

Page 3 of 19
Manichaeism and Gnosticism in the Papyri

This chapter deals with religious groups such as these as they existed in Egypt in the Ro­
man and late antique periods. (Traditional Egyptian religion, still vital in the early cen­
turies of the Christian era, is discussed in chapter 24.)

Papyrology has played a decisive role in our understanding of the religious movements of
the first centuries CE in Egypt and elsewhere in the Mediterranean. Because of the dis­
coveries of papyri, we have had a chance not only to look into the original writings of
some of these “heretical” groups and see them functioning in their social contexts but al­
so to learn from private letters how their followers behaved in their proper religious and
social environment. Papyri and parchments that contain both the literary texts in which
these groups believed and a variety of documents allow us to create a new picture of the
“heretics” and their beliefs beyond that offered by Epiphanius and his Panarion.

The Gnostic Writings from Nag Hammadi,

Oxyrhynchus, and the Area of Panopoiis
In late 1945 an Egyptian farmer found a large jar that had been abandoned near a cliff
close to his village. The name of the village was Nag Hammadi, situated on the left bank
of the river Nile, somewhat south of Abydos, at the point where the river bends toward
the east at an angle of 90 degrees. The jar contained thirteen ancient books, manufac­
tured from papyrus and covered by leather bindings that were still intact (figure 26.2).
The veracity of this account of the discovery of the Nag Hammadi Codices depends en­
tirely on what the finder of the jar told the police and scholars; it is up to us to decide
how much of it we want to believe.

Figure 26.2. The Nag Hammadi Codices in their

original covers.

The thirteen codices arrived on the Cairo antiquities market, and all but one, the Codex
Jung, which was first purchased by the Zurich-based foundation of the (p. 627) famous
psychoanalyst, were incorporated into the holdings of the Coptic Museum in Cairo. The
Codex Jung has in the meantime returned to Cairo as well.

The codices of this find are all written in the Coptic language. Dated documents that were
torn and recycled to strengthen the leather bindings of the books show that the codices
were bound and most likely copied some time after the middle of the fourth century CE.

Page 4 of 19
Manichaeism and Gnosticism in the Papyri

The texts themselves are older, although how much older is difficult to ascertain; there is,
however, no doubt that they were originally written in Greek and were translated into
Coptic at some point in the later third or early fourth century CE. Even more difficult to
establish is when they were first created in Greek, but it is unlikely that any of them
(apart from a fragment of Plato's Republic) goes back beyond the first century CE.

Why did these books end up in a jar near a cliff? The explanation of this riddle might, it
seems, be connected to their content. Most of the writings, fifty-one treatises altogether,
some of which are in several different versions, are Gnostic in the sense described earlier.
The assumption is that these books were once part of a library for which orthodox Chris­
tians were searching and that they were therefore hidden from the searchers. This is es­
pecially likely because some of the papyrus pieces that reinforce the bindings clearly
come from a monastery. The fact that the famous monastery of Pachomius was situated in
the village of Chenoboskion, not far from Nag Hammadi, has made this idea even more
enticing to many scholars. Were these books once owned by the monks of the earliest
monastery in Egypt, founded by Pachomius in 325? This idea, which has attracted both
positive and negative reactions from scholars, can neither be proved nor refuted. Pa­
chomius (and probably most of his followers) spoke Coptic, but we know from his biogra­
phy that Greek-speaking men from Alexandria joined the community in Upper (p. 628)
Egypt. Pachomius is said to have learned Greek himself but expected the newcomers to
learn Coptic (Vita in Greek, 361). The Greek texts may have been brought upriver by
those who wanted to join and been translated for use in the monastery. Against this as­
sumption is the fact that the Sahidic dialect in which these texts are written is not uni­
form, which seems to indicate that the translation was done in different places. The ques­
tion remains whether the monks of a monastery would have been open minded enough to
keep these texts in their library—or, rather, sophisticated enough to appreciate the
codices' contents. When we look at catalogues of monastic libraries as they are preserved
in papyri from late antiquity (Otranto 2000), we find that these lists contain only volumes
of the Old and New Testaments, legends of saints, and writings of the church fathers but
no works whose titles might reveal heretical content. But those lists all come from the
fifth or sixth centuries, when the uniformity of the Christian creed had been ordained and
sealed, and were compiled for official use; they may have omitted titles not considered
appropriate for reading by adherents of the orthodox beliefs of the official Christian
church. It is more likely, however, that those later libraries had been cleansed of every­
thing not orthodox (the “heretical” volumes having been removed to caches outside the
monasteries, as may have happened in the case of the Nag Hammadi Codices). The ques­
tion of the ownership of the Nag Hammadi Codices thus remains open.

Whoever collected these books must have had an interest in the explanation of the world
as the Gnostics saw it. He must also have been interested in secret writings, like those so-
called Hermetic texts in which the god Hermes Trismegistus explains the world (Codex VI
6, 7). The fact that a (bad) Coptic translation of a small part of Plato's Republic (588b–
589b) also fills several pages of one of the codices makes their owner still more interest­
ing to us (Codex VI 5).

Page 5 of 19
Manichaeism and Gnosticism in the Papyri

The Apocryphon of John occupies a prominent place among the other treatises. It appears
in three different versions in the codices of Nag Hammadi (Codex II, 1; III, 1; IV, 1; figure
26.3). A fifth-century Coptic codex housed in the papyrus collection in Berlin, which was
known long before the Nag Hammadi library was found (Codex 8205), also contains a ver­
sion of this treatise. It must have been very popular in Gnostic circles; Irenaeus, the bish­
op of Lyon, probably knew of this text when he composed his antiheretical book, Contra
Haereses, in 185 CE.

Figure 26.3. Nag Hammadi Codex II, 28–29,

Apocryphon of John.

According to its title, the Apocryphon of John presents itself as a secret revelation of John,
the son of Zebedee, who meets the resurrected Christ far away from all human dwellings
in the desert. This setting for a secret encounter has numerous precedents in Jewish reve­
lation literature. What Christ tells John is a mythological narrative about the creation, the
Fall, and the salvation of humankind. He explains the existence of evil in the world and re­
veals a way in which people can escape this unworthy condition, which came about
through the mixture of light and darkness. Much of the content is phrased in the vocabu­
lary of Greek philosophy. God is perfect and has no connection to the evil world. “He is
the invisible Spirit; it is not (p. 629) right to think about him as a god or something similar.
For he is more than a god, since there is no one above him, nor does anyone lord it over
him. He exists in nothing inferior, for everything exists in him” (Codex II, 1–3; NHLE 100).
From this perfect entity emanate Christ and Sophia, who encourage humans to return to
the original condition, which existed before light and darkness were mixed. The female
entity, Sophia, plays an important role, as do female entities in general in Gnostic sys­
tems. Evil comes into the world when Sophia disobeys God and brings forth the monster
Yaldabaoth, who—as himself a mixture of the light particles inherited from his mother
and earthly parts of darkness—creates the world. From here, the struggle between light
and darkness begins, which is the condition of humankind, according to all Gnostic sys­
tems. Finally, Christ is sent to save humanity by reminding people of their heavenly ori­

Page 6 of 19
Manichaeism and Gnosticism in the Papyri

gin. Christ speaks: “Arise and remember that it is you who hearkened and follow your
root, which is I, the merciful one, and guard yourself against the angels of poverty and
the demons of chaos and all those (p. 630) who ensnare you, and beware of the deep sleep
and the enclosure of the inside of Hades” (Codex II 1, 31; NHLE 116). He then gives or­
ders to John to write down everything he has just said and to give these words to his
(John's) companions.

Of a different flavor is the Gospel of Thomas, presented only once in the codices from Nag
Hammadi (Codex II, 2). This gospel is a collection of 114 sayings of Jesus, of which many
find more elaborate parallels in the Gospels of Matthew, Mark, and Luke. The discussion
of the origins of the New Testament gospels and a possible source, “Q,” which may have
contained a more primitive and no longer extant version of them was enormously stimu­
lated by this new text. Some of those sayings that do not parallel those in the New Testa­
ment have a Gnostic character. Saying 83 runs as follows: “The images are manifest to
man, but the light in them remains concealed in the image of the light of the Father. He
will become manifest, but his image will remain concealed by his light” (Codex II, 2, 47;
NHLE 127).

The sayings are introduced as “the secret sayings which the living Jesus spoke and which
Didymus Judas Thomas wrote down” (NHLE 118). They are thus presented as belonging
to an esoteric circle. But no particular Gnostic group can be identified in which these say­
ings might have been compiled. The Manichaeans' interest in this gospel must have been
considerable, since one of their first missionaries was called Thomas, and the concept of
a heavenly “twin” (didymos is the Greek word for “twin”) was popular with the
Manichaeans as well (mentioned later).

The Gospel of Thomas—like the other texts in the Nag Hammadi Codices—was clearly
translated from the original Greek into Coptic. In this case, we also have parts of the
Greek sayings preserved in papyri that do not belong to the Nag Hammadi find. Three pa­
pyri found at Oxyrhynchus contain twenty-one of Jesus's sayings, which are the same as
those from Nag Hammadi. These papyri date to the beginning or the middle of the third
century CE and are thus about one century older than the Nag Hammadi manuscripts.
When British papyrologists B. P. Grenfell and A. S. Hunt first published P.Oxy. 1 with eight
logia Iesu (sayings of Jesus) in 1897, they did not have the complete Coptic text with its ti­
tle from Nag Hammadi. Grenfell and Hunt's publication sparked a vivid discussion, for
they claimed that these sayings are extracts from a collection of Christ's sayings, not pas­
sages from a narrative gospel (which the Nag Hammadi Codex confirms); that the sayings
are not heretical (which is a matter of the reader's viewpoint); that they are independent
of the four Gospels in their present shape (which the codex confirms); and that they may
go back to the first century CE (which may be correct, but the Nag Hammadi codices give
no further help in answering this question).

For papyrology, the discovery of the logia Iesu at Oxyrhynchus and their publication in
1897 is a decisive historical date. This sensational find made papyrology and its heroes,
Grenfell and Hunt, popular not only in Britain but also worldwide and enabled them to

Page 7 of 19
Manichaeism and Gnosticism in the Papyri

continue their work in Oxyrhynchus in the hope of finding more material that would bring
people closer to the secrets of the New Testament texts. This hope was fulfilled some
years later, when two further (p. 631) fragments with logia were found and published in
1904 in volume four of The Oxyrhynchus Papyri. That these texts were indeed “heretical”
in the eyes of the Christian church from the fourth century on became manifest only
when the complete Coptic version was found at Nag Hammadi.

While the background of the Nag Hammadi find is still obscure, the findspot of the logia,
the town of Oxyrhynchus in Middle Egypt, is one of the best-known towns of Roman
Egypt (see chapter 2). Not surprisingly, in this lively environment all manner of different
Christian groups would develop. As early as the end of the third century, Oxyrhynchus
had at least two churches (P.Oxy. I 43v). Besides the logia Iesu from the Gospel of
Thomas, some additional theological fragments were excavated here by Grenfell and
Hunt, which surely once belonged to a member of a Christian community in which gnosis,
not faith, was believed to lead to salvation. Interestingly, P. Oxy. I 4, which has only twen­
ty-nine lines, some of which are very fragmentary, contains a theological treatise in Greek
about the “upper and the lower soul” and mentions the “prison of the body,” thus reveal­
ing its Gnostic background. P.Oxy. I 4 is dated to the late third or early fourth century.

Of the same date is P.Oxy. VIII 1081, which—like the logia Iesu—turned out to be a Greek
version of a complete Coptic Gnostic text called the Sophia Jesu Christi, preserved in Nag
Hammadi Codex III and in the great Berlin Codex 8502 (mentioned earlier). The main
topic of the Sophia Jesu Christi is the existence of an invisible, supercelestial region in­
habited by the Father and to which those will return who know the Father in pure knowl­
edge.

The Berlin Codex, which dates to the fifth century CE, was most likely found somewhere
around Panopolis in Upper Egypt. Thus, we have for the Sophia Jesu Christi the same situ­
ation as for the logia Jesu: one textual witness in Greek, dating to about the third century
or earlier and coming from the highly Hellenized city of Oxyrhynchus, with its close con­
nections to Alexandria, and two later Coptic versions from the fourth and fifth centuries
that come from Upper Egypt. Taking into consideration the ambiguity of archaeological
finds, which depend on a number of factors, a pattern seems to repeat itself here in show­
ing how the shift from Greek to Coptic developed first in the Hellenized centers of the
north and spread to the more rural Egyptian regions farther south. The same situation ex­
ists with regard to the Gospel according to Mary, first discovered in a small Greek frag­
ment of the early third century excavated at Oxyrhynchus (P.Ryl. III 463) and found in its
Coptic translation in Berlin Codex 8502.

The Greek versions, which undoubtedly were the original ones, circulated in places like
Oxyrhynchus (and certainly Alexandria) as late as the third century; afterward, the Coptic
versions spread and prevailed. Both the Codex Askevianus (of the late fourth century),
which contains the difficult text of the Pistis Sophia, and the Codex Brucianus (of the
same date), with the Books of Jeû, two great Gnostic Coptic codices now in the British Li­
brary and the Bodleian Library, Wt the same pattern.

Page 8 of 19
Manichaeism and Gnosticism in the Papyri

The transition from Greek to Coptic in the Gnostic writings is paralleled by a

(p. 632)

change in the relative importance of the two languages in Christian literature. There is no
apparent difference in this development between the Christian and the “heretical” writ­
ings. The earliest Coptic manuscripts of the New Testament date to the late third or early
fourth centuries CE, the period when the Gnostic writings, too, were translated into the
indigenous tongue.

The close parallels between the holy texts of these competing and conflicting groups are
also apparent in the way their books were made. Christians and Gnostics copied their
books into codices, not onto rolls. As a book form, the roll was gradually abandoned after
the second century CE; there is not a single New Testament or Gnostic text written on a
new roll. Hand in hand with the preference for this recently introduced book format goes
the custom of abbreviating the so-called nomina sacra in certain ways, such as taking the
first and last letters of the names of Jesus or God or the word for “spirit” or “master” and
indicating the abbreviation by putting a line over the remaining letters. No difference in
the copyists' style of writing can be detected. The scribes were most likely not trained for
only one community but worked for whoever employed them. Pagan texts of the time also
do not show different styles of writing. At that time, scribes of pagan texts did not use the
abbreviated nomina sacra. Only Christians and heretics shared common features in their
holy books.

Manichaeism in Egypt
From the fourth century on, the most successful religion apart from Christianity was
Manichaeism, and it remained so for several centuries. Fortunately, we are better in­
formed about the religion of Mani than about any other Gnostic system. Our knowledge is
based to a large extent on papyrological material that was excavated in Egypt. Yet, in this
case, the finds from Egypt are rivaled by those made in the oasis of Turfan in northern
China, where climatic conditions have preserved Manichaean texts and wall paintings
that complete the picture of what was once a world religion in the true sense. Languages
in which Manichaean writings are found range from Greek, Latin, and Coptic to Chinese,
Middle Persian, and Uigur, a Turkic language spoken in the region of the Northern Silk
Road. When Christianity became the official religion around the Mediterranean, Mani's
followers were persecuted. His religion traveled with the merchants along the Silk Road,
and it eventually became the state religion in the Uigur Empire, where it flourished until
the mid-ninth century. But Mani also had numerous followers in Italy and North Africa,
the most prominent of whom was Augustine, who for eight years found in this Gnostic
system the answer to the question of how evil comes into the world.

The singularly wide diffusion of this religious system meant that Manichaeans en­
(p. 633)

countered the myths and beliefs of many contemporary cultures, which had an impact on
the character of Manichaeism in different places: In Europe its followers gave their creed
a consciously Christian flavor, and in the East it had certain Buddhist characteristics. The

Page 9 of 19
Manichaeism and Gnosticism in the Papyri

very versatility of Mani's religion added to its success. The Manichaean writings found in
Egypt, written on papyrus or parchment, all have very close affinities to Christianity.

From its beginning, Manichaeism played an important role in the study of the history of
religions. Augustine wrote fervent invectives against his former coreligionists. In his Pa­
narion, Epiphanius dedicated an extensive essay to this dangerous enemy of Christianity.
The finds at Turfan, where at the beginning of the twentieth century German, British, and
French archaeologists excavated Manichaean monasteries, stimulated new interest. But it
was the discoveries in Egypt that made the study of Manichaeism a lively field of scholar­
ly research during the second half of the twentieth century and beyond. New finds contin­
ue to come to light in the oases of the western desert and elsewhere.

The Cologne Mani Codex

Scholars had long seen Manichaeism as a religion dependent mostly on Zoroastrian tradi­
tions, but the discovery of a small parchment book that now belongs to the papyrus col­
lection of Cologne University in Germany provided a new basis for the understanding of
Mani's cultural and religious background. The Cologne Mani Codex (figure 26.4) is
unique in not only its content but also its extraordinary size. It measures only 3.8 × 4.5
centimeters, and each page contains twenty-three lines written in a careful bookhand in
which the letters do not exceed a height of more than one millimeter. Even though the
codex may have shrunk from its original size (the difference in size cannot be very great),
it is still remarkable that scribes were able to put such tiny, neat letters onto the parch­
ment. They were probably well trained for small writing, as were the cutters of gems in
antiquity, who had to produce miniature figures on very small surfaces. They may also
have used small bottles filled with water, which served as magnifying glasses. The date of
this miniature book is still debated, but on palaeographical and historical grounds, the
fifth century CE seems more likely than the eighth or even ninth century. The place where
the codex was found in Egypt is not certain either, but Upper Egypt and the region
around Lykopolis seem most likely.

Figure 26.4. The Cologne Mani Codex, pp. 58

and 63 (enlarged).

Page 10 of 19
Manichaeism and Gnosticism in the Papyri

The 192 pages of the codex, some of which are very fragmentary, contain the biography
of the founder of Manichaeism. The text is written in Greek, a translation apparently
made in Egypt from the original Aramaic. The style of composition adds to the uniqueness
of this example of a biography from late antiquity. Through excerpts from writings by his
first disciples, we follow Mani's life. Each citation is headed by the name of the disciple—
some of them known also from other (p. 634) (p. 635) Manichaean sources. As eyewitness­
es (or at least claiming to be), the disciples report both the events that involve their mas­
ter and his very words. The authenticity that this procedure asserts corresponds to
Mani's claim of the superiority of his religion, which was based on the importance he
gave to the recording of everything he said and did. For instance, Mani said, “Indeed, all
the (apostles), my brethren who came prior to me: (they did not write) their wisdom in
books the way that I, I have written it. (Nor) did they depict their wisdom in the Picture (-
Book) the way (that I, I have) depicted it. My church surpasses (in this other matter also),
for its primacy to the first churches” (Keph. 370, 16–375, 15; Gardner and Lieu 2004,
265–268). Mani's religion was as much a religion of the book as Christianity. The holiness
of books in Manichaeism may help us to understand the singularly small size of the
Cologne Mani Codex. As the product of an extremely difficult scribal process, the book it­
self would have counted as proof of the holiness of its content and of the followers' dedi­
cation to the religion's founder, whose life was told in its pages.

The Cologne Mani Codex is far from complete. When we first meet Mani, he is already
four years old, and we lose sight of him on the last extremely fragmentary pages, when he
is twenty-six. The remaining years of his life must be reconstructed from other sources.
Mani was born in 216 in the city of Seleucia-Ctesiphon in Mesopotamia; when he was four
years old, his father took him away from his mother to live with him in a community of
baptists on the lower Euphrates. Here Mani experiences his first revelations, which are
brought to him by a heavenly twin. The twin is the key to Mani's religious system. He de­
scribes him in these words (Cologne Mani Codex 24; Gardner and Lieu 2004, 51): “I re­
ceived him (my twin) piously, and I obtained him as my own property. I believed him that
he belongs to me and that he is a good and useful counselor. I recognized him, and under­
stood that I am he from whom I was separated. I have borne witness that I myself am he
and am completely the same.” As in every Gnostic system, human beings come from the
realm of light, to which they must return. Mani's belief was that, at death, he would be re­
united with his heavenly twin and that not only he himself but all human beings as well
have a twin with whom they will be one again after death (with some restrictions, as we
will see later). Also of interest in Mani's song about the twin is the sequence of receiving,
believing, and recognizing, which in the climax describes the steps that lead to full gno­
sis.

Even as a child, Mani has also learned that all plants contain particles of light that human
beings are not allowed to harm if they want to become perfect Manichaeans. Plants weep
when they are harvested, and trees talk in anger when they are cut, making their suffer­
ings clear. For Mani's followers, his biography must have been an entertaining yet still se­
rious book to read or listen to. His biography is a pious narrative with a folkloristic tone.

Page 11 of 19
Manichaeism and Gnosticism in the Papyri

The baptists with whom Mani and his father lived are described as members of a Jewish-
Christian community on the Shat el-Arab River. The Cologne Mani Codex thus demon­
strates for the first time that, besides the Persian religion of (p. 636) Zoroaster (and
maybe more important than it), Christianity and Judaism were the most powerful roots of
Manichaeism.

At the age of twenty-four, Mani received a second great revelation from his twin, and af­
ter a dramatic discussion with the baptists, he left them and embarked on his missionary
travels, which took him from Mesopotamia, through Persia, and ultimately as far as India.
In this passage again, the codex reflects a number of folkloristic elements, such as when
Mani meets a hoary hermit whom he converts or when he appears before a king and his
attendants and convinces them all of the truth of his teachings.

Upon returning to Mesopotamia, Mani is welcomed by the Sassanid king, Shapur I, who
supported his religion. Here the Mani Codex breaks off. We know the rest in much less
detail from some sources in Middle Persian. Shapur's successors were less favorably dis­
posed, and in 276 Mani died in a prison in Seleucia-Ctesiphon after the priests of the
Zoroastrian cult, who recognized him and his religion as a serious rival, had denounced
him at court.

What made Mani's religion so successful? Certainly the charisma of its founder played a
role, as did his determination to promote his religion more vigorously than all others be­
fore him, claiming that he was the real Apostle of Light after Zoroaster, Jesus, and the
apostle Paul. He believed that the time of the oral transmission of teachings was over
(which had led to terrible controversies in Christianity) and that everything he had said
and done should be written down. This was a new and decisive shift away from the spo­
ken word to words that existed in writing and were therefore eternal. The folkloristic fla­
vor of his biography, as we see it in the Cologne Mani Codex, may have added to the pop­
ularity of his teaching, as his painted book, which showed the creation of the world and
the fight between light and darkness, also certainly did. In the Persian tradition, Mani is
still called the “Painter.” Another advantage of Mani's church was its strict hierarchic
structure, with one supreme leader, twelve teachers, seventy-two bishops, and so on, who
were sent out wherever a Manichaean church was founded. In the fourth century, a
Manichaean teacher was established in Egypt (mentioned later).

On the other hand, the difficulty of his system was the strict asceticism that it expected of
those who wanted to become electi, the chosen ones, who would advance the task of the
separation of light and darkness. To collect as many light particles in themselves as possi­
ble, the elects had to avoid harming light particles in any living entity. Thus, they were not
allowed to harvest plants, cut trees, bake bread, or even touch (they would have said “to
beat”) water. Naturally, they were not allowed to marry or have sex. Their food was
preferably organic material that contained a high percentage of light particles. Melons
were considered a particularly powerful source. As elects they wandered around, begging
for their daily food, and were allowed only to pray and sing or to copy the holy books.
This upper class of Manichaeans was supported by the catechumens, who fed and housed

Page 12 of 19
Manichaeism and Gnosticism in the Papyri

them. (p. 637) Catechumens could not expect to play a decisive role in the separation of
light and darkness but could hope to be reborn as elects in future lives (Mani had taken
the belief in the transmigration of souls from his Buddhist experience in India). The
lifestyle dictated to the elects probably stimulated a system of housing to which these
wandering ascetics could turn. It has been argued that the earliest monasteries (the old­
est one that is clearly dated is that founded in Egypt by Pachomius in 325) had their roots
in Manichaean guesthouses for their elects.1 But the Christians made of these guesthous­
es something completely different. On the one side we find the lonely elects, unable to
support themselves and expecting only to be fed and clothed, and on the other side were
the monks, who were dedicated to a communal life of work and prayer in which every­
thing was to be shared.

Mani's religion was likely introduced in Egypt during his lifetime. Some texts in Middle
Persian (Gardner and Lieu 2004, 111) tell of the first missionary, Addaius, who “came as
far as Alexandria.” He and other missionaries may have reached Egypt by ship across the
Red Sea from the Shat el-Arab River rather than overland through the desert to Alexan­
dria. Traveling by ship, they would have first reached the regions of Upper Egypt before
going downstream to Alexandria, and indeed it seems that the strongholds of
Manichaeism were in Upper Egypt in the area around Lykopolis (today Assiut), from
where Manichaeism reached the oases in the western desert. Most of the papyrological
finds concerning Mani's religion in Egypt originate from the region around Lykopolis or
most likely came from there.

Between 277 and 297 the neo-Platonist Alexander of Lykopolis wrote a well-informed
treatise against the Manichaeans. He may have had personal encounters with
Manichaean missionaries. A letter written on papyrus, most likely by the bishop of
Alexandria, also dates to the end of the third century. In it he warns Christians not to
trust the Manichaeans and stresses the absurdity of their teachings (P.Ryl. 469; Gardner
and Lieu 2004, 114–115); he condemns the “stupidity” of the ascetic practices of the
elects and quotes the prayer they repeat before eating: “Neither have I cast the bread in­
to the oven. Another has brought me this and I have eaten it without guilt.” He continues:
“Whence we can easily conclude that the Manichaeans are filled with much madness.”

Better than the literary antiheretical works, this letter shows the Christians' genuine con­
cerns about the then new religion originating in Mesopotamia. It attacks the
Manichaeans for practices that were fundamental to their concept of salvation. This be­
havior was openly displayed by the elects when they wandered from place to place to ask
for alms. Where Manichaeans were at home, the figure of the uncombed and smelly elect
became a common feature in everyday life and provoked feelings of dismay and much
mockery. It is surprising that we hear rather little of attacks or difficulties in the private
letters of Mani's followers.

Page 13 of 19
Manichaeism and Gnosticism in the Papyri

The Manichaean Codices from Medinet Madi (Fayyum) and the

(p. 638)

Private Documents from Upper Egypt and the Oases

Even though the first large find of Manichaean texts was made in the Fayyum, the Coptic
dialect in which these texts are written reveals that these books originated in the Lykopo­
lis area. In 1929, farmers in the southwestern Fayyum were plundering mud-brick houses
of the Roman period on the ancient site of Narmouthis (today Medinet Madi) in hopes of
finding fertilizer for their fields. They discovered a box filled with seven codices still
bound in their wooden covers. From the antiquities market in Cairo, four volumes were
brought to Berlin, while the other three became part of the Chester Beatty collection in
Dublin. This find, made forty years before the Cologne Mani Codex became known, pro­
vided new insight into what Manichaeism was and how “Christian” it could be. The
codices contained two of the canonical works of Mani's religion: the Kephalaia of the
Teacher (the main teachings of Mani, which once consisted of 520 pages with at least 172
chapters, now in Berlin; some very small fragments in Vienna) and the Epistles of Mani
(Berlin). From his pupils there were the Synaxeis of the Living Gospel (Berlin). The find
also included a Psalm Book (once containing more than 600 pages; Dublin), Homilies
(Dublin), and a Church History of at least 250 pages (formerly in Berlin, now lost). Coptol­
ogists recognized that these texts had been translated from a Syriac original and had
been copied into the codices around 400 CE. Besides the fact that many of the pages of
these Manichaean books are horribly worn or quite fragmentary, the entire codex of the
Church History disappeared from Berlin at the end of the Second World War. In a prelimi­
nary report Carl Schmidt and Hans Jakob Polotsky had said enough about the content of
this codex to make apparent its affinity in style and content to what we have now in the
Greek Mani Codex in Cologne (Schmidt and Polotsky 1933).

Jesus plays an important role in the Kephalaia of the Teacher, Mani. In chapter 8, Mani
talks about the path that Jesus took in his descent to earth, incorporating the Christian
concept of salvation into the Manichaean cosmos and bringing together Christ and the
Manichaean entities from the realm of light (8, 36, 27–37, 27; Gardner and Lieu 2004,
218–219): “Once again, the Light-Man speaks to the congregation that is sitting in front
of him: When Jesus the son of Greatness came to this world, at the time that he unveiled
the greatness, he boarded ten vehicles! He journeyed in the universe by them. The first
vehicle is the light ship, since he received instructions from the Ambassador there. The
second carriage is the ship of the First Man, since his dwelling is established there. The
third is the Pillar of Glory, the Perfect Man, since he shone forth there.… After he had as­
sumed these ten, he came and manifested himself in the flesh. He chose the holy church
in four vehicles. One is all the holy brothers. The second is the pure sisters. The third is
all the catechumens, the sons of the faith. The fourth is the catechumens, the daughters
of the light and truth.” A large portion of the codices from Medinet Madi is now available
in new editions and translations.

Until fairly recently, papyrological finds that illustrated Manichaean life in Egypt
(p. 639)

beyond the scriptures of its church were rare. Only two letters are recognized (not with­
out dispute: see chapter 25) as having been written by followers of Mani. The first, P.Harr.

Page 14 of 19
Manichaeism and Gnosticism in the Papyri

I 107, is a late third or early fourth-century letter of unknown provenance written by a

certain Besas to his mother. In it the son sends greetings to his mother and refers to the
“Father, God of Truth” and the “Paraclete Spirit.”2 The other letter comes from fourth-
century Oxyrhynchus (P.Oxy. XXXI 2603) and awaits republication.3 A Manichaean back­
ground has been proposed. On a high intellectual level, the writer of this letter recom­
mends his brethren to be received by the addressees, using the image of a mirror as an
argument for mutual trust. At the end he sends greetings to the “brethren with you, both
elect and catechumens” A third Manichaean letter is currently being prepared for edition.
It comes from Oxyrhynchus and also contains greetings to elects and catechumens. A
Manichaean community is thus attested in Oxyrhynchus at least for the fourth century.
Some small Syriac fragments written in a specifically Manichaean script add to this pic­
ture (Burkitt 1925, 111–119).

A more detailed idea of how Manichaeans lived in their religious communities and also
participated in the daily life of their (most likely predominantly) Christian environment
has been provided by the excavations in the Dakhleh Oasis, about 180 kilometers south­
west of Assiut. Since the early 1990s, Australian scholars have uncovered the remains of
the ancient village of Kellis (figure 26.5). Manichaeism is attested there until the late
fourth century, when the village was abandoned. In addition to letters, literary texts writ­
ten on papyrus or wooden boards have also been found. Both kinds of texts are presented
in either Greek or Coptic; of special interest is a word list that gives Manichaean techni­
cal terms in both Syriac and Coptic. This word list has again stimulated debate about
whether indeed all Manichaean texts went through two stages of translation, first from
Syriac into Greek and then into Coptic, or whether they were directly translated from
Syriac into Coptic.

Literary texts from Kellis include a Greek version of a psalm already known from the Cop­
tic psalm book from Medinet Madi, some of Mani's epistles in Coptic, which were lost in
the Medinet Madi codices, and a very interesting prayer in Greek, titled “Prayer of the
Emanations.” It lists all of the divine entities that play a role in the separation of light and
darkness and are sent down by the Father of Light to rescue the first man and mankind.
Whether these Greek texts are versions that were used at an earlier stage and then aban­
doned, or whether both languages were used side by side remains undetermined.

Page 15 of 19
Manichaeism and Gnosticism in the Papyri

Figure 26.5. House 3, Room 6, at Kellis with an­

cient material scattered on the floor.

The Manichaeans of Kellis also wrote in Greek or Coptic in their private lives (figure
26.6). We have more than thirty private letters written in Coptic by followers of Mani
(more may yet be published) and one in Greek that is undoubtedly of Manichaean back­
ground. It is a striking feature of the Coptic letters from Kellis that many of them have
the addressee's name written in Greek (as if the “letter carrier” would understand that
language better) and that the salutations are also phrased in Greek, while the main cor­
pus of the letter follows in Coptic. Most (p. 640) of the Coptic letters were found in “House
3,” thus acquainting us with one Manichaean household in the middle of the village. No
elect is named as such: The people in House 3, both men and women, all seem to have
been catechumens. That the senders (and recipients) of these letters were followers of
Mani is clear from the references to the elect, the catechumens, and the Paraclete. It
seems that the theological idea of Mani or his twin (ultimately the same person) as the
Paraclete whom Jesus had announced (John 14:26) was common knowledge in the
Manichaean communities. The private letters from Oxyrhynchus also refer to the founder
of the religion as the Paraclete.

Page 16 of 19
Manichaeism and Gnosticism in the Papyri

Figure 26.6. P.Kell.Copt. 25, letter by Matthaios

to his mother, Maria (from House 3, Room 6, at
Kellis).

The tone of the letters is solemn, with people often talking piously with one another. The
letters from the family archive of Makarios, which date to between 350 and 370, are char­
acteristic. When Makarios receives letters from his sons, who were dwelling at the time in
the Nile valley, we even get a glimpse of the highest Manichaean official in Egypt. “I will
go with the Teacher to Alexandria” (P.Kell Copt. 29.13–15), we read there, or “my brother
Piene is learning Latin with the Teacher” (P.Kell.Copt. 20.24–26). Is it possible that Piene
was preparing to become a Manichaean missionary in the west? Books and their produc­
tion play an important role in the lives of these people. Books (unfortunately without any
titles) are exchanged (P.Kell Copt. 20, 36; 26, 28 ff) and copied (P.Kell.Copt. 22, 66; 24,
36; and 34, 23). Despite this interest in the Holy Scriptures, however, the letters say noth­
ing about the realm of light and its (p. 641) (p. 642) messengers, in contrast to the “Prayer
to the Emanations” (mentioned earlier), which shows the importance of the myth in the
Manichaean church, as does the psalm book from Medinet Madi. Being a Manichaean
was serious business; the act of providing alms to the elect is referred to in two letters:
P.Kell.Copt. 15 and P.Kell.Gr. 63. For these people, their church is the “HolyChurch,” and
some of them may have had to suffer for their beliefs. The warnings not to let a letter fall
into the hands of “someone,” as well as the need for secrecy, do not necessarily refer to
the difficult circumstances of the person as a Manichaean but rather to the position of a
Manichaean believer in a difficult family situation. The appearance of a magical spell that
accompanies one letter is astonishing because magic in all forms was prohibited to the
Manichaeans; this particular spell is a curse to be used to separate a loving couple.

Further finds may even better illustrate the coexistence of Christians and Manichaeans in
this environment. The excavation of the Manichaean monastery that is mentioned in an
account book of rents found at Kellis but has not yet been located could also give us new
insight into the relationship between elects and catechumens in their daily lives. The
Page 17 of 19
Manichaeism and Gnosticism in the Papyri

twenty-first century might bring the lives of these persecuted people even more into fo­
cus through the books and private letters that still await excavation.

Bibliography

Nag Hammadi
Otranto, R. 2000. Antiche liste di libri su papiro. Rome: Edizioni di storia e letteratura.

Robinson, J. M., ed. 1990, rev. ed. The Nag Hammadi Library in English, Translated by
Members of the Coptic Gnostic Library Project of the Institute for Antiquity and Chris­
tianity. Leiden: Brill. Cited in the text as NHLE.

Till, W. C., ed. 1972. Die gnostischen Schriften des Koptischen Papyrus Berolinensis 8502,
2nd ed. Berlin: Akademie Verlag.

Turner, J. D., and A. McGuire, eds. 1997. The Nag Hammadi Library after Fifty Years: Pro­
ceedings of the 1995 Society of Biblical Literature Commemoration. Leiden: Brill.

Veilleux, A. 1980. Pachomian Koinonia. Vol. 1, The Life of Saint Pachomius and His
(p. 643)

Disciples. Kalamazoo, Mich.: Cistercian Publications.

Manichaeism
Burkitt, F. C. 1925. The Religion of the Manichees. Cambridge: Cambridge University
Press.

Gardner, I., A. Alcock, and W.-P. Funk, eds. 1999. Coptic Documentary Texts from Kellis,
vol. 1 (=P.Kell. V). Dakhleh Oasis Project Monograph 9. Oakville, Conn.: Oxbow.

Gardner, I., and S. N. C. Lieu, eds. 2004. Manichaean Texts from the Roman Empire. Cam­
bridge: Cambridge University Press.

Gardner, I., A. Nobbs, and M. Choat. 2000. “P. Harr. 107: Is This Another Greek
Manichaean Letter?” ZPE 131: 118–124.

Koenen, L. 1983. Manichäaische Mission und Klöoster in Ägypten. In Das Römisch-byzan­

tinische Ägypten, 93–108. Aegyptiaca Treverensia 2. Mainz am Rhein: P. von Zabern.

–––, and C. Römer, eds. 1988. Der Kölner ManiKodex: Über das Werden seines Leibes.
Opladen: Westdeutscher Verlag.

Lieu, S. N. C. 1994. Manichaeism in Mesopotamia and the Roman East. Leiden: Brill.

Schmidt, C., and H. J. Polotsky. 1933. Ein Mani-Fund in Ägypten. Berlin: Akademie der
Wissenschaften.

Page 18 of 19
Manichaeism and Gnosticism in the Papyri

Notes:

(1.) Koenen (1983) has strongly argued for this connection between Manichaean houses
and the early monasteries.

(2.) The main arguments for a Manichaean background of this letter are found in Gard­
ner, Nobbs, and Choat (2000, 118–124, and plate III).

(3.) Lieu (1994, 98n316).

Cornelia Römer

Cornelia Römer is Director of the Papyrus Collection and Papyrus Museum, Austrian
National Library, Vienna.

Page 19 of 19
The Future of Papyrology

The Future of Papyrology

Peter van Minnen
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011

Subject: Classical Studies, Greek and Roman Papyrology, Material Culture Studies
Online Publication Date: Sep 2012 DOI: 10.1093/oxfordhb/9780199843695.013.0027

Abstract and Keywords

This article looks at the future of papyrology. It draws heavily on two position papers Pe­
ter van Minen gave at international papyrological congresses: in Copenhagen in 1992 and
in Vienna in 2001. The first section starts with the immediate context from which the pa­
pyri derive as physical objects. The second section concentrates on the texts written on
them. The final section focuses on the wider context that produced the texts. The article
concludes by noting the need for better communication between papyrologists.

Keywords: Peter van Minnen, international papyrological congresses, physical objects, papyri, papyrologists

IN this final chapter I take a look ahead. In doing so I draw heavily on two position papers
I have given at international papyrological congresses: in Copenhagen in 1992 (Van Min­
nen 1993) and in Vienna in 2001 (van Minnen 2007; such congresses are held under the
auspices of the Association Internationale de Papyrologues every three years). In the first
presentation, which dealt with the history of papyrology, I made the past normative; in
the second (“The Millennium of Papyrology (2001–)?”), I made the future normative. In
this chapter I again do the latter because readers are more indulgent when it comes to
the future; the past has to be manipulated too much to be acceptable as normative.

Papyrologists have made predictions or projections about their future before, but they
have usually been reluctant to commit themselves too far (see McGing 2006, 246–248).
The title of my first paper (“The Century of Papyrology [1892–1992]”) actually derives
from one of the earliest predictions about papyrology ever made: By 1900, Ludwig Mitteis
and Theodor Mommsen (see Martin 2000; Gonis 2006) had both claimed that the twenti­
eth century would be the “century of papyrology,” just as the nineteenth century had
been the “century of epigraphy.” In my first paper I discussed how accurate this predic­
tion had been and suggested ways in which papyrologists could improve their communi­
cation with other disciplines. The present handbook goes a long way toward doing just
that.

Page 1 of 17
The Future of Papyrology

In the second paper, I calculated how much work remains to be done. I concentrated on
the editing of unpublished texts, what I called the “core business” of papyrologists. Even
if only half of all unpublished texts in the more than 1,400 (p. 645) known collections
worldwide from “Aachen to Zutphen” (an estimated 1,000,000–1,500,000, of which almost
half are held by the Egypt Exploration Society) are publishable, it would still take papy­
rologists ten times as long as it took them to publish the estimated 72,500 published texts
in the hundred years since about 1895 (broken down into 50,000 Greek and Latin docu­
ments; 7,500 Greek and Latin literary texts; as well as 7,500 Coptic; 3,500 demotic and
abnormal hieratic; 3,000 Arabic; and 1,000 Aramaic and Pehlevi texts). Hence, the “mil­
lennium of papyrology.”

My statistics on literary and documentary texts from papyri and ostraca published or re­
published in five-year periods from 1890 to 1999 (van Minnen 2007, 714) can be updated
as follows. In the period 2000–2004, about 4,500 texts were published or republished,
and 2005 and 2006 were exceptionally good years, with about 1,500 and 2,500 texts pub­
lished or republished. The total of published texts now stands at about 80,000.

On average, 12 volumes (monographs) with papyrus texts come out every year, and this
has been the case since about the mid-1970s. There is every reason to believe that this
publication rate, with an average of about seven hundred Greek and Latin texts published
each year, as well as texts in other languages at irregular intervals, is going to continue in
the foreseeable future. At least the five-year period since 2000 has confirmed the predic­
tion I made in 2001.

In these statistics I ignored “normal” hieratic papyri because they date from before the
papyrological “millennium and a half” (from the earliest abnormal hieratic and demotic
papyri in the early seventh century BCE to the latest Greek papyri in the late eighth cen­
tury ce). I also ignored papyri with texts in yet other languages such as Syriac, Armenian,
and Gothic because there are only a handful of such texts to begin with. A Punic papyrus
of the sixth century BCE can be added to the list of oddities: an amulet found inside a
bronze case with the head of a falcon at Tal-Virtù, near Malta (on this “Maltese falcon,”
see Rocco 1975).

Editing the unpublished texts in collections worldwide and republishing texts in need of
revision will keep papyrologists busy for centuries at least. That much is clear. Given the
tendency, more marked in the past than at present, to publish relatively more Greek liter­
ary papyri than other papyri and relatively more Greek and Latin documentary papyri
than papyri in other languages, as well as relatively more Ptolemaic and early Roman
documentary papyri than documentary papyri from late antiquity, the current backlog in
the publication of documentary papyri, especially those from late antiquity and/or in oth­
er languages, is even bigger than that for the kinds of papyri that have traditionally re­
ceived more attention from papyrologists. But none of this changes the bottom line: There
are enough papyri of every kind, from any period and in any language, to keep papyrolo­
gists busy for quite some time.

Page 2 of 17
The Future of Papyrology

The only problem I see is not that there are not enough texts but that there are not
enough papyrologists. By increasing their number, one could reduce (p. 646) the amount
of time it will take to publish the remaining texts, but that will not be easy. Papyrologists
who edit texts use a number of specialized skills they have acquired in the course of a
decade or more. Ideally, the acquisition of the philological skills (knowledge of Greek and/
or other ancient languages) should take place before the acquisition of the specialized
skills (palaeographical, editorial, and historical) needed to read, edit, and use papyrus
texts, but this is not always possible. It is also not easy to speed up the process, and in­
creasing the numbers of trainees at the limited number of institutions that actually offer
courses and other forms of training in papyrology is a financial and logistical problem.

In recent years, a consortium of ten institutions in North America has started a series of
papyrological summer institutes of the kind that was very successful in the 1960s. The ex­
pectation is that this will continue after the current round(2003–2012; for announce­
ments see http://www.papyrology.org/). Shorter summer programs are occasionally orga­
nized at a variety of European institutions. All of this is aimed at making a generation of
younger students of the ancient world at least aware of papyri and in some cases to “con­
vert” them to papyrology (for a case of “enforced apostasy” from papyrology, see P.Merton
II, p. vii). The pool of potential “converts” may decline or expand, depending on the gen­
eral level of enrollment in subjects such as the classics, ancient history (with Greek and
Latin), Egyptology, and classical Arabic. If, say, enrollments in the classics go down, papy­
rologists will have to make a greater effort to reach students before they settle for
“Homer and Vergil” (and not much else). It seems likely that Egyptology and classical
Arabic will attract increasing numbers of students in the foreseeable future. Papyrolo­
gists will need to make sure that their material can be used responsibly by ancient history
students who have little or no knowledge of ancient languages (they are appearing in in­
creasing numbers) by streamlining the way they present papyrus texts and making trans­
lations widely available.

Papyrologists do or could do many other things besides editing and “disseminating” texts.
In fact, I earlier calculated the number of editors of papyrus texts in the last decades of
the twentieth century at only sixty-five (van Minnen 2007, 707), whereas a couple of hun­
dred papyrologists usually show up at international congresses. Clearly, papyrology can­
not be so narrowly defined as to exclude most of its practitioners.

There are very few full-time papyrologists, scholars who do nothing else: Most teach an­
cient Greek or ancient history besides. What do all of these papyrologists do? They edit
and reedit texts from papyri and/or with their help study some aspect of the ancient world
(e.g., its literature or the social history of Egypt in the Graeco-Roman period). Sometimes
they even go to Egypt to dig up more papyri and to find out more about the context from
which they derive. There have always been such papyrologists, but early on, they did a lot
of harm while doing a lot of good: They dug up hundreds of thousands of papyri, which
will keep papyrologists busy for another millennium, but they also destroyed the original
context from which these papyri derived without recording it properly. Since the 1930s,
better (p. 647) recordkeeping at excavations in Egypt has generated a huge quantity of ad­

Page 3 of 17
The Future of Papyrology

ditional data to match what one reads in the texts. A lot of this still needs to be explored,
adding enormously to the task at hand, but the most recent excavations will no doubt re­
sult in models for using data from older excavations, which are often incomplete in any
case.

In the section that follows, I start with the immediate context from which the papyri de­
rive as physical objects. In the section after that, I concentrate on the texts written on
them, and in the final section I focus on the wider context that produced the texts.

Papyri as Ancient Artifacts

Papyri and ostraca have been found mainly in Egypt and less so in other parts of the an­
cient world. Early on, papyri were dug up by the local population and by papyrologists in
search of papyri (and little else). Early archaeologists followed their trail, but it was not
until the 1930s that serious excavations (not just to find more papyri but also to record
the finds in context) were undertaken. World War II disrupted this promising develop­
ment for decades. In the last couple of decades, however, new excavations have been un­
dertaken with remarkable success (for a survey see Bagnall 2001).

The first serious excavations concentrated on sites on the outskirts of the Arsinoite nome,
such as Karanis and Tebtynis, which had yielded many papyri before. Special circum­
stances there have preserved much organic material. In the course of the Roman period,
the gains on the desert made in the early Ptolemaic period were lost, and many sites on
the outskirts were returned to the desert for a millennium and a half. In recent years, the
expansion of agriculture in Egypt has regained some of these losses, but there are still
sites left to search for papyri and other materials (see the urgent call for “rescue” excava­
tions in Gallazzi 1994), which will allow a detailed reconstruction of many aspects of life
in these towns based not just on texts but also on the whole range of materials found.
This has always been the goal of serious excavations in the Arsinoite nome, but so far
none has delivered on its promise (I outlined a strategy to come to grips with the early ex­
cavations at Karanis in van Minnen 1994). In the future, however, one may expect to be
able to put the various kinds of materials so far studied in isolation from one another back
together once the excavators sit down to make sense of the whole.

In recent decades, the focus has shifted somewhat to sites in the Eastern Desert and in
the oases in the Western Desert. The former are mainly, but not exclusively, Roman army
camps and the like, which have so far yielded large quantities of (p. 648) ostraca and a
wealth of other materials (on the exemplary work at Mons Claudianus see, e.g., O.Claud.
III). Excavations on the Red Sea coast have revealed, for example, Berenice, a town of
major importance to the trade with East Africa and India from the early Ptolemaic period
to late antiquity (see Wendrich et al. 2003). Sites in the oases of the Western Desert are
even more promising because they tend to be more “normal” towns, such as one finds in
the Nile valley, except for the oases' special hydrological circumstances (on Kellis, the
best-published site so far, see the Dakhleh Oasis Project: Monograph series).

Page 4 of 17
The Future of Papyrology

We have already begun to understand quite a bit more about the Roman army system in
the Eastern Desert, mainly through the work of Cuvigny (see also chapter 2). In the near
future one can also expect to learn quite a bit more about sites in the Western Desert
(e.g., Mothis), which have so far not been foremost in the study of Graeco-Roman Egypt.
For the delta, however, one will probably never know much about the kinds of things tak­
en for granted in the Nile valley and the Arsinoite nome and presumably soon in the oases
of the Western Desert because the water table there was raised as early as the nineteenth
century—before papyrological excavations had started—and this development has contin­
ued apace ever since. The archaeology of nonorganic materials there will make some
progress, but that is not papyrology.

What the expansion into the desert areas and the exploration of new sites in the Arsinoite
nome and even the Nile valley will do to papyrology is mainly twofold. In the first place, it
will redress an annoying imbalance. Ever since the end of the nineteenth century, papy­
rologists have drawn mainly on the same finds from a limited number of sites. All of the
statistics so far produced (see Rémondon 1966 through Habermann 1998) therefore show
the same kinds of bias. Expanding into new territories will generate at least new biases to
match the old ones.

Second, the exploration of new sites will continue to add to the stock of unpublished ma­
terial. As a matter of fact, more papyri and ostraca are currently found each year than are
published by papyrologists. For as long as more are found in Egypt, the daunting backlog
of 1,000,000–1,500,000 unpublished items will therefore not diminish, no matter what pa­
pyrologists do.

An added effect of the expansion into new territories is a more traditional one: Artifacts,
including papyri, will disappear from excavations and reappear on the market (cf. chapter
21). Nowadays, responsible institutions no longer invest in “hot” papyri, but there is too
much money in irresponsible hands. “Hot” papyri often lack a provenance, and the con­
text in which they were found is deliberately effaced.

Once papyri are retrieved, they need to be conserved on site and in the museum where
they will be kept. Conservation of papyri has taken a turn for the better since the
mid-1970s, and this will no doubt continue in the foreseeable future. Another phenome­
non that is also likely to go on indefinitely is the increased access to information about
and images of papyri. First Yale University, then the University of Michigan, pioneered
the electronic cataloguing of papyri in (p. 649) their collections. Then a more ambitious
project was pioneered at Duke University, where the largely unexplored collection was
conserved, catalogued and digitized. By the end of 1995, Duke was the first collection to
go online as a whole—to go “public.” At that point, a collaborative effort linking the most
important collections in the United States got under way, the Advanced Papyrological In­
formation System (APIS). The much larger collections at Yale, Michigan, the University of
California—Berkeley, Columbia University, and Princeton University, later joined by oth­
ers around the globe, have made many more images of and much more information about
papyri in their collections available. My guess is that currently about 10 percent of all

Page 5 of 17
The Future of Papyrology

published texts are now accounted for in APIS-type projects, the bulk of these in APIS it­
self. I use the term “accounted for” advisedly as not all projects provide the same level of
detail or coverage for published texts. Coverage of unpublished material in APIS-type
projects is much spottier.

All of this material is accessible through the Web sites of the individual collections, but
most of it is also available as a whole at the Web site at Columbia University http://
www.columbia.edu/cu/lweb/projects/digital/apis/; now also at http://www.papyri.info). As
time progresses, ever more papyri in collections worldwide will become accessible in
some form through digitized images, and incomplete records will be added to over time.
The quality of papyrological work that deals with published texts will improve because it
will be possible to base it not just on the information in the printed edition but increasing­
ly also on high-quality images. This has already revolutionized the work with texts from
certain collections, and the same will happen with all of them.

Making images of unpublished papyri available will definitely make it possible for out­
siders without access to the originals to work on a “virtual collection.” This has already
led to the publication by papyrologists from all over the world of dozens of papyri at Duke
University.

One possible effect of APIS-type efforts to make images of papyri available in bulk is to
give the renewed interest in palaeography a boost (see chapter 5). Palaeography is more
than a dating tool, but even as a dating tool it is badly in need of revision. For some
“papyrologies” (e.g., Coptic papyrology), a careful study of dated texts could generate the
chronological backbone.

Papyri as Texts
Most papyrologists are mainly interested in the texts written on papyri. True, they some­
times interact with originals, but the texts are the main thing papyrologists worry about.
There is nothing wrong with that. Texts are the papyrologist's facts (p. 650) (cf. Youtie
1963). Most of them study papyri because they have a passionate interest in the past and
because they are trained to construe that past from texts. Texts indeed provide a mean­
ingful link with the past because they are bits of communication. When reading, one lis­
tens in on a conversation of long ago, as it were. I do not deny that one can also “read”
other artifacts and that in a postprocessual world pots do have a soul (contrast Wilam­
owitz-Moellendorff's obiter dictum that “Töpfe haben keine Seele”: Wilamowitz-Moellen­
dorff 1913, 5). But most papyrologists are not archaeologists but philologists by training,
and therefore this section on texts is twice as long as the one on papyri as artifacts.

Texts are also accessible to anyone who knows the ancient language they are in or can
use translations responsibly. This means that potentially all students of the ancient world
can, if only occasionally, become papyrologists. The problem with this has always been
mainly twofold. In the first place, most papyrological publications are obscure, certainly
in comparison with epigraphical publications, where the great series (IG, CIV) at least
Page 6 of 17
The Future of Papyrology

provide a home for the bulk of the texts. Only The Oxyrhynchus Papyri can claim a some­
what similar status, mainly because of the way this series has been marketed since the
end of the nineteenth century. Many libraries otherwise devoid of papyrological publica­
tions have a set of P.Oxy. The recent development of electronic tools has brought most
published papyrus documents in Greek and Latin and some in Arabic within the reach of
all. In some cases, translations are available as well, either through APIS-type projects
(which deal with the material collection by collection) or through a comprehensive data­
base (as for Arabic documentary papyri). Eventually all documentary papyrus texts will
become available with translations linked to them in some form. If making the papyrus
texts more “user-friendly” (through descriptions that retrace the flow of the text and es­
pecially annotated translations) is done responsibly, chances are that papyrus texts will
be used responsibly by future generations of scholars, even those who do not have the
specialized skills to deal with the material directly in the original languages.

In the second place, papyrus texts are difficult for outsiders to access because of the spe­
cial formatting imposed on the representation of texts in printed form. These are purely
conventional (the so-called Leiden system) and limited in number but evidently a major
obstacle for many. Things here have been further complicated since the 1980s. In putting
the printed texts of Greek and Latin documents on papyri and ostraca into electronic form
in the Duke Data Bank of Documentary Papyri (DDBDP), additional interventions in their
representation were rendered necessary. The orthography of many words had to be cor­
rected—a “smart” searching program that would recognize aberrant forms did not yet ex­
ist. All kinds of other information (e.g., on scribal corrections or corrections of readings in
the various addenda et corrigenda to the printed editions) were inserted with the help of
other kinds of formatting.

Originally, the DDBDP was meant as a huge index to the printed volumes, but when it was
completed (for the time being) in June 1996, its original purpose had (p. 651) long been
superseded. It had turned out to be not just a huge index but also the premier tool for edi­
tors of Greek and Latin documentary papyri. Increasingly, papyrus collections had been
exhausted as far as complete and legible documents were concerned. Around the time
this was happening, the DDBDP made it possible to identify and publish even fragmen­
tary texts with confidence. The relatively simple searching program allowed one to recog­
nize a few words or even a string of letters as deriving from a particular type of docu­
ment, for which one then found more complete parallels in the DDBDP. This is the most
important function of the most important tool papyrologists currently use, and it will re­
main that indefinitely. I think the exceptionally prominent status of this electronic tool
warrants a few additional comments.

When the DDBDP first went online (now at http://www.papyri.info), a user-friendly inter­
face changed the way the texts are represented in significant ways. Even papyrologists
now think that what the interface shows on the screen is the text they can quote. It is
nothing of the sort, however, because the DDBDP was never meant to give the text but
merely to provide an index to the printed texts. The formatting is so different from the
one found in most printed editions that only those with a trained eye can “read” the heav­

Page 7 of 17
The Future of Papyrology

ily doctored DDBDP “texts” as such by “retranslating” what they see on the screen (a lit­
tle bit like “reading” PSI volumes and “translating” the symbols { } and < > used there to
their exact opposite to match the way they are used elsewhere). It will take the concerted
effort of generations of papyrologists to rectify this in the DDBDP itself, and they would
still need a smart indexing program to navigate the results.

Ideally, the texts in the DDBDP should also reflect current critical understanding of the
texts by incorporating corrections to the printed texts made after the editio princeps. This
has already happened to some extent. In the mid-1990s, corrections to older text editions
published up to 1920 were in fact incorporated from the Berichtigungsliste der griechis­
chen Papyrusurkunden I–IX and other sources (e.g., Chrest.Mitt. and Chrest.Wilck.) in an
effort to correct the critical imbalance in the corpus of published texts. Older text edi­
tions contain relatively many misreadings, and it would have been misleading to include
their texts as they were originally printed without taking the many, often substantial pub­
lished corrections into account. The revision of older editions included in the DDBDP will
have to continue as more volumes of the Berichtigungsliste come out (so far, X and XI).
Essential help can also come from APIS-type projects, which should review texts pub­
lished from papyri in their collection also with an eye to their electronic representation
and the incorporation of corrections from the Berichtigungsliste and other sources.
Where they have not yet done so, they would have to go over the substantial body of ma­
terial already covered a second time.

A reality check might be helpful at this point: It takes several papyrologists currently five
years to reedit 500 Greek documents critically and responsibly. To do (p. 652) this for all
55,000 published Greek and Latin documents would take them half a millennium. Not all
texts are in need of reedition right away, but all of them will be within the next half mil­
lennium. Part of the effort that is going into APIS-type projects is aimed at supplying help
that is lacking in some editions. Ideally, future (re-)editions should provide from the start
all the kinds of help one might possibly need. To this end, in the appendix to my paper for
the papyrological congress in Vienna (van Minnen 2007, 712–714),1 I have presented sev­
eral sets of “ten commandments” for preparing, publishing, and using papyrus editions
that should go a long way toward standardizing and improving current editions.

To educate users (papyrologists, as well as outsiders) about what currently constitutes a

text in papyrology, there is a desperate need for a “diplomatic” handbook for the various
“papyrologies” in existence. Such handbooks exist for medieval documents (e.g., Guyot­
jeannin, Pycke, and Tock 1993 for Latin documents) but not for ancient texts. Handbooks
for Greek and Latin, demotic and abnormal hieratic, Coptic, and Arabic documentary pa­
pyri are an urgent desideratum, perhaps less so for Aramaic and Pehlevi. Handbooks for
the literary papyri (with known or new texts) in the main languages are also needed. The
older handbooks take too much for granted, and the newer ones often provide more bibli­
ographical coverage than information that one cannot find elsewhere.

There is nothing wrong with having diferent sets of common practices for documents (as
opposed to literary texts) or even for new literary texts (as opposed to known ones), but

Page 8 of 17
The Future of Papyrology

users have never been told this (e.g., the “Note on the Method of Publication and Abbrevi­
ations” printed in every volume of P. Oxy. lists practices used for the documents but not
for the literary texts that immediately follow).

The DDBDP currently does not include all Greek and Latin documentary texts that fall
within its definition. Some documentary texts were omitted because they were mistaken
for literature (e.g., dream reports); others were excluded because they were mistaken for
inscriptions (e.g., the texts on larger sherds in O.Douch). Moreover, a whole category of
texts, often called “subliterary” (e.g., magical texts), has been omitted even when they
contain documentary material as well (unlike magical handbooks, individual spells refer
to actual people, who may also occur in “regular” documentary texts). It would be better
if this material were also available in database form, either in the DDBDP or in a separate
database (or both). Some of the Greek magical texts on papyrus have now made it to the
Thesaurus Linguae Graecae, where they can be searched; the more “literary” subliterary
texts can be searched at http://cpp.arts.kuleuven.be/. For the entire corpus of magical
texts on papyrus there is now a printed dictionary (Muñoz Delgado 2001), but this is not
as useful as a fully searchable database. For the time being, “regular” literary papyri
(Homer, etc.) do not have to be included in a separate database. Standardized texts
(p. 653) of these are already searchable in the TLG. But within the context of APIS-type

projects, literary texts need to be included as full texts, thereby making it possible for tex­
tual critics to manipulate the texts of each witness.

The many APIS-type projects pose problems of their own, not least in that they contain
unique scholarly content not found elsewhere. They are extremely labor intensive and not
easy to keep up. That is one of the structural differences with the databases construed
from published texts such as the DDBDP. APIS-type projects and the publication-based
projects will therefore develop side by side for the foreseeable future rather than in tan­
dem. The former will, however, provide a critical check on the latter, and the latter will be
a useful repository of all kinds of useful information for the former. Fairly simple links be­
tween the two kinds of projects will not absolve users from doing most of the work of
piecing the information together themselves.

The “great leap forward” in the use of papyrus texts happened in the mid-1990s, when
the DDBDP replaced the awkward dictionaries and indices used until then, and images of
the published papyri in at least some collections (University of Michigan, etc.), as well as
other kinds of help, started to become available for instant use by scholars working away
from the collections. Increasingly, APIS-type projects will add more images and other in­
formation for each item included in their collections. In addition to images of at least the
published texts, they will also provide translations, copied or updated from the publica­
tions or newly made, introductions that retrace the flow of the texts, and copious annota­
tions, all of which may be lacking in the printed editions. The images are primarily aimed
at specialists, but the translations and other aids can be used by a far greater number of
scholars. Even those not ordinarily engaged in papyrological research can start using pa­
pyri responsibly (for an example, see Dijkman 2003, a study of the village scribe in Ro­
man Egypt based on the limited number of translated documents available in APIS). Even­

Page 9 of 17
The Future of Papyrology

tually all of the collections will thus be covered. The moment when images and transla­
tions of all published papyri will be available electronically to the entire scholarly commu­
nity is not too far of. That would complete the “great leap forward” made possible by the
digital revolution of the mid-1990s.

Papyri as Facts
Papyrus texts exist in many forms, as is clear by now, but whichever kind is used, papyrol­
ogists still need to integrate the data and turn the texts into usable facts. In my paper at
the international congress of papyrologists at Copenhagen in 1992 (p. 654) (van Minnen
1993), I sketched the various ways in which this is or could be done. Here I can highlight
only a few of these (see now also Bagnall 1995). Even so, this section is again almost
twice as long as the one on papyri as artifacts, and this should come as no surprise com­
ing from the pen of a papyrologist with historical inclinations.

Sometimes papyrologists literally have to put the pieces back together. If fragments of
one papyrus have been dispersed over several collections, a composite text has to be de­
vised or even published before one can use it well for some other purpose. This alone
takes as much time as editing an unpublished text. More commonly, single items from a
larger cache of documents have been dispersed over several collections, and these have
to be put back together. Papyrologists have increasingly focused on these so-called
archives, caches of papyri deliberately put together in antiquity (see Martin 1994). It is
best not to be too strict in the application of the term archive as texts may have been
found together not because they originally formed part of the same “archive” (or library)
but because they were deliberately discarded together, as in genizas, or more accidental­
ly, as in dumps. Even such seemingly hodgepodge caches can be interesting if studied as
a whole, as I showed in the case of a traditional Egyptian geniza: the so-called temple li­
brary from Tebtynis that contained hieratic and demotic literature (in actuality discarded
books from a variety of mostly private libraries) (see van Minnen 1998, 168; disputed by
Ryholt 2005). The composition of a dump can also be interesting. The more recent exca­
vations carefully record the stratigraphy of dumps, and even for the older excavations
(e.g., Karanis) data exist to reconstruct (somewhat more imaginatively, to be sure) how
the dumps there came about.

As a by-product of APIS-type projects, more data will become available about the acquisi­
tion of papyri in various collections. This will provide ever more scope for “museum
archaeology” (see chapter 10), which indicates the possible links between texts by tracing
them back to the same dealer or even purchase (e.g., as part of the German Kartell, which
bought papyri for several collections in Germany in the early twentieth century; cf. chap­
ter 3).

For the time being, putting texts from a single archive together will remain the painstak­
ing work it has always been. Archives can reveal a lot more about ancient society than a
single text. A family archive allows one to reconstruct the life of a family over several
generations (e.g., P.Fam.Tebt., to cite an early example of this kind of thing) and often
Page 10 of 17
The Future of Papyrology

gives one a better idea of the family's social position. Likewise, individual items from an
official archive are often bafffling, but once they are reassembled, they become meaning­
ful or at least understandable. The publication of P.Polit.Iud. has finally cleared up the
role of the Jewish politeumata in Ptolemaic Egypt.

Apart from texts that were deliberately put together in antiquity, there are others one can
assemble oneself. Putting similar texts (or those that relate to a particular social institu­
tion) together in a series often allows a more convincing (p. 655) picture to emerge. The
prime example is the corpus of wet-nursing contracts in C.Pap.Gr. I. Much more can be
done with this material now but only because of the work that went into assembling the
texts from a wide variety of sources.

Larger archives or types of texts for which too many examples exist will be more difficult
to assemble, but the potential of the electronic medium will allow more progress in this
area. The very large Zenon archive from the third century BCE was captured before the
development of electronic tools in the so-called Guide to the Zenon Archive of 1981. Simi­
lar projects have been contemplated for other large archives. The Dioscoros archive of
the sixth century CE is on the verge of being captured electronically. Fournet has gath­
ered images and other data on all of the published and most of the unpublished texts.
Once this detailed and painstaking work is available to other scholars, they will be able to
produce much better work, whether they are researching some aspect of late Roman soci­
ety based on texts in the archive or reediting a particular type of text. Fournet himself
has done the latter for the poems of Dioscoros (P.Aphrod.Lit., section IV) and is about to
publish a reedition of the petitions from the archive. It will not be possible to reedit all of
the texts belonging to the Dioscoros archive in one go, but they deserve to be reedited ac­
cording to modern standards, and the scholarly world will gradually come around to do
just that.

A papyrological conference held in December 2005 in Strasbourg focused on the

Dioscoros archive, and there it was clear how much is unknown, how much papyrologists
think they know but do not really, and how much they do know but have so far failed to
put together. Connecting the dots the way it was done at this conference is bound to
change our understanding of late antiquity in many ways. The concerted approach of the
Dioscoros archive by papyrologists of different persuasions has already been helped by
the electronic tools in existence, especially the DDBDP, but the electronic tool devised by
Fournet will be of even greater assistance. The Dioscoros archive is dispersed over many
collections, most of which are not currently contemplating APIS-type projects. Fournet's
database is a kind of virtual APIS-type project, a database that captures an ideal “collec­
tion”: the archive as it was originally discovered.

I have used the Dioscoros archive merely as an example. All archives (currently over 400)
are in the process of being captured in the Leuven Homepage of Papyrus Archives (http://
www.trismegistos.org/arch/index.php). This is less detailed than Fournet's database for
the Dioscoros archive, but all the kinds of things that database contains could potentially
also be made available for the other archives.

Page 11 of 17
The Future of Papyrology

Papyrologists collect texts from printed or electronic sources to write synthetic accounts
of various aspects of the ancient world (see chapter 21). I would like to stress two points:
the importance of expanding the database for any of these aspects and the focus on
Egypt.

In the first place, for any aspect of the ancient world, the number of available papyrus
texts is not yet closed by any means. So far only a small portion of all (p. 656) papyri have
been made accessible to scholars, as I explained earlier. For any topic the available data­
base can and will be expanded by a factor of five or (sometimes much) more. This is more
important than one thinks. Adding yet another tax receipt is by itself not going to change
our knowledge of the ancient world substantially, and even if minor improvements are
made to our knowledge by the continued editing of similar papyrus texts, at some point
the law of diminishing returns (cf. McGing 2006, 248) kicks in.

But all of this depends on the kind of returns one expects. Even simply increasing the
quantity of the available evidence for any aspect of the ancient world is going to improve
the quality of what one knows about it. Numbers provide a greater statistical reliability
for the conclusions one draws from the evidence. Traditionally students of the ancient
world have had to be satisfied with a limited data set for almost everything they were in­
terested in and with an ever-increasing level of sophistication in the interpretation of the
same evidence. What papyrus texts offer papyrologists is a much better deal. Not only
will there be an ever-increasing scope for sophistication in the interpretation of papyrus
texts, as with any other type of evidence, but papyrologists will also be able to put more
and more data in series and derive statistically better conclusions from them. Numbers
do count. An example of a consensus papyrologists seem to have reached not as to the
overall interpretation of the findings but as to the quality of the interpretation is the sta­
tistical use of the data included in the census documents from Roman Egypt (Bagnall and
Frier 1994). With nearly 300 census documents and more than 1,000 individuals, the con­
clusions are far more interesting than with, say, 30 documents and 100 individuals. Just
imagine how much more secure the conclusions will be when there are 3,000 census doc­
uments listing 10,000 individuals. That day will come, and future papyrologists will be
able to apply much more sophisticated statistical procedures and see whether a chrono­
logical division of the material provides insights into the demographic development over
time (at present, a chronological division of the data would undercut their statistical use­
fulness because there is no critical mass for each of the shorter periods one can distin­
guish within the first three centuries of Roman rule).

In the second place, one should not forget that the majority of papyrus texts come from
Egypt and illustrate its history—social, cultural, or otherwise. It is true that papyrus texts
have come to light in places outside Egypt as well. Nonetheless, we should not lose sight
of the difference of scale. For Egypt, the number of available texts on any topic is just
about overwhelming. Elsewhere one is at best dealing with useful snapshots. The Latin
ostraca from Bu Njem (O.Bu Njem) reveal a great deal about a Roman army unit, about
the literacy of its peculiar brand of society, and so on. But in Egypt there are multiple “Bu
Njems,” and in Libya itself the major source of information even on the Roman army is

Page 12 of 17
The Future of Papyrology

not the ostraca but inscriptions, and it is the same for many aspects of ancient life outside
Egypt. In (p. 657) other parts of the ancient world, inscriptions will always be a more im­
portant (or at least more numerous or diverse) source of information than papyri or ostra­
ca.

For many papyrologists, “Egypt” used to mean “Greek-speaking” Egypt. Texts in some
form of Egyptian, abnormal hieratic, demotic, and Coptic were the domain of Egyptolo­
gists and Coptologists. Likewise, the abundant parallel documentation in Arabic for early
Arabic Egypt was the domain of Arabists. In practice, few Egyptologists, Coptologists,
and Arabists cared about papyri. Since the 1960s, however, things have changed quite a
bit. Not that there are now large numbers of Egyptologists, Coptologists, and Arabists
editing unpublished texts from papyri (the demotists are the largest group, and they are
up against the smallest number of unpublished texts), but most papyrologists, classicists
by training, have become aware of the fact that they can ignore the parallel documenta­
tion in languages other than Greek and Latin only at their peril. The demotists have led
the way. In more recent years, very promising developments have taken place in Coptic
and Arabic papyrology. Efforts to collect Coptic documents not published in monographs
have resulted in three sturdy volumes on the model of the Greek Sammelbuch, and Arabic
papyrology now has its own series of international congresses and a database (http://
orientw.unizh.ch/apd/project.jsp).

Substantial descriptive databases are available for published Greek and Latin documents
(the Heidelberger Gesamtverzeichnis, or HGV, http://aquila.papy.uniheidelberg.de/
gvzFM.html); for published demotic and abnormal hieratic documents (http://
www.trismegistos.org/daht/index.php); and for published Coptic documents (http://
dev.ulb.ac.be/philo/bad/copte/base.php?page=accueil.php). More important, papyrolo­
gists need DDBDP-type databases for languages other than Greek and Latin because they
must be able to search the published texts in the original language. So far only Arabic has
tried to follow in the Greek and Latin footsteps of the DDBDP. The much smaller corpus of
published documents in Arabic makes it likely that one will see the completion of this
database very soon. It will provide an enormous stimulus to the publication of more Ara­
bic documents, of which there are relatively many in existing collections worldwide. I re­
alize that there are difficult problems for demotic and Coptic (just think of the various
Coptic dialects or the competing ways of transliterating demotic), but none of this is in­
surmountable (a smart indexing program could solve these problems). It will never take
more than a fraction of the time it took to compile the original DDBDP with its 50,000
published Greek and Latin documents to capture the 17,000 published documents in oth­
er languages or the 8,000 literary texts or the 5,000 recently published Greek and Latin
documents.

The focus on Egypt brings me back to the first section of this chapter—on papyri as arti­
facts. A greater awareness of the fact that the papyri come from somewhere in Egypt and
that it makes a diference whether one can identify that place either from the text or from
its archaeological provenance will improve the quality of papyrological scholarship. Even
if the general provenance of the majority of texts is known (mainly thanks to the old exca­

Page 13 of 17
The Future of Papyrology

vations at Oxyrhynchus and in the (p. 658) Arsinoite nome), increasingly more papyri will
have a more precise provenance as archaeological information becomes available for
newly found texts.

If I may draw one conclusion from the foregoing, it is that there is now more to do than
anyone imagined in the 1970s. Papyrologists have come to realize that the number of un­
published texts far outstrips that of published texts and that the number of publishable
texts is now far larger than ever before, mainly because the DDBDP for Greek and Latin
documents (and eventually similar databases for documents in other languages) allows
papyrologists to make sense of fragmentary texts. Papyrologists have also come to realize
that it is useful to provide at least limited access to unpublished material in collections
and even more access, especially in the form of images, to published materials; they also
now recognize that it is often easy enough to collect the dispersed information on pub­
lished texts in electronic form, which is especially useful as it can quickly make informa­
tion about the bulk of the published material available. Papyrologists have also learned
that it pays to reassemble texts from excavations or archives and to redo older text edi­
tions with the help of the tools now available.

For each of these realizations, there is a matching prediction. In the foreseeable future,
the number of unpublished texts in collections around the globe will be reduced by 4,500
in each five-year period (except that more texts will continue to be found in Egypt). The
total number of published texts will pass the 100,000 mark by about 2030, if one does not
count unpublished texts accessible in some form through APIS-type projects. The total
number of publishable texts is currently about ten times as many as have been published,
and this material will keep papyrologists busy for at least a millennium. An updated ver­
sion of the DDBDP, a version that includes the older material inadvertently excluded in
the current version, as well as parallel databases for documents in other languages and
for other types of texts, will make the work of papyrologists easier, better, and more com­
prehensive. Increased access to images of published and unpublished papyri will make all
kinds of papyrological work better and more reliable and will also allow papyrologists to
do more work away from physical papyrus collections. Electronic access to unpublished
texts will also allow the discovery of more links between them or with published material
than hitherto, when the discovery of such links was mainly serendipitous. Virtual guides
to archives or dossiers will soon provide the same kind of services as the Guide to the
Zenon Archive, as well as even better services: Texts and images, not just data about the
texts, will also be provided. Better data from excavation and a renewed attempt to come
to terms with older excavations will enable a clearer grasp of the context from which pa­
pyri derive. Virtual collaboration between papyrologists at different institutions will result
in better work because many archives and dossiers and certainly many types of texts add
up to too many items for one papyrologist to handle alone.

(p. 659) Still, the most important conclusion from all of this is the need for better commu­
nication. Papyrologists use different sets of common practices to represent texts in print.
These “tricks” have never been fully explained and have often scared potential users
away. In the digital era the problem has been compounded because the digital medium

Page 14 of 17
The Future of Papyrology

has increased the confusion. The development of electronic tools has added enormously
to the number of such practices. A “text” in the DDBDP is not the same thing as a text in
printed form, and one text can appear as multiple discrete “dates.” It is electronically
possible to smooth out all of the mismatches, but my prediction (that it will take half a
millennium to do so) is surely more realistic than the expectation that technology will
somehow fix our problems for us or that the input of “volunteers” will revise texts effi­
ciently and promptly. What we need in the short run is a “diplomatic” handbook for each
kind of text (literary, documentary, etc.) and for each language that explains the mis­
match between texts and “texts” and all the other things papyrologists either take for
granted or, in too many cases, are not even aware of themselves.

The future is indeed normative and delightfully so. If papyrologists do even half of the
things I have listed in the previous paragraphs (preferably in less than half a millennium),
one will be rewarded with an ever better knowledge of all kinds of aspects of life in the
ancient world and especially with the unexpected: The history of papyrology is not (only)
unremitting drudgery (and ever more of it is on its way, for at least a millennium) but (as
Keenan shows in chapter 3) also an unbroken string of exciting discoveries that usually
take one completely by surprise (as did the text I reedited in van Minnen 2000). The past
will continue to surprise into the distant future as far as the eye can see.

Bibliography
Bagnall, R. S. 1995. Reading Papyri, Writing Ancient History. London: Routledge.

———. 2001. “Archaeological Work on Hellenistic and Roman Egypt, 1995–2000.” Ameri­
can Journal of Archaeology 105: 227–243.

———, and B. W. Frier. 1994. The Demography of Roman Egypt. Cambridge: Cambridge
University Press.

Dijkman, J. 2003. “De dorpsschrijver in Romeins Egypte.” Tijdschrift voor Geschiedenis

116: 5–30.

Gallazzi, C. 1994. “Trouvera-t-on encore des papyrus en 2042?” Proceedings of the 20th
International Congress of Papyrologists, ed. A. Bulow-Jacobsen, 131–135. Copenhagen:
Museum Tusculanum Press.

Gonis, N. 2006. “Mommsen, Grenfell, and the ‘Century of Papyrology.”’ ZPE 156: 195–
196.

Guyotjeannin, O., J. Pycke, and B.-M. Tock. 1993. Diplomatique médiévale. Turnhout: Bre­
pols.

Habermann, W. 1998. “Zur chronologischen Verteilung der papyrologischen Zeug­

(p. 660)

nisse.” ZPE 122: 144–160.

Page 15 of 17
The Future of Papyrology

Martin, A. 1994. “Archives privées et cachettes documentaires.” Proceedings of the 20th

International Congress of Papyrologists, ed. A. Bülow-Jacobsen, 569–577. Copenhagen:
Museum Tusculanum Press.

———. 2000. “Das Jahrhundert der Papyrologie?” APF 46: 1–2.

McGing, B. 2006. “Papyri.” In The Edinburgh Companion to Ancient Greece and Rome,
ed. E. Bispham, T. Harrison, and B. A. Sparkes, 238–250. Edinburgh: Edinburgh Universi­
ty Press.

Muñoz Delgado, L. 2001. Léxico de magia y religion en los papiros magicos griegos.
Madrid: Consejo Superior de Investigaciones Científicas, Instituto de Filología.

Rémondon, R. 1966. “LʼÉgypte au 5e siècle de notre ère: Les sources papyrologiques et

leurs problèmes.” Atti dellʼXI Congresso Internazionale di Papirologia, 135–148. Milano:
Istituto Lombardo di Scienze e Lettere.

Rocco, B. 1975. “Un talismano bronzeo da Malta contenente un nastro di papiro con is­
crizione fenicia,” Studi Magrebini 7: 1–18.

Ryholt, K. 2005. “On the Contents and Nature of the Tebtunis Temple Library: A Status
Report.” Tebtynis und Soknopaiu Nesos: Leben im römerzeitlichen Fajum, ed. S. Lippert
und M. Schentuleit, 141–170. Wiesbaden: Harrassowitz.

van Minnen, P. 1993. “The Century of Papyrology (1892–1992).” BASP 30: 5–18.

———. 1994. “House-to-house Enquiries: An Interdisciplinary Approach to Roman Kara­

nis.” ZPE 100: 227–251.

———. 1998. “Boorish or Bookish? Literature in Egyptian Villages in the Fayum in the
Graeco-Roman Period.” JJP 28: 99–184.

———. 2000. “An Official Act of Cleopatra (with a Subscription in Her Own Hand).” An­
cient Society 30: 29–34.

———. 2007. “The Millennium of Papyrology (2001–)?” Akten des 23. internationalen Pa­
pyrologenkongresses, ed. B. Palme, 703–714. Vienna: Verlag der Österreichischen
Akademie der Wissenschaften.

Wendrich, W. Z., R. S. Tomber, S. E. Sidebotham, J. A. Harrell, R. T. J. Cappers, and R. S.

Bagnall. 2003. “Berenike Crossroads: The Integration of Information.” Journal of the So­
cial and Economic History of the Orient 46: 46–87.

Wilamowitz-Moellendorff, U. von. 1913. Sappho und Simonides: Untersuchungen über

griechische Lyriker. Berlin: Weidmann.

Youtie, H. C. 1963. “The Papyrologist: Artificer of Fact.” Greek, Roman, and Byzantine
Studies 4: 19–32.

Page 16 of 17
The Future of Papyrology

Notes:

(1) . The stray quotation at the top of p. 714 (“inconsistency is too common to be crimi­
nal”) goes with the four sets often commandments” on pp. 712–713.

Peter van Minnen

Peter Van Minnen is Associate Professor of Classics and Ancient History, University
of Cincinnati.

Page 17 of 17
Index

Index
The Oxford Handbook of Papyrology
Edited by Roger S. Bagnall

Print Publication Date: Dec 2011 Subject: Classical Studies

Online Publication Date: Sep 2012

(p. 661) Index

Page numbers in bold indicate figures.

Abbasids, 452, 458–61

Abbott, Nabia, 454, 456–57, 461, 462, 467–68
abbreviations
Christian papyri, 592, 611n11, 612n14
Christian private letters, 607
Gnosticism, 632
Greek and Latin, 170–77
Pentecost speech, 612n14
Scholar's texts, 275, 279n17
slurs, 177
and symbols, 176
Aberystwyth (Wales), 69
Abinnaeus, Flavius, 68, 135, 141, 229, 249, 423
Abraham of Hermonthis, 439–40, 434, 445
Abusir al Malaq, 33–34, 345, 404
Abydos, 397, 407
academic mathematics, 342–43
accentuation, 268, 326, 330
accounts, 370–71
Achaemenid monarchy, 398, 473
Actium, 311, 315n9
Acts of the Apostles, 595–96
ad Demonicum, 330, 595, 600
Adams, J. N., 423–24, 446n8
Addaius, 637
administrative documents
Arabic papyri, 461–63
bilingualism, 408, 524–25
declarations, 379–81
Ptolemaic Egypt, 524–30

Page 1 of 64
Index

katoikic land, 386

Latin, 422
mummy cartonnage and, 400
Near Eastern papyri, 483
official correspondence, 375–76
petitions, 377–79
public documents, 373–74
adoptions, 366
Advanced Papyrological Information System (APIS), 199, 291–92, 649, 652–54
adverbs, 295–99
Aeschylus, 24, 268, 269, 286
Afghanistan, 475, 486–87n6
afsh, 36, 40, 50
agoranomic documents, 548–50, 556, 558n13
agoranomos, 364–65, 369, 484, 543–46, 556
agriculture
accounts, 370
and capitalism, 501
census of, 527
contracts, 368–69
dead lands, 466
and excavations, 647
Islamization, 463
multilingualism, 412
one-sixth tax (apomoira), 526
survey of land, 525
taxation of, 530–32
thesauroi (granaries), 530
AIP (Association Internationale de Papyrologues), 65, 69–70, 510n11, 644
Aland, K., 598
Alcaeus, 62, 131, 268, 273, 278, 285
Alcman, copies of, 275, 285
Alcubierre, Roque Joaquín de, 304–6
Alexander of Lykopolis, 436, 608, 633, 637, 638
Alexander the Great, 180, 184, 320, 335n11, 396, 399, 404, 405, 473, 476, 535, 541, 575, 579,
590
Alexandria
administration of Egypt, 526
administrative documents, 375–76
Alexandrian chancery, 103, 104, 105
Alexandrian majuscule, 127, 129, 131, 136
and Athens, 502
birth certificates, 380
as capital, 400
cartonnage, 33–34
chancery style, 105, 120–21, 133, 141, 142
and Constantinople, 535
declarations, 384

Page 2 of 64
Index

decline of, 464

demotic documents, 556
founding, 523
and Greek language, 433
imperial cults, 575
legal Latin, 425
Octavian captures, 396
Ali, Mohamad, 31–32
aliases, 172–73
Almagest, 349–51
(p. 662) almanac, 348–49, 351

alms, 464, 637, 642

Alopex archive, 224, 227
alphabet, 150, 167, 322–24, 327, 428, 431
Amenothes, 570–71
American Society of Papyrologists (ASP), 510n11
amicitia papyrologorum, 69
Amon, 565, 577, 582
amphitheatric papyrus, 6
amphora, 15, 16, 17
amulets, 372, 456–57, 467n13, 584, 591, 595–96, 611n8, 645
anagraphê, 546–48
Analecta Papyrologica, 75
Anastasi, Giovanni, 60
anepigraphic artifacts, 41
Annales school, 507
annotations, 272–77
Anonymus Londinensis, 355–56
Antinoopolis, 31–32, 34, 194, 242, 379–80, 523, 532
Antiochus II, 525
Antioch, 333, 371, 474, 487n12, 487n14
antiquities
antiquities laws, 497–98
Antiquities Service, 34
archives, 224, 225, 226
bakshish principle, 38
desecration of, 30–31
“hot” papyri, 648
Medinet Madi codices, 638
Near Eastern papyri, 478
papyrus cartels, 66–67, 73
Antonine Constitution, 423, 425, 485, 543, 554–55, 557
Apa Johannes, 436–37, 438
Aphroditê/Aphrodito papyri, 241, 242, 383, 434, 439, 440, 444, 453, 501
Apis bull, 405, 411, 414, 569–70, 582, 584, 586
Apocryphal literature, 598–600
Apocryphon of John, 628, 629, 630

Page 3 of 64
Index

Apollonios, finance minister of Ptolemy II, 103, 218, 226, 234–35, 243, 398, 413, 414, 474, 524,
528, 572. See also Zenon archive
Apollonius (mathematical writer), 340
Apostolic Fathers, 600
apostrophe, 262, 599
apparatus, 62, 202–3, 293, 300
Arabia, 479, 483, 523–24, 535
Arabic papyri
administrative documents, 461–63
Arabo-Nabataean, 474–76
astronomy, 457, 467n12
Checklist of Arabic Documents, 211, 454
conversion, 463–64
documentary papyrus, 458–66
editing of, 455, 657
International Society for Arabic Papyrology (ISAP), 454
last of, 467n6
Latin terminology, 460
Near Eastern papyri, 488n29
and poetry, 457
publication of, 453–56
and rural history, 465–66
study of, 453
Thousand and One Nights, 457
Wadi Daliyeh, 476
Arabs. See also Arabic papyri
conquest of Egypt, 433, 435–36, 441, 464, 481
Coptic documents, 436
handwriting development, 136
names, 193
ostraca, 16
Ptolemaic era, 406
regnal years, 183
Aramaic
after Alexander, 400
Cologne Mani Codex, 633
customs accounts, 404
Engadi caves, 479
Near Eastern papyri, 489n44
parchments, 11
Persian empire, 395–99
Petra papyri, 480
shipping, 412
Simon Choseba dossier, 479
technical vocabulary, 407
arbitration, 38, 387, 439, 440, 441, 447n28, 490n55
archaeobotanists, 41–42
Archaeological Institute of America, 499

Page 4 of 64
Index

archaeology
anepigraphic artifacts, 41
and archives, 226–27, 245–46
context of Oxyrhynchus excavations, 62
emergence of, 36–38
handwriting development, 102
Karanis excavations, 35, 38–39
museum archaeology, 228–29
archaeozoologist, 42
Archimedes, 339–40, 343
architecture, 32, 303, 314n2, 566
archives. See also bilingual archives
administrative documents, 377–81
antiquities excavation, 224, 225, 226
The Archive of Aurelius Isidorus, 73
archivization, 498
Babatha archive, 50, 51–52, 367, 476, 479
bilingual archives, 407–9
classifying, 243, 245
composition of, 237–40
definition, 217–18, 654
Dionysios archive, 548
Dioscoros archive, 439, 440
vs. dossier, 219
Dryton archive, 225, 229, 240, 388n25, 499
Dura-Europos, 477–78
editions, 229–30
(p. 663) excavations, 47–49

guides, 230, 247

Heroninos archive, 231, 235, 362
Karanis, 245–46, 423
Leuven Homepage of Papyrus Collections (LHPC), 230, 248, 655
Melitian monastery of Hathor, 435
museum archaeology, 228–29
loss of official archives, 360–61
original depositories, 219–24
Pachymios archive, 439
Pathermouthis archive, 439
real estate, 364
reconstruction of, 226–29, 247
Seleucids, 487–88n23
Totoes archive, 223
types, 231–37
usefulness of, 216, 247–48
virtual collaborations, 658
Arden papyrus, 257, 258
Arethusa, 295
Aristophanes, 132, 167n13, 269, 329

Page 5 of 64
Index

Aristotle, 167n13, 268, 269, 288, 309, 311, 356, 488n24, 614n55
aroura, 177, 179, 185–86, 234, 489n47, 528, 531
Arsinoe Philadelphos, 525
Arsinoite nome (Fayyum). See also Karanis; Tebtynis
administration of Egypt, 530
Christianity, 600–601, 605
death notices, 381
declarations, 384
deeds of divorce, 367
dialect, 209
dike-work certificates (penthemeros), 384
excavations, 648
first Fayyum find, 32
geography, 184–85, 523
Greek soldiers' settlement, 401
irrigation system, 384
lease contract, 369
liquid measurement, 188
nomarchs, 529–30
prosopography, 194
Roman era, 532
artaba, 177, 179, 186, 187, 382, 407
asceticism, 636–37
ASP (American Society of Papyrologists), 510n11
Assiut (Lykopolis), 44, 637, 639
Association Internationale de Papyrologues (AIP), 65, 69–70, 510n11, 644
astrology
and astronomy, 349–52
and Christianity, 586
Hellenization, 580–81
and Old Coptic, 431
papyri, 339
Roman period, 345–46
temples, 566, 584
astronomy, 339, 343–46, 349–52, 457, 467n12, 566
Aswan, 32, 184, 290, 397
asylum, 575, 581
Athanasius of Alexandria, 593, 599–600, 608–9, 616n97
Athena, 577, 583, 578
Athenaion Politeia, 288, 299n4
Athenian, measure, 187
Athenian ostraca, 14, 15
Athens, 12, 157, 276, 313, 320, 502, 550, 577
Attic dialect, 150, 154–55, 157, 159–61, 187–88, 288, 298, 327, 331
auditoria (classrooms), 321
Augustan papyrus, 6
Augustan writing style, 111
Augustine, 335n5, 590, 632–33

Page 6 of 64
Index

Augustus, 6, 34, 45, 109, 190, 214n12, 264, 374, 381, 484, 575–76
Aurelian, 474–75
Austrian National Library, 60, 99
authors, 329–32
autopsy, 379
Babatha archive. See archives
Babylonia, 350, 473, 487n23, 580
Bacchias, 40
Bacchylides, 232, 273, 278n13, 285
Bactria, Bactrian language, xviii, 11, 461, 473, 482, 490n61
Badr, 458
Baghdad, 452, 458
Bagnall, R. S., 496, 500
Bagnall, W. S., 227
Bahr Yusuf, 523
Bailey, D. M., 39, 43
bakshish principle, See also antiquities, 38
balloting, 14
Banaji, J., 501, 512n41
banking transactions, 364, 371
baptism, 615n71
Bar Kokhba
archives, 49–50, 51, 52, 219
coinage, 489n47
document cache, 49–50
Hebrew language, 474
legal documents, 490n72
Near Eastern papyri, 476, 479, 489n44
Barns, J. W. B., 593
Bartoletti, Vittorio, 71
Basile, Corrado, 8, 25n8
basilikos grammateus, 376, 378, 384–85, 526, 527, 533
Bayardi, O. A., 305
Becker, C. H., 454
beer, 43, 206
Bell, H. I., 66, 68, 70, 229, 599, 616n99
Belzoni, G. B., 497
Berenice, 648
Berichtigungsliste der griechischen Papyrusurkunden, 65, 198, 212, 651
(p. 664) Berlin Codex, 630–31

Berlin Museum, 60–61

Berliner Klassikertexte (BKT), 211
Berytus, 334
biblical majuscule, 128–29, 130, 131, 132, 136, 145, 146n11
Bibliographie papyrologique, 209
Bibliotheca Teubneriana, 310
bikos, 186
bilingual archives, 241, 242. See also bilingualism

Page 7 of 64
Index

antiquities trade, 224, 226

guides, 230
Hellenized Egyptians 216
legal documents, 387
Leuven Homepage of Papyrus Collections (LHPC), 230
publication of, 229–30
title deeds, 228
bilingualism See also diglossia
administrative documents, 408, 524–25
in Alexandria, 400
Apa Johannes letter, 438
Arabization, 459
archives, 407–9
Babatha archive, 479
Christian papyri, 612n26
contracts, 409, 445
dreams, 414
education, 442–44
family archives, 409
Glaukias, 411, 414
Greek-Arabic documents, 461–62
Kellis papyri, 429, 434, 437
koine Greek, 327, 483, 591
Latin-Greek, decline of, 423
legal documents, 377, 437
literacy, 444–45
literary papyri, 444
monasteries, 434
Nessana dossier, 459, 480–81
Palmyra, 487n10
Pehlevi documents, absence in 420–21
school exercises, 334, 443
school texts, 442–44
scribes, 408
Soknebtynis dossier, 504
tax receipts, 408
translations, 445
urban/village status, 434
Zenon archive, 408–9
billeting, 410–11
Bingen, Jean, 69, 72–73, 246
biography, 283, 291, 633, 635–36
birth certificates, 360, 379–80, 389n57, 423, 462
birthdates, 240, 346–48
bitumen, 80
Blanchard, A., 170, 173n1, 175
blotting paper, 82–83, 86, 89–90, 97
Boak, A. E. R., 35, 67, 73, 615n81

Page 8 of 64
Index

Bodmer papyri, 287, 443, 612n26

Bonaparte, Carolina, 309
Bonaparte, Napoleon, 30, 55n2, 309–10, 314n5, 316n25
book hands, 105, 257, 324, 347, 459, 632
Book of the Dead, 21, 83, 586
Book of the Temple, 568, 586
bookrolls. See also desiderata
as ancient writing material, 19
Arden papyrus, 258
authors, 268, 269
book hands, 257
columns, 257, 259, 260, 261, 263
computer-assisted tomography, 314
dimensions of, 259
Herculaneum papyri, 264, 305–8, 313, 315n9
lectors, 265
length of, 264–65
literary papyri, 267
lyric texts, 277n2
Maas's law, 258, 261
manufacture of, 256–57
margins, 263, 277n5
as material for codex, 25
multispectral imaging (MSI), 313
and muse, 20
production of, 18–23
punctuation, 261–63
replacement of, 24, 256, 430, 631
scholar's texts, 273–74
title tags, 264
tombs, 44–45
Books of Jeû, 631
Borchardt, L., 44
Borgia, Stefano, 59. See also Charta Borgiana
Boubastos, 44, 92, 94, 95, 185
boulê (town council), 184, 375, 390n81, 533
Bowman, A. K., 513n66
B-R uncial, 145
brackets, 202, 203, 455
British Library, 145, 212, 631
British Museum, 47, 61, 66, 69
Brussels, 67–69
Bu Njem, 140, 446n5, 656
Buchis bull, 569–70, 586
Buddhism, 633, 637
Bureaucracy. See also agoranomos; basilikos grammateus
Arabic, 461–62
Byzantine period, 535

Page 9 of 64
Index

Ptolemaic era, 400

scribal training, 407
business correspondence
Christian, 602, 603
private documents, 362
registration system, 243
Zenon archive, 234–35
Byzantine papyrology. See papyrology
Byzantine period
administration of Egypt, 534–35
(p. 665) Aphroditê/Aphrodito papyri, 501

arbitration, 387, 490n55

archives, 241
dry measure, 186
Greek and Coptic documents, 154–55
leases, 369
loan contracts, 369
magical texts, 584
Near Eastern papyri, 476, 484–86
school exercises, 330
cadastral reference system, 528
cadastre of Aphrodito, 383
Caecilius Statius, 311
Cahen, Claude, 454
Cailliaud, Frédéric, 31
Cairo antiquities market, 33, 60, 66, 67, 626
Cairo codex, 287
Cairo Museum (Egyptian Museum), 5, 226
calamus, 18, 19, 26n12
calcium burns, 87
calcium carbonate, 87, 88, 89
Calderini, Aristide, 198
calendar, 159, 180–83, 209, 214n12, 339, 344–46, 350, 482
calligraphic writing. See also Augustan style; biblical majuscule; Gnomon of the Idios Logos
Antonine papyri, 127
Coma Berenices of Callimachus, 112
development of, 101–2
Ebers papyrus, 353
emergence of, 106–7
Latin handwriting development, 138–39, 141
literary papyri, 112, 113, 115, 116, 117
Oxyrhynchus papyri, 123, 126
Roman uncial, 128
Callimachus, 48, 112, 129, 206, 264–65, 269, 277n6, 287–88, 329
Cambyses, 395–96, 398
cantina deipapiri (Tebtynis), 227, 228
Cappadocians, 595
Caracalla, 374, 381

Page 10 of 64
Index

carbonized papyrus. See also papyrus; Petra papyri

archives, 231
conservation of, 91–98
Herculaneum papyri, 306–8
official correspondence, 375
restoration techniques, 84, 95–98
cardboard backing, 85–86
Carians, 397–98, 404–5, 412
Carmen de bello Actiaco, 139
Caromemphites, 404–5
Carter, Howard, 47
cartonnage
reused archive, 221
conservation and restoration of, 88, 89–91
crocodile mummies, 45, 46, 47
dating of papyri, 88
epigrams, 287
under Greek rule, 400
Gurob cemetery, 33–34
handwriting and, 102
mummy, 40, 43, 80, 87–88, 89–91, 102, 285, 287, 406
production of, 87
Sappho poetry, 285
Tebtynis, 45
Timotheos papyrus, 286
Caspians, 397, 399
Catalogue of Paraliterary Papyri (CPP), 291
catechism, 331, 600
catechumens, 601, 636–40, 642
catfish, 42
Catullus, 287, 290
caudate hastas, 111
cave of horror, 479, 489n48
cave of the letters, 49–50, 51, 219, 479
Cavenaile, Robert, 212
cedar oil, 83
cellulose treatment, 82–83
census, 227, 373, 380–1, 462, 479, 527, 574, 656
Centre de Documentation de Papyrologie Littéraire (CEDOPAL), 199, 292
Centro Internazionale per lo Studio dei Papiri Ercolanesi, 71
ceramics, 4, 42
Chalcedon, 446n19
Champollion, Jean-Francois, 31, 55n3, 60
chancery style
Byzantine period, 135–36
imperial writing style, 120–21, 125
Latin handwriting development, 139–42, 143
writing development and, 111

Page 11 of 64
Index

Zenon archive, 105

chaos, 561–62, 565, 629
Chariton, 288
Charles III, 304, 307
Charta Borgiana, 30, 59, 98
Chartae Latinae Antiquiores, 212
Checklist of Arabic Documents, 454
Checklist of Editions of Greek, Latin, Demotic, and Coptic Papyri, Ostraca, and Tablets, xviii,
211, 299n8, 388n7, 397, 446n16
Chester Beatty papyri, 443, 596, 597, 612n26, 638
child-donation contracts, 508
choachytai/choachytes 60, 219, 230, 235–36, 565, 571
Choat, M., 601–2
choinix (pl. choinikes), 186, 187
chôra, 528, 543–44
chous, 188
chrêmatismoi, 545–46
Chremonidean War, 525
Christianity. See also codices; Constantine; conversation manuals; nomina sacra
amulets, 591, 595–96, 611n8
Apocryphal literature, 598–99
Arabization, 460
business correspondence, 602, 603, 604
(p. 666) Chalcedonian, 433, 446n19

chancery style, 136

condolence letters, 363
contracts, 387
and Coptic, 431–34
cryptograms, 607, 615 n88
and demons, 587
Dioscoros archive, 242
documentary papyri, 601–7
and Egyptian religion, 584, 586–87
and Gnosticism, 625–26
heretics, 623–24
legal documents, 461
literary papyri, 591–601
and Manichaeism, 637, 642
Melitian schism, 608–10
names in papyri, 193
as official state religion, 623
order to arrest, 604–6, 615n76
and Oxyrhynchus excavations, 61–62
patristic literature, 599–601
persecution of, 373, 605–8
private letters, 601–3, 607, 614n62, 616n89
restoration of property, 534–35
school exercises, 328, 593–95, 612n26

Page 12 of 64
Index

school texts, 600

scribes, 592–93
The Shepherd of Hermas, 591, 600, 614n58, 614n59
and slavery, 365
Chronique dʼÉgypte, 67
chronology, 180–83
Chrysippus, 107, 311
Chrysostom, John, 324
Church History, Manichaean, 638
Cicero, 141, 163, 167n12, 289, 311, 317n34, 334, 336n17, 428
circumcision, 380–81, 389n60, 573, 586
citizenship, 161, 380, 423, 542, 543, 554–55, 557, 605
civitas, 534
Clarysse, Willy, 230, 238, 246
classical literature, 24, 135, 150, 266–67, 282–300, 309, 328, 600, 612n23, 612n26
Claudian papyrus, 6
Clement of Alexandria, 598–600
Cleopatra, 34, 376, 396, 570
cleruchs, settlement of, 401–2, 412, 528. See also klerouchos
clothing, 42, 371, 436, 462, 465, 513n64, 572, 606, 615n71
codex. See also Chester Beatty papyri
ḥadīth collections, 456–57
authors, 268, 269
biblical majuscule, 129
Christianity, 266–67, 592, 611n8
Codex Askevianus, 631
Codex Brucianus, 631
Codex Jung, 626–27
Codex Washingtonianus, 596
columns in, 266
development of, 23–24, 25, 265–67, 278n8
Jewish, 592
Matthew, 591
New Testament, 595–96
Oxyrhynchus excavations, 62
patristic literature, 599–600
Pauline epistles, 598
Psalms, 593, 594
punctuation, 266
reconstruction of, 25
replaces bookroll, 24, 224, 256, 266–67, 430, 631
restoration techniques, 91
tax receipts, 224
tax registers, 383
Tura codices, 599–600
waxed tablets, 12
coinage, 37, 189–91, 489n47
Collart, Paul, 71–72, 73

Page 13 of 64
Index

collation, 272, 275, 277

Collombert, Philippe, 402
Cologne Mani Codex, 633, 634, 635–36, 638
Columbia University, xx, 66, 649
columns (selides, paginae), 14, 21, 59, 200, 204, 206–7, 214n10, 257, 258, 259, 260, 261, 263,
266, 278n9, 308, 309, 316n20, 330–31, 348, 353, 377, 490n74
Coma Berenices of Callimachus, 112
commentary (hypomnema), 114, 200, 208–9, 275, 277, 291, 325, 364, 379
commerce, 39, 419, 464–65, 590
common script, 140, 141–42
comparanda, 297–98
compound words, 173, 277n4, 296
compulsory services, 364, 379, 384–85
computer-assisted tomography, 314
concertina tablets, 11, 13, 14, 23, 27n42
condolence letters, 362–63
Conservation Service, 31
consonants, 152–53, 162–63, 173, 277n4, 324, 444, 599
Constantine the Great, 189, 191, 422, 433, 535, 554–55, 586, 616n97, 623
Constantinople, 129, 136, 373, 382, 422, 501, 535, 555
Constitution of Athens, 44, 61
consulate, 183
context, 499, 645–46
Contra Haereses, 628
contracts. See also notaries (agoranomoi); bilingualism
adoptions, 366
agriculture, 368–69
Arabic papyri, 465
contract keeper (syngraphophylax), 363
contracts of deposit, 369
divorce contracts, 238, 367
freemen, 365–66
Hellenization, 542
loans, 238, 239, 364, 368–70, 386, 388n28, 437, 439, 479, 543–44, 548, 554, 558n14
marriages, 366
Pehlevi documents, 419
property ownership, 367–70
(p. 667) Roman policy in Egypt, 557

slavery, 365, 368

tabellio, 364, 386, 485
wet-nursing contracts, 655
conversation manuals, 428, 444
conversion, 433, 461, 463–64, 466
copper, 18, 87, 383,
Coptic. See also Gospel of Thomas
Alexandrian majuscule, 131
in Arabic Egypt, 436, 453, 459–60, 467n3
Arabic names, 193

Page 14 of 64
Index

Arabization and conversion, 463

archives, 240–41, 242
Checklist of Editions, 211
child-donation contracts, 508
and Latino-Greek in Christian milieus, 428–31
calendar, 180
deeds of surety, 387
development of, 435–37, 439–42
documents, 3, 4, 16, 192, 199, 229, 359, 434–442, 446n16, 453, 460, 657
editing of, 657
electronic databases, 199
emergence of, 430–31, 432
first dated document, 440
and Greek, 430–45, 435, 442, 443, 631
Greek pronunciation, 153
handwriting development, 444
as language of the monks, 433–34
legal documents, 441, 442, 541, 460
see Lexikon der lateinischen Lehnwörter, 429
as liturgical language, 460
Manichaeism, 639–40
Nag Hammadi Codices, 428, 627–31
place names, 185
school exercises, 328
Sophia Jesu Christi, 630
technical vocabulary, 194
Coptic Museum (Cairo), 627
copybooks, 328, 335n9, 375
Cornelius Gallus, 139, 290, 522
Cornell University, 66
Corpus dei Papiri Greci di Medicina, 354
Corpus Inscriptionum Iranicarum (CII), 419
Corpus Papyrorum Graecarum, 212
Corpus Papyrorum Latinarum, 212
Corpus Papyrorum Raineri, 211
corrector, 274
cosmic order, 561–62, 565
costrels, 42
Cotton, Hannah, 480, 483, 486n1
court proceedings, 238, 376, 543, 546–48, 550–52, 556–57, 607
creation, 624–26, 629, 636
Cretans, 398–99, 411, 582, 584
Cribiore, Raffaella, 320, 325, 335n7, 500, 593, 595, 612n26
Crisci, E., 481
crocodile as religious symbol 373, 405, 565, 569, 578–79, 580, 583
mummies, 46, 47, 80, 221–22, 226, 231, 247, 566, 574
Cronache Ercolanesi (CErc), 310, 314n1, 317
cryptograms, 607, 616n88

Page 15 of 64
Index

cubit, 4, 6, 185, 186, 187, 218, 341

cults. See also Amenothes
imperial, 575–76, 576
mortuary, 565
sacred animals, 569–70
Tutu, 583
Virgin Mary, 586
cuneiform script, 350, 352, 477, 488n32
currency, 158, 175, 177, 189, 190–91, 485
“Curse of Artemisia,” 405
cursive writing
abbreviations, 171
Arabic papyri, 455
bookrolls, 257
Byzantine period, 135–36
common script, 140, 141–42
cursivization, 109, 112
and decipherment of texts, 200–202
emergence and development of, 101–2, 106–7, 109, 111, 241
Greek, 443, 444
Latin handwriting development, 139–40
Latin papyri, 137
Near Eastern papyri, 481
Pehlevi documents, 419
Ptolemaic style, 106
restoration techniques, 203
Roman cursive, 115, 120, 121, 122
school exercises, 334
customs duties, 383
Cuvigny, Hélène, 648
Cyrenaeans, 411–12
Cyril of Alexandria, 599–600
Cyrus, prefect, 422
Dakhla Oasis, 437, 583, 639, 648
tablet books, 12, 14, 23
Dakhleh Oasis Project, 56n12, 648
Daniel, Book of, 591, 600
Daninos, Albert, 47
dating
abbreviations, 175, 177
papyri of Psalms, 593, 595
Christian private letters, 614n62
chronology, 180–82
commentary and, 209
fiscal year, 183
New Testament, 596, 613n31
Old Testament, 595
palaeography, 649

Page 16 of 64
Index

regnal years, 486

dative, 154, 157, 159, 160, 165
Davis, Natalie Zemon, 508
(p. 668) Davy, Humphry, 307

DDBDP (Duke Data Bank of Documentary Papyri), 192, 194, 198–99, 212, 521, 649–59
dead lands, 466
Dead Sea, xviii, 453, 473, 479, 489n44
caves, 56n21, 475–76, 478
scrolls, 474, 476
death notices, 381
decans (constellations), 345, 580–81
Decius, 584, 586, 605–6
Edict, 373
declamations (meletai), 325, 332–33, 335n15
declarations, 177, 227, 228, 243, 367, 379–81, 383–84, 389n54, 441, 479, 497, 547, 606
declensions, 153–55, 157, 165, 171–72, 325
decorations, 322–23, 576
deeds of surety, 386–87
Deir el-Bahari 236–37, 571, 586
Deir el-Medina, 223
defensor civitatis, 534
dekaprotoi, 533–34
dʼElboeuf, Prince of Austria, 303–4
demes, 185
demons, 562, 572, 583–84, 586–87, 600
Demosthenes, 26n34, 161, 167n13, 204, 264, 268, 269, 335n15
demotic writing
abbreviations, 175
administrative documents, 524–26
agoranomic documents, 558n14
at Alexandria, 556
archives, 220, 221, 222, 226, 229–30, 236, 240–41, 242
astronomy, 346, 351–52
bookrolls, 19, 21
in Checklist of Editions, 211
court proceedings, 547–48
decline of, 236, 241, 242, 430, 549, 555–57
Demotic Chronicle, 284
dream books, 571–72
editing of, 657
Egyptians and Amazons, 284
horoscopic astrology, 347–8, 351–2, 581
legal documents, 399, 406, 541–43, 546–49
literary papyri, 283
mathematics, 342
measures, 186
medical texts, 353
Myth of the Eye of the Sun, 284

Page 17 of 64
Index

Oracle of the Lamb, 284, 576

oracle questions, 579
ostraca, 14–15, 245
Papyrus Amherst, 397–98
in the Persian empire, 399
resistance to Greek loanwords, 506–7
rush pen, 408
Tebtynis, 513n60
war of scepters archive, 219
in Zenon archive, 408–9
denarius, 158, 189, 190–91, 263, 485
Derveni papyrus, 45, 92, 94, 103, 288, 313
Deutsches Papyruskartell, 66
di Sangro, Raimondo, 307
dialects, 149, 159, 192, 209, 262, 405, 407, 431, 482, 611n3, 628, 638, 657
diaries (commentarii), 360, 374, 384
diastole, 262
dichôra, 43
dicolon, 261–62
Didymus Judas Thomas, 630
Didymus the Blind, 593, 599–600, 614n54, 614n68
Diegeseis, 287
Diels, H., 310
Diem, Werner, 454, 459
Dietrich, Albrecht, 454
digitization, 649, 653
diglossia, 327, 419, 421–30, 427, 434, 446n8. See also bilingualism
dikastêrion, 546
Diocletian. See also Edictum Diocletiani de pretiis rerum venalium
and bilingualism, 424
coinage, 190–91
Latin professorships, 425
Latinization, 422, 428
persecution of Christians, 606–8
property ownership, 386
regnal years, 183
Romanization, 422
taxes, 382
Diodorus's circle, 270–71, 277, 278n13
Diodorus Siculus, 309
dioiketes, 474, 524, 526, 528, 531–33
Dionysia dossier, 551–52
Dionysios Petosorapis, 576–77
Dionysios, son of Kephalas, archive, 226, 230, 235, 548
Dionysius of Alexandria, 601, 615n82
Dionysius Thrax, 295–99, 332
diorthôtês model, 274
Dioscoros of Aphroditê

Page 18 of 64
Index

archive, 241–42, 246, 655

bilingualism, 427, 434, 444
Coptic, 439, 440, 447n27
handwriting development, 241–42
Kôm Ishqaw, 49
Dioskourides, 378, 402–3
diphthongs, 150, 152, 162
diplomatic handbook, 652, 659
diplomatic transcript, 203–5, 293–94
diplomatics, 480, 482, 484–85. See also double document
disegni, of Herculaneum papyri, 307
diviso acu, 8, 26n8
divorce contracts, 238, 367, 388n22, 554–55
Dizionario dei nomi geografici, 198
document cache, 49–50
(p. 669) documentary papyri. See also Duke Data Bank of Documentary Papyri abbreviations, 171

Arabic papyri, 458–66

book hands, 263
Christian, 601–6
classifying, 359–60
dating, 209
decipherment, 202
dreams, 652
Dryton archive, 499
editing of, 211
Greek handwriting development, 103
Greek language, 149–50, 154, 157–58
imperial writing style, 120, 125
Latin, 139, 142, 149–50, 162–64, 645
numerical data, 342
Pehlevi, 419–21
publication of, 358–59, 645
restoration techniques, 206–7
severe style, 71, 132
translations of, 650
types of, 361
Wörterbuch, 198–99
Doric dialect, 286, 405, 407
dossiers. See also choachytai/choachytes; Kronion and Isidora; Nessana dossier
definition, 218–19
Kôm Gamman, 513n56
Near Eastern papyri, 479
ostraca, 219, 246
Soknebtynis priests, 504, 513n55
dots
diacritical, in Arabic papyri, 455
Leiden system, 202, 203, 214n7
punctuation, 261, 277n3

Page 19 of 64
Index

ruling dots, 261

double documents, 485, 544, 548. See also diplomatics
double names, 579
dowry, 363–64, 366–67, 369, 410, 479, 550–51, 556
drachma/ drachmê (weight/currency), 175, 177, 179, 188, 189, 190–91, 272, 485, 574–75
dreams, 240, 245, 414, 566, 571–72, 577, 582, 584, 652
Nectanebo's, 400
Nektembes, 414
Drovetti, Bernardino, 55n1, 60
dry measure, 186–87
Dryton archive, 499
D-scholia, 331–32
ductus, 101, 103, 105, 107, 109, 110, 111, 114, 120, 127, 129, 131, 136, 141–42
Duke Data Bank of Documentary Papyri. See DDBDP
Duke University, 649
Dura-Europos papyri
administrative documents, 385–86
archives, xvii, 47
literary papyri, 300n12
military history, 385–86
Near Eastern papyri, 475–78
ostraca, 477
parchment, 11
published, 74
Sassanid invasion, 477
writing style, 120, 125
Yale excavations, 67–68
Easter, 602, 615n71
Ebers papyrus, 353–54
eclipses, 346, 352
Edessa, 474, 478, 483, 487n10
Edfu, 219, 241, 397–98, 400, 453, 502, 562, 565, 569, 573–74
donation text, 502, 573–74
Edgar, C. C., 226
edict, xvii, 123, 373–4, 381–82, 389n45
Edictum Diocletiani de pretiis rerum venalium, 257, 263
education, 320–33, 442–44, 524. See also Cribiore, Raffaella, school exercises
Egypt. See also irrigation system; Islamization
administrative documents, 524–30
antiquities laws, 497
and Antonine Constitution, 554–55, 557
Arab era, 435, 441, 466, 523–24, 535
astrology, 346–52
astronomical papyri, 343–46
bilingualism, 327, 395–445
Cambyses conquers, 395–96
charred papyrus, 92–98
and Christianity, 431–34, 590–622, 627–28

Page 20 of 64
Index

chronology, 180–83
cleruchic settlement, 528
coinage, 189–91
conquest of, 180, 399, 590
Egypt Exploration Fund/Society (EEF/EES), 33–34, 36, 61–63, 645
epikrisis registers, 380
ethnic identity, 432–33
geography, 184–85, 522–24
and Gnosticism, 598, 613n44, 624, 626–32
Greek documentary language, 162
Greek population of, 401
Hellenization, 524, 530, 542–43
insurgencies, 576
land registration, 546
Latin papyri, 137, 162
law of the Egyptians, 550–52, 555–57
legal documents, 541–45
literacy, 359–60
literary papyri, 282–84
Macedonian conquest, 158
Manichaeism, 632–42
military history, 386
monasteries, 608, 637
multilingualism, 395–445
municipalization, 532–33
ostraca, 14–16, 40–44, 53, 245–46, 322–24, 343
(p. 670) papyrus origin, 5

peregrines, 543
publication of papyri, 211
regnal years, 182–83, 491n77
Roman era, 422, 484–85, 504, 532–33, 553–55
salt tax, 382
Sassanids, 419
Seven Nomes, 523
Sonderstellung problem, 64–65, 500, 512 n40
status of Egyptians, 530
taxes, 382–83
writing development and, 109, 111–12
Egyptian language. See also demotic; Coptic
administrative documents, 524–25
Arabization of Egypt, 460
decline of, 430–31
legal documents, 542
names, 191–93
Ptolemaic era, 406
technical vocabulary, 194
Egyptian religion. See also cults
amulets, 584

Page 21 of 64
Index

and Christianity, 584, 586–87

demons, 583–84
disappearance of, 582–87
dreams, 571–72
Edfu temple, 219, 502, 562, 569
Greek inscriptions, 581–82
Hellenization of the gods, 578–80
interpretatio graeca, 522–23, 577–78
listening gods, 568, 570
magical texts, 583–84
oracles, 563, 571
sacred animals, 569–70
saints, 570–71
temple administration, 573–76
variations, 565
wisdom literature, 583
worldview, 561–64
Egyptology, 60–61
Eighteenth International Congress of Orientalists, 68
eisthesis, 262
ekdosis, 366
ekphrasis, 332
ekthesis, 262
electi, 636
electrolysis, 97
electronic databases. See also Advanced Papyrological Information System; DDBDP (Duke Data
Bank of Documentary Papyri); Leuven Database of Ancient Books (LDAB)
Bibliographie papyrologique, 209
Catalogue of Paraliterary Papyri (CPP), 291
images, 648–49
Leuven Homepage of Papyrus Collections (LHPC), 230
Mertens-Pack, 285, 292, 300n12
elegy, 285, 290, 294
Elephantine, 103, 220, 343, 397–98, 409–10, 420, 565, 582
Eloquent Peasant, 283
emancipations, 365–66
employment contracts, 364, 369–70
emporitic papyrus, 6
encomium, 332–33
endogamy, 504, 506, 513n58
Engadi, 479
enguetes (surety), 530
enkyklios paideia, 321, 328
Ennius, 311
enteuxis, 377–78, 526,
enzyme treatment, 86, 90–91
epagomenê, 180, 181, 182
ephemeris, 349

Page 22 of 64
Index

Epicurus, 303, 310–13, 317n34

epicycle, 350
epigrams, 105, 261, 287, 309, 311, 317n34, 498
epigraphê, 533
epikrisis, 208, 360, 363, 380, 389n59, 504
Epiphanius of Salamis, 623–26, 632
Epistles of Mani, 638
epistolary courtesies, 160
epistolary writing, 330
epistrategos, 184, 385, 551
epoikion, 184
epsilon-theta style, 114, 119
Erasistratus, 355–56
Erinna, 286
erôtêmata, 331
ethics, 498
ethnic identity, 432–33
ethnos (caste), 524–25
êthopoiiai, 332–33
Euclid, 340, 342–43
Eudoxus papyrus, 344–45
Euripides, 105, 268, 269, 286–87, 327, 309, 327, 329, 603
Eusebius of Caesarea, 600–601
evolving hand, 324, 330–31
exactor civitatis, 390n81, 534
excavations. See also individual sites
context destruction, 646–47
Egypt Exploration Fund (EEF), 33–34
house-to-house-approach, 48
illicit excavations, 33–35, 40, 67, 247
multidisiplinary, 41–44
papyrus conservation, 81
pre-WWI, 36
rescue excavations, 647
Exodus, Book of, 591, 595, 598
fables, 283, 330, 488n27
Fachwörter, 198
Fackelmann, M., 94, 96
(p. 671) family archive

Aramaic, 398
definition of, 218
description of, 233–34
editing of, 654
legal documents in, 238
mixed marriages and, 411–12
of mortuary priests, 565
of soldiers, 241
family law, 557

Page 23 of 64
Index

“Fannius' clever workshop,” 6–7

Fayyum. See also Arsinoite nome; Karanis; Tebtynis; Zenon archive
administration of Egypt, 529
Arabic papyri, 453, 465
archives, 232
Bahr Yusuf, 523
Egyptian religion, 565
Fayyum portraits, 500, 511n35
Fayyum Villages Project, 523
first Fayyum find, 60
gate of justice, 568
Kimân Faris, 60
Medinet Madi codices, 638
miracle year, 61
mummy cartonnage, 401
naming of, 522–23
Pehlevi documents, 419, 420
sebbâkhîn, 32–33
toparchies of, 534
Febvre, Lucien, 496–97
female students and teachers, 328
Ferdinand, 310
Ferguson, C. A., 446n8
Fikhman, I. F., 70–71, 74
Filmoplast, 85, 97
fiscal year, 183
foodstuffs, 41–42
foreclosures, 543–44, 556
Fouad papyri, 71–72
Fournet, Jean-Luc, 655
fractions, 177, 186, 328, 340–41, 350
fragmentary texts, 226, 658
Frantz-Murphy, Gladys, 454, 466
freemen, 365–66
frog lamps, 43
Frösén, Jaakko, 95–98
full transcript, 203–4, 206
funerary practices, 80, 372, 500, 511n35, 565
Fusṭāṭ, 453, 462, 465
Gadamer, H.-G., 497
Gaius Cassius, 18
Galen, 355–56
Galerius, 534
Gallazzi, Claudio, 40
gallnuts, 18
Gamble, H. Y., 267
garbage dumps
archives, 218, 221, 231, 654

Page 24 of 64
Index

context, 40, 48, 80, 222, 247

Didymoi, 54
importance of, 50–55
Mons Claudianus, 41
oracle questions and, 571
ostraca, 245–46
Oxyrhynchus excavations, 62, 273–74
papyrus storage, 80
Gardner, I., 601–2
Geb names, 514n77
gender, 155, 192, 512n51
Genesis, Book of, 123, 129, 565, 591, 595, 598–99, 604
Geneva, 66
Geneva papyri, 212–13
Geniza documents, 464
geniza, 513n57, 654
geography, xvii, 44, 184–85, 198, 209, 478, 502–3
administrative, 521–35
geometrikon, 186
geometry, 342–43
gesso plaster, 87, 90
Ghoran papyri, 528
Gigante, Marcello, 71, 310, 316n28
Glaukias, 414
glossaries, 150, 194, 331, 334, 428, 444, 481, 490n57
Gnomon of the Idios Logos, 123, 127, 374
Gnostics
Books of Jeû, 631
Coptic, 433
in Egypt, 598, 613n44
and the gospel of John, 598
as heretics, 624
Nag Hammadi Codices, 627
Pistis Sophia, 631
private letters, 601
goldbeater's skin, 308
Gospel according to Mary, 631
Gospel of Thomas, 598–99, 629–30
Gradenwitz, Otto, 63
Graeco-Latin koinê, 135
Graeco-Roman Memoirs, 208
Graf, Theodor, 32, 60
graffiti, 162, 397, 407, 431, 478
grammarians, 72, 167n3, 271, 275, 277n3, 279n17, 321, 323, 325–26, 328–33
grammatical treatises, 288, 291, 295–99
grapheion (record office), 231, 364, 549, 556
Greek. See also interpretatio graeca
abbreviations, 170–74, 175, 176

Page 25 of 64
Index

administrative documents, 524–25

agoranomic documents, 556
in Apa Johannes letter, 438
in Arabic Egypt, 452–53
archives, 230, 240–41, 242
astronomy, 344–45, 350
and Bactrian, 473
(p. 672) bilingualism, 327, 423–24

bookrolls, 21, 256

as business language in Egypt, 590
and Coptic, 430, 432–45, 435, 443, 631
Corpus Papyrorum Graecarum, 212
“Curse of Artemisia,” 405
decline of, 441
documentary language in Egypt, 162
documentary papyri, 149–50, 154, 157–58
double documents, 482
dream books, 572
and Egyptian religion, 568
Gospel of Thomas, 599
Graeco-Latin koinê, 135
graffiti, 407
grammatical treatises, 295–99
grapheion registers, 556
Greek-Arabic documents, 461–62
handwriting development, 102–5, 107–14
horoscopic astrology, 347, 581
inscriptions, 581–82
Ionian notation, 340
koinê 135–36, 149–50, 298, 327, 430, 482–83, 591
kômê, 184
and Latin, 158, 428
legal documents, 376, 422, 541–43, 550, 556–57
literary papyri, 284–89, 428, 645
as liturgical language, 433
loanwords, 163
magical texts, 652
and Manichaeism, 639
mathematics, 339, 352
medical texts, 352–56
monastic language, 433–34
morphology and syntax, 154–57
names, 192
Nessana dossier, 481
notaries, 545–46, 566
novels, 288
as official language, 399–400, 419, 422–23
and Old Coptic, 431

Page 26 of 64
Index

oracle questions, 584

patristic literature, 600
petitions, 542
phonology, 150–53
as prestige language, 396
private letters, 361
as sacred language, 590–91
scholar's texts, 320–21
school exercises, 325, 334
technical vocabulary, 194
transliterations, 152–53, 167n5, 427
vocabulary, 157–58
weight system, 188–89
wills, 367, 422, 554
Grenfell, Bernard P.
and crocodile mummies, 46–47, 222
Fayyum finds, 34
historiography, 498–99
logia Iesu, 629–30
Oxyrhynchus excavation, 52, 53, 54, 61–63
papyrus editing, 211
on principles of afsh, 36
scientific archaeology, 36–38
translations, 207–8
Grohmann, Adolf, 453–54
Gronewald, M., 599
guardianship, 366, 484, 555
Guéeraud, Octave, 72
gum arabic, 18, 82–83
gunpowder, 32
Gurob, 33–34, 345
gymnasium, 380, 402, 504, 506, 511n35, 513n66
ḥadīth collections, 456–57
Hadrian, 56n13, 193, 232, 364, 373, 385, 523, 532, 574
Haensch, R., 489n44
Hagedorn, U. and D., 94
hagiography, 498
hamma, 186
handwriting development
archives, 227
Coptic, 444
declarations, 379
Dioscoros of Aphrodito, 241–42
Graeco-Latin koinê, 135
Near Eastern papyri, 481, 485
normative script, 127
official correspondence, 376
school exercises, 321–22, 327, 330–31, 335n7

Page 27 of 64
Index

severe style, 131–32

Handy Tables, 349, 351
Harpochrates, 565, 567, 577–79, 583
Harrauer, H., 88
Harris Homer, 114
Hawara, 44, 220, 409, 600
Hawara Homer, 128
Hayter, John, 306–7, 315n11
Hebrew, 22, 50, 406, 474, 479, 483, 489n50, 490n72, 590
Hebrews, Letter to, 591, 596, 597, 603, 604, 613n29, 613n38
Heidelberger Gesamtverzeichnis der Papyrusurkunden Ägyptens (HGV), 3, 145n1, 199, 446n5,
522n1, 657
Hellenes, 524
Hellenica Oxyrhynchia, 288
Hellenization, 542, 543–46, 548, 578–79, 631
hellenocentricity, 498
Hellenomemphites, 404–5
Hendriks, I. H. M., 9–10, 26n10
henotheism, 584
Herakleopolite nome, 33–34, 316n28, 529
Herakles, 567, 577
Herculaneum papyri. See also Philodemus
bibliography, 316n28, 317
bookrolls, 264
calligraphic writing, 112, 114
(p. 673) carbonized papyrus, 94–95

Centre Internazionale per lo Studio dei Papiri Ercolanesi, 71

editing of, 308–10, 315n15
Epicurus, 310–13
excavations, 312
handwriting development, 102, 138, 139
Latin papyri, 311–12, 315n9
library, 303–6
Obbink-Delattre method, 315n12
papyri, 59, 306–9, 315n15
polyptych, 23
restoration techniques, 92
Stoics, 311
heresy and heretics, 623–24, 626, 628, 630–31
Hermes Perpheraios, 206
Hermetic texts, 628
Hermopolis
agoranomic documents, 548
Egyptian religion, 565
excavations, 31
gunpowder factories, 32
Idumaeans, 405
illicit excavations, 33, 35

Page 28 of 64
Index

medical texts, 355

tax rolls, 383
Hero, 339–40
Herodotus, 293–94, 398
Herondas, 61, 287
Heroninos archive, 48–49, 231, 235, 362
Herophilus, 355
Hesiod, 268, 269, 286, 329, 335n9, 488n27, 504
hexameters, 200, 259, 286, 333, 335n12, 581
Hibeh parapêgma, 344
hieratic
decline of, 430–31
and demotic, 399
editing of, 657
literary papyri, 283–84
manufacture of, 6
medical texts, 352–54
and Old Coptic, 432
at Tebtynis, 236
hieroglyphs, 55n3, 60, 402, 405, 430, 566–67, 573–74, 581
high stops, 295
Hijra year, 183
Hipparchus, 349
Hippocrates, 160, 355
historical notes (historiai), 330
historical papyri, 372–73
historiography, 351, 458, 495–509, 6, 512n40
Hobson, Deborah, 496, 498
Hogarth, D. G., 34, 38
Holy Trinity, 624
Hombert, Marcel, 67–69
Homer
bookrolls, 21, 277n1
in Dioscoros archive, 242
Harris Homer, 114
plus verses, 288
rhetoric, 332
school exercises, 268, 325, 327, 329–31
subliterary papyri, 285
Homilies (Manichaean), 638
homophony, 151, 154, 156
Hopkins, Simon, 455, 459
Horace, 317n35
Horden, P., 512n50
horizontals, 174
horoscopic astrology, 346–48, 351, 372, 581
Horus, 565, 572, 574, 583, 585
house-to-house-approach, 48

Page 29 of 64
Index

humidor, 82
Hunt, Arthur S.
Christian papyri, 595
crocodile mummies, 46–47, 222
Fayyum finds, 34, 55
hierarchy of contexts, 48
historiography, 498–99
logia Iesu, 629–30
Oxyrhynchus excavations, 52, 61–63
papyrus editing, 211
principles of afsh, 36
and scientific archaeology, 36–38
translations, 207–8
Husselman, Elinor M., 73
hydrochloric acid, 87
hydrogen sulfide, 304, 307
hydrolysis, 79
hymns, 405, 581, 611n8, 614n69
of Callimachus, 287
Hyperides papyrus, 61, 114, 118
hyphen, 262
hypographê, 549, 556
hypomnema, 114, 200, 275, 277, 364, 379
hypomnematographos, 526
hypotheseis, 286, 288–89
ibis, 569–70
idios logos, 386, 526, 575
Idumaeans, 405
IFAO (Institut Français dʼArchéologie Orientale), 71–72
Iliad, 44, 264
illegitimacy, 380
illicit excavations. See excavations
illustration, 209–10
inclined ogival majuscule, 132–33, 136
incoming documents, 237–38
index, 624, 650–51
indiction, 183, 534–35
infanticide, 403
inheritance model, 274
inheritance tax, 553–54
ink, 18, 82, 84, 300n16
(p. 674) inkwells, 52–53

inscriptions, 581–82, 652, 655–56

Institut Français dʼ Archéologie Orientale (IFAO), 71–72
Institut für Altertumskunde, Cologne, 72
Institute of Papyrology, Sorbonne, 72
intercalary months, 180, 183
intermarriage, 403

Page 30 of 64
Index

internal documents, 238

International Committee of Papyrology, 68
International Society for Arabic Papyrology (ISAP), 454
interpretatio graeca, 522–23, 577–78
Invernizzi, A., 487–88n23
invitations, 363
Ionian notation, 340
Ionians, 397–98, 404–5, 412
iota adscript, 262–63
Iranian, 407, 477
Iraq, 452, 463
Irenaeus of Lyons, 628
iron-gall ink, 18, 82
irrigation system, 32, 59, 218, 384–85, 407, 466, 523
Isidora. See Kronion and Isidora
Islam, 458, 460–61, 463–64
Islamization, 462–464
isnāds (chains of transmitters), 457
Isocrates, 14, 114, 161, 330, 595, 600
Istituto Papirologico “G. Vitelli”, Florence, 71
Iulius Isocrates, strategos, 213
ius trium liberorum, 555
Japanese tissue paper, 84, 85, 88, 96
Jensen, C., 310
Jericho, 475–76, 479, 488n29, 489n44
Jesus
in Apocryphon of John, 628–29
and Egyptian cults, 586
and Gnosticism, 625
Kephalaia of the Teacher, 638
logia Iesu, 598–99, 629–30
and Manichaeism, 640
sayings of Jesus, 62, 267 (see also logia Iesu)
Jews. See also Bar Kokhba; interpretatio graeca
Babatha archive, 476
and the codex, 592
in Elephantine, 397–98
Greek language, 483
immigrants to Egypt, 398, 409
Islamization, 464
Judaism, 267, 624, 628, 635–37
legal documents, 461
liturgical language, 406
marriages, 366
names, 193
Old Testament, 592
politeuma, 405, 654
revolt (115–117 CE), 232

Page 31 of 64
Index

Septuagint translated, 590

use of parchment, 473
John, Gospel of, 593, 595–96, 598
Johnson, J. de M., 34, 38, 44–45, 615n69
Jonah, 129, 131
Jones, A. H. M., 65
Jördens, Andrea, 219
Jouguet, Pierre, 63, 68, 71
Journal of Egyptian Archaeology, 70
Judaism. See Jews
Judea, 480, 483
juridical documents, 238, 240
juristic papyrology, 64, 145, 374
Justinian
cadastre of Aphrodito, 383
use of Greek language, 422–23
legal documents, 427
regnal years of, 183, 491n77
tax receipts, 480
taxes, 382
Kahun papyrus, 353
Kallistos, 608–9
Kampichler, J., 95
Karabacek, Joseph, 453
Karanis
The Archive of Aurelius Isidorus, 73
archives, 245–46, 423
bookrolls, 39
context, 511n30
excavations, 647
hierarchy of contexts, 48
house-to-house-approach, 226–27
inkwells, 52–53
sebbâkhîn, 35
tax rolls from, 383
University of Michigan excavations at, 38, 67
Karnak, 43–44
katochos, 414, 508, 515n88, 572, 575, 582, 584
katoikic land, 368, 386, 528
Keenan, James G., 508
Kellis, Tutu cult, 583
Kellis papyri
accounts, 370
bilingualism, 434, 437, 444
Coptic, 433
Dakhleh Oasis Project, 648
excavations, 640
Manichaeism, 601–2, 639

Page 32 of 64
Index

private letters, 641

Kelsey, Francis W., 38–39, 66
Kelsey Museum of Archaeology, Ann Arbor, 73
Kenyon, Frederic G., 61, 63, 288
Kephalaia of the Teacher, 638
keramion, 42, 188
Kerkeosiris, 46, 219, 222, 371, 383, 401, 564
Kerkesoucha Orous, 364
Khan, Geoffrey, 454
(p. 675) Khirbet Mird, 476, 488n29

Khoury, Raif Georges, 454, 456

Kilpatrick, G. D., 593
kimân, 34
Kimân Faris, 60
kinship terms, 160
klerouchos 528
Koenen, Ludwig, 94
koinê Greek. See Greek
kollêsis, 9, 19–21, 81, 265–66. See also tomos synkollêsimos
kolophônion, 42
Kôm Gamman, 513n56
Kôm Ishqaw, 49
kômai (villages), 184, 523
komarches, 526, 527, 531, 533
kotylê, 187, 188
Krokodilo, 53
Krokodilopolis 231, 523, 548, 557n2, 568
Kroll, Joseph, 72
Kronion and Isidora, 503, 505, 506, 513n58, 513n60, 514n77
Kronos, 507, 514n76, 514n77
Kulikowski, M., 512n41
La Graufesenque (Condatomagos, Gaul), 141
land registration, 544–45, 556, 558n7, 603–4
language, 395–96, 399, 407–9, 459–60, 482–83, 541–43. See also bilingualism; individual lan­
guages; multilingualism
laokritai, 546, 558n9, 567
Latin. See also hypotheseis
and Arabic papyri, 460
archives, 241
bilingualism, 423–24
birth certificates, 379
at Bu Njem, 655
chancery style, 143
Chartae Latinae Antiquiores, 212
conversation manuals, 428
Corpus Papyrorum Latinarum, 212
cursive writing, 137
diglossia, 421–30

Page 33 of 64
Index

documentary papyrus, 149–50, 645

in Egypt, 162–66
handwriting development, 137–45
in Herculaneum papyri, 311–12, 315n9
as language of power, 424, 428, 446n8
Latinization, 429–30
legal documents, 553, 557
literary papyri, 282, 289–90, 422, 423, 427–28
loanwords, 158
and Manichaeism, 632
medical texts, 352–56
morphology and syntax, 163
in Near Eastern papyri, 483
as official language, 422–3
oracle questions, 584
Oxyrhynchus letter, 164–66
papyrus decipherment, 201
regnal years, 183
rhetoric, 334
semiuncial, 144
transliterations, 427
uncial writing, 145
Vulgar Latin, 162–63, 166
wills, 542
law of succession, 481, 553–54, 557
law of the Egyptians, 543, 550–52, 555, 556–57
lawyers, 428
LDAB (Leuven Database of Ancient Books), 3, 4, 21, 25, 145n1, 266, 291–92, 299n11, 300n13
leaf-tablet, 13, 14
leap years, 181, 214n12
leases, 369, 549
lectors, 265
legal documents. See also contracts; notifications (prosangelmata); scribes
administrative documents, 376–77
agoranomic documents, 548–50, 556
anagraphê, 546–48
Arabic papyri, 460–61
arbitration, 387
banking transactions, 364
Bar Kokhba, 490n72
bilingualism, 377, 387, 437
Coptic, 437, 441, 442, 541
demotic writing, 399, 541–43, 549
double documents, 544
employment contracts, 369–70
family archive, 238
Gnomon of the Idios Logos, 374
Greek language, 376, 422, 541–43, 550, 556–57

Page 34 of 64
Index

Hebrew, 490n72
Hellenization, 542, 548
language, 541–43
Latin, 424, 425, 427
law of succession, 553–54
law of the Egyptians, 543
marriages, 238, 544
misthôesis, 369–70
Near Eastern papyri, 483–85
order to arrest, 377
private, 363–69
Ptolemaic era, 376, 542
parchment, 473
Tebtynis, 364
wills, 367, 542
Leiden system, 68, 202, 203, 211, 455, 650
lemma, 200, 325, 331
Leriche, P., 478
Le Roy Ladurie, Emmanuel, 507
Leuven Database of Ancient Books. See LDAB
Leuven Homepage of Papyrus Collections (LHPC), 230, 655
Levi Della Vida, G., 454
(p. 676) Lewis, Naphtali, 71–72, 231, 483

Lex Youtie, 207

Lexikon der lateinischen Lehnwörter, 429
Libanius of Antioch, 333
libelli (petitions), 606, 615 n81
library, 218, 305–6
Libya, 523–24, 655–56
ligatures, 120, 136, 137, 174
limestone, 14
linen, 3, 4, 45, 87
linguistic interference, 460
liquid measurement, 187–88
listening gods, 568, 570
literacy, 320–28, 359–60, 407, 444–45
literary papyri
Arabic, 456–57
in archives, 240
bilingualism, 444
book hands, 257
bookrolls, 267
calligraphic writing, 112, 113, 115, 116, 117
Catalogue of Paraliterary Papyri (CPP), 291
chancery style, 123
Christian, 429, 591–601
codices, 278n9
decipherment, 200, 203

Page 35 of 64
Index

definition, 282–83
Derveni papyrus, 92, 94
vs. documentary papyri, 359, 407
in Dryton archive, 499
editing of, 198, 211
Greek, 103, 149, 284–89, 428, 645
Latin, 138, 289–90, 422, 423, 427–28
Manichaeism, 639
Near Eastern papyri, 475
Oxyrhynchus, 34, 62–63, 268
prose, 288–89
Ptolemaic style, 105
punctuation, 262
restoration techniques, 204, 205, 300n16
scholar's texts, 272
school exercises, 321, 334
severe style, 132
study of, 65
stylization, 150
Tebtynis, 236
litterae caelestes, 140, 142
litterae communes, 142
liturgical service, 375, 382, 385, 390n80, 390n87
liturgies, 291, 405–6, 433, 460
livestock, 382
Livian papyrus, 6
loan contracts. See contracts
loanwords, 158, 460
Lobel, Edgar, 63, 293, 300n14, 511n27
logia Iesu, 598–99, 629–30
logistes, 534
Louvre, 60, 212
love poems, 284
Lowe, E. A., 145
Lucian, 269, 276–77
Lucretius, 311
Luijendijk, AnneMarie, 615n76
Luke, Gospel of, 591, 596, 598
Lumbroso, Giacomo, 63
Lykopolis, 436, 608, 633, 637–38
Lydian, 399
lyric texts, 277n2
Maas's law, 258, 261
Macedonia, 158, 180, 182–83, 473, 590
magical texts, 583–84, 595–96, 602, 642, 652
magnetometry, 41
Mahaffy, John P., 61, 63, 522
Mai, Angelo Cardinal, 60

Page 36 of 64
Index

majuscule, 128–29, 139, 141–42, 145, 481. See also biblical majuscule
Makarios, 640
Manchester, University of, 454
Manichaeism
asceticism, 636–37
catechumens, 638–40
and Christianity, 636–37, 642
Church History (lost), 638
Cologne Mani Codex, 632, 633, 634
Coptic, 433
Easter, 602, 615n71
in Egypt, 631–37
Epistles of Mani, 638
Homilies, 638
and Judaism, 636
Kephalaia of the Teacher, 638
Latin, 429
Mani, founder, 625, 635–37, 640
Medinet Madi codices, 351, 581, 637–42
Panarion, 624–26
private letters, 601–2, 639–40, 641
Psalm Book, 638
and Serapion of Thmouis, 615n69
Synaxeis of the Living Gospel, 638
Zoroastrianism, 636
Manning, J. G., 501–2
manure, 32, 35
marble, 486n2
margins
bookrolls, 259, 263, 277n5
codices, 267
and papyrus decipherment, 199
and restoration techniques, 206, 214n10
in school exercises, 329
Margoliouth, D. S., 454
Mariette, Auguste, 31, 44
Mark, Gospel of, 598
marriage
agoranomic documents, 558n14
Babatha archive, 479
(p. 677) betrothal agreement, 439–41

contracts, 366
divorce contracts, 554–55
endogamy, 504, 506, 513 n58
intermarriage, 403
law of the Egyptians, 550–52
legal documents, 238, 544
mixed marriage, 410–12

Page 37 of 64
Index

Near Eastern papyri, 489n4

Marrou, H.-I., 320
Martin, Victor, 68
Martinez, David G., 49
martyrdom, 606
Masada, 476
Maspero, Gaston, 33–36, 43–44, 70
mathematics, 339–43, 352–53, 566
mation, 186
Matthew, Gospel of, 591, 598
Maximianon, 43
maxims, 325, 327
Mazzochi, Alessio Simmacho, 306–7
measures, 175, 185–89, 421
medical texts, 291, 352–56, 378–79, 457, 506
medicine, 566, 572, 585
Medinet Madi codices, 351, 581, 637–42
Melitian monks. See monasteries
Memphis
after Greek conquest, 400
Egyptian religion, 565
ethnic identity, 404
Greek inscriptions, 582
imperial cults, 575
multilingualism, 398–99, 414
shipping, 412
Menander, 49, 105, 114, 133, 242, 268, 269, 287, 309, 327, 329, 443
Mendesian archive, 47
mephitic gas, 304
mercury, 307
meridarches, 526, 527, 529
merides, 184, 528, 529, 532
Merkelbach, Reinhold, 72, 206
Mertens-Pack, 285, 292, 300n12
Mesopotamia, 120, 342, 352, 478, 635–36
metallic inks, 18, 26n12
methyl cellulose, 83–84
metropolis (capital), 184, 228, 376, 378, 380, 522–23, 548, 605
metropolite-gymnasial class, 506
microfiche, 210
Middle Euphrates, 475–78
Middle Kingdom, 283, 570–71
Middle Persian. See Pehlevi
midollo, 307–8, 315n14
military. See also soldiers
administration of Egypt, 526
administrative documents, 385–86
Arabic papyri, 462

Page 38 of 64
Index

bilingualism, 423
Bu Njem ostraca, 655
Dura-Europos archive, 477
Latin papyri, 162
legal documents, 377
Pehlevi documents, 419–20
petitions, 378
taxes, 382
Millar, Fergus, 478
Mimes of Herondas, 44
Min, 578, 580
minuscule, 140, 141–42, 145, 481
minutes, 374–75
misspellings, 151–52, 165
misthôsis, 369–70
Mitteis, Ludwig, 63–64, 644
mixed marriage, 409–12
mixed papyri, 299n7
mna, 189–90
mnemonic devices, 327, 330
modius Italicus, 187
Mommsen, Theodor, 63, 644
monasteries and monasticism. See also Pachomius
archives, 219, 236–37
bilingualism, 444
child-donation contracts, 508
in Egypt, 433–34
Manichaeism, 632, 637, 642
Melitian monks, 220, 435, 608–10, 616n97
Nag Hammadi Codices, 627
Near Eastern papyri, 476
origin of, 608
patristic literature, 599–600
The Shepherd of Hermas, 600
monograms, 172–73, 175, 177
monopoly, 530
monotheism, 584
Mons Claudianus
diet and provisions, 42
excavations, 40–41, 648
garbage dumps, 246
ostraca, 15, 16, 43, 648
quarries, 56n13
school exercises, 322
Montevecchi, Orsolina, 71
months, 180, 181, 182–83, 344–45, 413, 482
Montserrat codex, 428
Morgan, Teresa, 320, 335n7

Page 39 of 64
Index

morphology, 154–57, 163, 165

Moscow papyri, 341
Mueller, Katja, 502–3
multilingualism
agricultural shipping, 412
Arabic papyri, 454
Arabization, 459
contracts, 409–10
in Egypt, 418–19, 445
(p. 678) absence in Pehlevi documents, 419–21

Persian empire, 395–99, 409

Ptolemaic era, 399–406
multiplication, 340
multispectral imaging (MSI), 98, 313
mummies. See also cartonnage
and Christianity, 586
crocodile mummies, 45–47
Egyptian religion, 565
labels, 4, 12, 371–72
sacred animals, 570
transport of, 371–72
municipalization, 532–33
muriarouros, 528–29
Museo Nazionale Archeologico, Naples, 60
museum archaeology, 228–29, 246, 654
Muslims, 432
Myos Hormos, 42, 322
mythographical texts, 331–32
Nabataeans, 406, 479
Nag Hammadi Codices
cartonnage, 437
discovery of, 626
Gospel of Thomas, 598–99, 629–30
illustrations of, 627
language, 428–29
Sophia Jesu Christi, 630
Nahal Hever
Babatha archive, 50, 51–52
diplomatics, 482
Engadi caves, 489n47
excavations, 56–57n21
Near Eastern papyri, 478–80
Salome Komaïse, 476
Semitic languages, 483
writing development and, 120, 124
Nahman, Maurice, 66–67
Namenbuch, 65, 198
names

Page 40 of 64
Index

abbreviations, 172–73
Egyptian gods, 578–79
in Egyptian papyri, 191–93
ethnic identity, 402–3, 433
and ethnic origin, 398–99
in the Fayyum, 522–23
Islamization, 463
in Nessana dossier, 481, 483
Oxyrhynchus letter, 165–66
Petra papyri, 483
Roman citizenship, 160
Tebtynis, 514n77
Nanaion, 549
Naples, 303, 314n5, 315n15
Narmouthis, 581–82, 638
narrative tradition, 508
natron salts, 79–80
naubion, 186
Naukratis, 402, 404
Naville, Edouard, 33, 47, 231
necrologies, 74
necropolis, 103, 219, 404, 571
Neith, 577, 578
Nelson, C. A., 310, 316n25
Nessana dossier, 459, 475–76, 480–81, 486n2
Neugebauer, O., 346–47, 350
New Kingdom, 284, 568–72
New Testament
Chester Beatty papyri, 597
codices, 266, 595–96
Coptic, 631
dating, 596, 613n31
Greek vocabulary, 157
koinê Greek, 149
parchment rolls, 22
patristic literature, 599–600
Nicaea, 608
Nicole, Jules, 61, 63
Nile, 412, 523
Nimrud, 27n42
Nobbs, A., 601–2
nomarches, 232, 525, 526, 528–31
nomes
basilikos grammateus, 527
Byzantine period, 534
definition, 184
geography of Egypt, 522–24
Pathyrite nome, 545–46, 548, 556

Page 41 of 64
Index

Roman era, 532–33

strategos, 528
nomina sacra, 200, 592, 631
nominal phrases, 192
Nongbri, B., 596
normative script, 138–39
Norsa, Medea, 70–71
notaries (agoranomoi)
bilingualism, 441, 444
contracts, 364, 542
Egyptian priesthood, 573
land registration, 545–46
Near Eastern papyri, 484, 485
property ownership, 383–84
Roman era, 549–50
temples, 566
transliterations, 427
notifications (prosangelmata), 378
Notitia Dignitatum, 534
novels, 268–69, 288
numerical literacy, 327–28, 340
Obbink, Dirk, 498
Obbink-Delattre method, 315n12
obliques, 174–75
obols, 190–91
Octavian, 181–82, 396
(p. 679) octavo, 24

official archives, 231–33

official doctor, 378–79
Officina dei Papiri, 310, 314n5
oikonomos, 526, 527, 529, 531
Old Babylonian period, 352
Old Coptic, 430–31, 432
Old Testament, 22, 590, 592, 595, 624
omens, 352
one-sixth tax (apomoira), 526
onomastics, 192, 463, 478, 480, 583, 586
onomatopoeic, 151
open text, 299
Oracle of the Lamb, 284
oracles, 372, 563, 571, 579–80, 584, 586
oral tradition, 458, 461, 636
oratory, 288
order to arrest, 377, 604–6, 615n76
Oriens, 485–86
Origen, 595, 599
Orrieux, Cl., 231
orthography, 163, 650

Page 42 of 64
Index

Osiris, 565
Oslo, 66
ostraca
abbreviations, 170
accounts, 371
administrative documents, 375
Amenothes cult, 571
amphora, 15, 17
as writing material, 3, 14–17
Arabic, 453
archaeobotanists, 41–42
archives, 245–46
from Bu Njem, 655
dossiers, 219, 246
dreams in, 572
Dura-Europos archive, 477
frequency of use, 4
garbage dumps, 53, 246
geometrical theorems, 343
horoscopic astrology, 347, 351
from Krokodilô, 53
language evolution, 149
from Maximianon, 43
medical texts, 354
military history, 385–86
from Mons Claudianus, 15, 16, 43, 648
Near Eastern, 473, 486n2
and Old Coptic, 431
published, 645
school exercises, 322
shaped, 16
survey of land, 525
tax receipts, 15–16, 63, 383
from Tebtynis, 40
temple archives, 237
temple oaths, 567
outgoing documents, 237–38
“Oxford Dioscuri”, 34
oxidation, 79, 81
Oxyrhynchus
abbreviations, 175
Alcaeus poetry, 285
astronomy papyri, 350
birth certificates, 379
bookrolls, 264
death notices, 381
Egyptian priesthood, 586
garbage dumps, 50, 52

Page 43 of 64
Index

Gospel of Thomas, 598–99

Greek letter, 158–62
Grenfell and Hunt excavations, 34, 54
Latin letter, 164–66
literary papyri, 268
logia Iesu, 598–99, 629–30
Manichaeism, 639
minutes, 375
oracle questions, 372
The Oxyrhynchus Papyri, 211, 499, 511n27, 650
papyrus fragments, 37
prosopography, 194
regnal years, 183
ruling dots, 261
sacred animals, 569
Sarapion, 513n56
scholar's texts, 270–77
translations, 207–8
Pachomius, 428, 600, 627–28, 637
Pachymios, 439
Pack, R. A., 199, 292
Paderni, Camillo, 305–6, 315n9, 315n12
pagan texts, 266, 631
pagarchy, 535
pagi, 184, 533–35
paginae, 21
palaeography
Arabic papyri, 456, 459
digitization of papyri, 649
grammatical treatises, 295
Kellis papyri, 602
Near Eastern papyri, 481
scholar's texts, 273
Palestine, 120, 123, 133, 234, 480–81, 490n55
Palmyra, 475, 487n10
Panarion (Epiphanius of Salamis), 624–26, 632
Panopolis, 612n26, 630
Papaccio, Valerio, 314n3
paper, 3, 4, 453
papier-mâché, 87
Papiri della Società Italiana, 211
Papyri graeci regii Musei Taurinensis, 60
papyrological millennium, 368–69
papyrologists, 61, 68, 73–74
(p. 680) papyrology

Analecta Papyrologica, 75
Annales school, 507
Arabic, 452–55

Page 44 of 64
Index

Archiv für Papyrusforschung, 63

archives, 217–18
Bingen definition, 72–73
Byzantine papyrology, 70, 132, 135–36, 439
development of, 59–61
excavations, 648
Greek authors, 288
hellenocentricity, 498
and historiography, 498–99
integration, 500–507
international organization, 67–75
Leiden convention, 68
metapapyrology, 73
microhistory, 507–8
Near Eastern, 484–86
necrologies, 74
papyrological millennium, 61
papyrus cartels, 66–67
publication backlog, 645
and religion, 626
summer institutes, 645
unpublished material, 658
papyrus See also carbonized papyrus; Derveni papyrus; documentary papyrus
in Babylonia, 473
bookroll manufacture, 19, 256–57
carbonized, 83, 92
Christian documents, 591, 604–5
codices, 265–66
conservation of, 80–81, 648
Cornelius Gallus papyrus, 139, 290
cultivation of, 4–5, 452–53
damage to, 79–80
frequency of use, 4
manufacture of, 5–8, 9–10, 26n10
mills, 467n2
mummy cartonnage, 87–88
oracle questions, 563, 571
origin of, 5
restoration techniques, 81–86
schoolbooks, 326
papyrus texts. See also Arabic papyri; Dura-Europos papyri; literary papyri; Petra papyri; sublit­
erary papyri
archives, 220, 221, 222
as artifacts, 645–46, 657–58
Berliner Klassikertexte (BKT), 211
carbonized papyrus, 92–98
Checklist of Editions, 211, 299n7, 299n8
commentary, 208–9

Page 45 of 64
Index

conservation of, 81–86

decipherment, 199–203
digitization, 649
diplomatic handbook, 659
editing of, 197–99, 210–11, 644–45
as facts, 653–59
handwritings, 101–2
historical papyri, 372–73
hot papyri, 648
illustrations of, 209–10
last of, 453, 467n6
Near Eastern papyri, 474–77, 480–81, 483–86, 486–87n6, 489n44, 491n77
papyrological habit, 360
reediting of, 651–53
restoration techniques, 203–4, 206–7
sciences, 338–39
as texts, 649–53
transcription, 202
translation of, 207–8
Paraclete Spirit, 640
paragraphos, 261
paraliterary texts, 140
parallels, 204, 206
Paraloid B-48, 88–90
parapêgma, 344
parchments (pergamena)
as ancient writing material, 3
bookrolls, 22
codices, 266, 278n9
manufacture of, 11
medical texts, 354
Near Eastern, 473
and papyrus codex, 27n44, 265
school exercises, 322
temple archives, 237
title tags, 264
Parsons, P. J., 293, 294, 615n76
Parthenon, 56n13
Parthians, 477
Pathermouthis archive, 439
Pathyrite nome, 545–46, 548, 556
patristic literature, 599–601
Pauline epistles, 428, 596, 598
Pehlevi, 419–21, 420
Peloponnesian War, 288
Pentecost speech, 612n14
peregrines, 543, 553
Pergamenes, 11

Page 46 of 64
Index

persecution, 605–8, 623

Persia, 293–94, 395–99, 460, 624, 632, 636
The Persians, 405
Pestman, P. W., 70, 214n7, 228, 229–30, 238, 558n7
petitions
Abinnaeus, Flavius, 423
administrative documents, 377–79
archives, 237, 360
Christian persecution, 606
koinê Greek, 591
Near Eastern papyri, 485
(p. 681) Ptolemaic era, 542

Zenon archive, 406

Petra papyri
carbonized papyrus, 92, 95, 480
Near East papyri, 476
papyrus restoration, 93
regnal years, 486
restoration techniques, 94, 95
Petrie, William Matthew Flinders, 33, 36–38, 44, 47, 61, 220
Peyron, Amedeo Angelo Maria, 60
Philadelphia, 383
Philae, 431
Philip the Arabian, 533
Philo of Alexandria, 590
Philodemus
bookrolls, 264
handwriting development, 102
Herculaneum papyri, 306, 309, 311–13
and Piso Caesoninus, 317n34
writing style, 114
philologists, 36, 62, 650
Phoenicians, 397, 412
Phoenico-Egyptians, 405
phonology, 150–53, 163
photographic documentation, 97–98
physicians, 354–55, 378–79
Piaggio, Antonio, 92, 307–9
Pindar, 114, 269, 273, 285, 309, 329, 582
Pira, Gaetano la, 307
Piso Caesoninus, Lucius Calpurnius, 311–13, 317n34, 317n35
Pistis Sophia, 631
Pithom, 33
plague, 373, 501
Planatol BB, 88–90, 97
plates, 209–10
Platonic fragments, 160, 627–28
Plautus, 286–87

Page 47 of 64
Index

Plexiglas, 86, 97
Pliny the Elder, 4–8, 11, 467n2
plus verses, 288
Plutarch, 293–94
poetry
Arabic papyri, 457
bilingualism, 434
Greek, 285–87
love poems, 284
rhetoric, 321, 333–34
school exercises, 329–30
Simonides, 292–95
politeuma, 405–6, 654
poll tax. See taxes
Polotsky, Hans Jakob, 638
Polybius, 149
polyptych, 19, 23
polytheism, 582–83, 585
polyvinyl acetate (PVA), 97
Pompeii, 140, 162, 165, 303
porcellino, 313
Portici, 306, 310
Posidippus papyrus, 106, 107, 287
postal service, 375
postscript, 270
potsherds, 14
pottery, 42
power of attorney, 607–8
Prayer of the Emanations, 639, 642
prayers, 284, 457
Prayers of Serapion, 602, 615n69
Preaux, Claire, 68
Preisendanz, K., 75
Preisigke, Friedrich, 65, 198–99, 211–12
prepositions, 157
priesthood
administration of Egypt, 525
circumcision, 380–81
imperial cults, 575
mortuary priests, 565
notaries, 573
Roman era, 504, 506, 533, 574–75
taxation of, 576
theagoi, 565, 566
Princeton University, 66
private archives, 48–49
private documents, 361–72. See also private letters
private letters

Page 48 of 64
Index

Arabic papyri, 455–56

as archives, 216
Christian, 595, 601, 603, 614n62, 616n88, 616n89
Coptic, 437
destruction of, 16
Gnostics, 601
language evolution, 150
Latin papyri, 162, 423
lost documentation, 360–61
Manichaeism, 601–2, 639–40, 641
Pehlevi documents, 419
restoration techniques, 206
problem texts, 341–42
progymnasmata, 325, 332–33
pronunciation, 151–52, 163, 167n5
property ownership
Christian documents, 605
cleruchic settlement, 401
contracts, 367–70
Diocletianic reforms, 386
land registration, 399
Ptolemaic era, 383–84
taxes, 383–84
temples, 565–66, 573–74
Prophecies of Neferti, 284
prose, 288–89, 321, 333–34
prosopography, 193–94, 213, 227, 246
prostagmata, 374, 547–48
proto-Gospel, 267
(p. 682) protokollon, 19, 263

Psalms, Psalter, 443, 592–93, 594, 595, 638–39, 642

psogos, 332
Ptolemaic era
abbreviations, 171–73, 175
administration of Egypt, 526
administrative documents, 374
astronomy, 581
bronze coinage, 190
bureaucracy, 400
census, 381, 574
Chremonidean War, 525
chronology, 181
conveyance tax, 558n7
demotic writing, 241
Egyptian language, 158, 406
geography, 184–85
geography of Egypt, 522–24
Greek, 105, 106, 107, 152, 155

Page 49 of 64
Index

imperial cults, 575, 576

land registration, 545–46
leases, 369
legal code, 555
legal documents, 376, 542
liquid measurement, 188
mixed marriage, 412
multilingualism, 399–406
petitions, 377–78
private documents, 361
property ownership, 367–68, 383–84
Ptolemais founded, 523
Second Syrian War, 525
Septuagint translated, 590
taxes, 382, 530–32
temple administration, 573–74
trade, 648
Ptolemaios (katochos), 240, 414, 572, 582, 584
Ptolemy (mathematician), 349–51
Ptolemy X, 376
public documents, 372–87
punctuation
bookrolls, 261–63
codices, 266
dicolon, 261–62
dots, 261, 277n3
Leiden system, 214n7
literary papyri, 262
papyrus decipherment, 201, 203
scholar's texts, 273
sigla, 261
Punic, 645
Purcell, N., 512n50
pyroclastic flow, 306
pyrolized, 83–84, 92
Pythagoras, 342
Qamadir, 44–45
Qasr Ibrim, 290
Q source, 629
quarto format, 24
Quintilian, 11–12
quires, 24, 265, 278n8
Qumran, 476, 478–79
Qurʼāan, 452, 456, 467n10
Qurrah ben Sharik, 136
radon, 304
Ragab, Hassan, 5, 8, 10, 25n8
Raghib, Yusuf, 454

Page 50 of 64
Index

Rainer, Archduke, 32, 60, 453

Ramesseum papyri, 353
rank-size rule, 503, 512n50
rapid hand, 324–25, 333
Rathbone, D. W., 231, 235, 513n66
Ray, John D., 399, 506
real estate, 228, 364, 368
receiving measure, 186–87
recommendation letters, 424, 426
recto/verso, 102, 322, 592–93, 594, 595, 604
reed brush, 7, 26n12
reed pen (calamus), 7, 18, 19, 140, 408
reformed documentary hand, 266
regnal years, 182–83, 482, 485–86, 491n77
reincarnation, 637
reintegration edicts, 381
religion. See also individual beliefs
accounts, 371–72
Apis bull, 414
Arabic papyri, 457
circumcision, 380–81
Derveni papyrus, 288
epikrisis registers, 380–81
Hibeh parapêgma, 344
invitations, 363
leases, 369
liturgies, 385
multilingualism, 412, 414
Muslims, 432
Papyrus Amherst, 397–98
Resina, 303
restoration techniques
cartonnage, 89, 90, 91
literary papyri, 205, 300n16
margins, 207, 214n10
papyrus editing, 203–4, 206–7
papyrus conservation, 81–86
Revel, Jacques, 497
Revelation, Book of, 600–601
Revenue Laws, 524–25, 530–32
Reynolds, L. D., 612n26
rhetoric, 321, 325, 328–29, 332–34, 335n7
Rhind papyrus, 341–42, 353
Richter, Tonio Sebastian, 508
Roberts, C. H.
abbreviations, 593, 611–12n11
codex, 267
(p. 683) Gnostics, 613n44

Page 51 of 64
Index

New Testament, 596

papyrus manufacture, 8
The Shepherd of Hermas, 600
Rome
abbreviations, 172, 174, 177
administration of Egypt, 532–33
administrative documents, 374
agoranomic documents, 549
and Arabia, 479
army camps, 647–48
astrology, 346–52, 580–81
astronomy, 345–46
Bu Njem ostraca, 655
census declaration, 381
chronology, 181, 182
circumcision, 381
citizenship, 160, 379–80
conquest of Egypt, 402, 504
demotic writing, 241
Dura-Europos, 477
Egyptian priesthood, 504, 506, 574–75
epikrisis registers, 380
Fayyum, 523
Graeco-Latin koinê, 135
handwriting development, 107–14, 133
imperial writing style, 115
inscriptions, 581–82
katoikic land, 368
law of succession, 553–54
law of the Egyptians, 550–52, 555
legal documents, 376–77
liquid measurement, 188
magical texts, 584
modius Italicus, 187
Near Eastern papyri, 473–74, 483–84
oracle questions, 579–80
petitions, 378
regnal years, 182–83
right of three children, 366
school exercises, 330, 333
Sonderstellung problem, 64–65
stipulation clause, 554
Syriac language, 483
temple administration, 567, 573–74
tomos synkollêsimos, 243, 244
wills, 367
Rosetta Stone, 187, 575
Rosini, Carlo Maria, 309

Page 52 of 64
Index

Rostovtzeff, Michael I., 67–68, 74

rotulus, 22
Royal Correspondence in the Hellenistic Period, 74
rubbish dumps. See garbage dumps
Rubensohn, Otto, 33–34
Ruffini, G. R., 502
ruling dots, 261
rural history, 465–66
rush pen, 408
Rylands, fragment of Gospel of John, 593
Sabinus, 18
sacred animals, 565, 566
sacrifices, 373, 584, 586, 606–7
saints, 570–71
Saitic papyrus, 6, 8
sales tax, 384
Salome Komaïse, 476, 479
Salt, Henry, 60
salt tax. See taxes
saltpeter, 32
salvation, 624, 637–38
Samaritan revolt, 476
Sammelbuch, 65, 198, 212, 657
Sappho, 62, 285
Saqqara, 44, 103
Sarapion alias Apollonianos, 513n56
Sarapis, 564, 572, 578–80, 583
Sassanids, 419, 477
Sayce, A. H., 44
Sayings of Jesus. See Jesus, logia Jesu
Scherer, Jean, 72
Schmidt, Carl, 63, 638
schoinion, 185
scholar's texts, 270–77, 279n17, 320–21
scholia, 287, 299, 299n4, 325–26, 331–32
school exercises
archives, 240
bilingualism, 327, 334, 443
Christian papyri, 328, 595, 612n26
evolving hand, 324, 330–31
grammarians, 321, 325–26, 328–32
handwriting development, 321–22, 327, 330–31, 335n7
Homer, 268, 325, 327, 329–31
Latin, 334, 428
numerical literacy, 327–28
Old Coptic, 432
papyri, 320–25
rapid hand, 324–25

Page 53 of 64
Index

rhetoric, 321, 325, 332–34, 335n7

scholia minora, 315, 326, 331–32
school hand, 322–24
school texts, 321–22, 442–44, 488n29, 600
schoolbooks, 326
teacher's hand, 322–24
wooden tablets, 322–23, 326
Schow, Niels Iversen, 59–60
Schubart, Wilhelm, 131–32
Schütz, C. G., 310
sciences, 338–39
scorzatura, 307
Scotch tape, 85, 97
scribes
agoranomos, 548–50, 556
annotations, 273
(p. 684) Arabic papyri, 457

astronomy, 581
basilikos grammateus, 526, 527
bilingualism, 408
Christian, 592–93
codices, 266, 631, 633, 635
court proceedings, 547–48
dating papyri, 209
declarations, 379
diorthôtês model, 274
Herculaneum papyri, 313
Komogrammateis, 527
language of, 407
Latin, 138
legal documents, 363, 542
Maas's law, 258, 261
mathematics, 339–42
notaries, 546
oracle questions, 584
papyri, decipherment of, 200, 202
payment for, 257
Ptolemaic era, 555
regnal years, 183
restoration techniques, 206–7
scholar's texts, 270, 273
school exercises, 324
scriptura interior/exterior, 482, 486–87n6
stipulation clause, 554
topogrammateis, 527
scriptio continua, 261–62, 323
scriptio plena, 262
sculpture, 500, 511n35

Page 54 of 64
Index

sebbâkhîn, 32–34, 35, 224, 226, 236

Second Syrian War, 525
Seidl, Erwin, 231, 233
Seleucids, 473, 475, 477, 487–88n23
semicursive writing
Byzantine period, 135
development of, 101–2, 107, 109, 110
Greek handwriting development, 108
imperial writing style, 120
Ptolemaic style, 106
Semitic languages, 193, 397, 407, 474, 483
semiuncial, 142, 144
Semtheus, 567, 577
Seneca, 290
sentimental correspondence, 240
Septimius Severus
administrative documents, 374–75
contracts, 364
municipalization of metropolis, 184, 533
oracle questions, 571
petitions, 360–61, 378
Septuagint, 590, 593, 594, 604
Serapeum of Memphis, 31
Serapion of Thmuis, 615n69
Seven Nomes, 523
severe style, 131–32, 133, 134
sextarius, 188
Seyrig, H., 475
Shapur I, 636
Shepherd of Hermas, 600
shipping, 404, 412
shorthand, 170, 377
shrines, 564, 568
Sicily, 452
Sickler, Friedrich, 314n5, 315n17
sigla, 261, 273, 275
sigma, 300n15
sign-entry almanacs, 348–49
Silk Road, 632
silver, 191
Simon Choseba, 479
Simonides, 285, 292–95, 300n14
Sims-Williams, N., 486–87n6
single quire, 24, 592
sinusoids, 174
sitologos, 382–83, 389n65
Skeat, T. C., 267, 599, 613n31
skin (writing material)

Page 55 of 64
Index

Afghanistan, 486–87n6
diplomatics, 482
Dura-Europos archive, 477
frequency of use, 3, 4
Ionian writing material, 398
Near Eastern documents, 473–74
slavery
Arab commerce, 465
contracts, 365, 368
emancipations, 365
Greek ownership, 403
medical practice, 572
mixed marriage, 410
sacrifices, 607
taxes, 382
Zenon archive, 234
slurs, in writing, 177
Smith, H. S., 353–54
social network theory, 502, 506
Società Italiana per la Ricerca dei Papiri Greci e Latini in Egitto, 71
Soknebtynis, 503–7, 514n76, 514n77, 578, 580
Soknobraisis archive, 236, 372
Soknopaiou Nesos, 33, 574–75, 579
soldiers
archives, 236
billeting, 410–11
cleruchic settlement, 401, 412, 528
family archive, 241, 246–47
katoikic land, 368
military correspondence, 237
petitions, 386
private letters, 362
Vindolanda writing tablets, 385
solidus, 189, 191
(p. 685) solstices, 346

Sonderstellung problem, 64–65, 500, 512n40

Sophia, 629
Sophia Jesu Christi, 630
sophists, 334
Sorbonne, 72
Soter tomb, 511n35
Sothic cycle, 180
Souchos, 565
speech, 399, 407–9
square root, 342
Stanford School, 501
Stanwick, P. E., 511n35
stasis (issues), 333

Page 56 of 64
Index

stipulation clause, 554

Stohler-Zimmermann, A., 88
Stoics, 311
stones, 3
Story of Sinuhe, 283
Story of the Shipwrecked Sailor, 283
Strabo, 522–25, 528
strategos, 526, 528–29, 533–34, 568
stylus, 11, 19
sublinear asterisk, 315n15
subliterary papyri
Arabic papyri, 456–57
Checklist of Editions, 299n7, 299n8
Diegeseis, 287
editing of, 652
Egyptian, 282–84
grammatical treatises, 288, 295–99
Greek papyri, 291
Mertens-Pack, 285
summonses, 376–77
superlatives, 155
superscription, 172–74
supralinear letters, 172–74
surveys, 186, 525
suspension, 171–72, 175
Syene, 397, 409
syllables, 323–25
symbols, 175, 176
synallagmatographoi, 549
Synaxeis of the Living Gospel, 638
syngraphophylax, 544
syntax, 154–57, 163
Syria
Edessa, 474, 483, 487n10
Medinet Madi codices, 638
religion, 572
war of scepters, 372
Zenon archive, 234
system B, 350
Taeneotic papyrus, 6
Al-Tahtawi, Rifaʼa, 31, 55n3
Tait, W. J., 26n12
talents, 189–91
Talit, 46–47
Tal-Virtù, Malta, 645
Tanis, 92
tax receipts
abbreviations, 171

Page 57 of 64
Index

archives, 245
Bactrian, 473
bilingualism, 408
codex, 224
Karnak, 43–44
Near Eastern, 473
ostraca, 15–16, 383
Petra papyri, 480
Wilcken ostraca, 63
taxes. See also tax receipts
age of liability, 379
in Arabic papyri, 461–63
archives, 237
conveyance tax, 545, 558n7
death notices, 381
epigraphê, 533
epikrisis registers, 380
fifteen-year cycle, 534–35
funerary, 235
harvest tax, 574
inheritance tax, 553–54
measures for, 186
one-sixth tax (apomoira), 526
ostraca receipts, 15–16
payment for, 371
Petra papyri, 486
Philip the Arabian, reforms, 533
and plague, 373
poll tax, 380–82, 462–63, 504, 574–75
privatization of, 535
privileges, 406
protest about, 463
Ptolemaic era, 382, 530–32
receiving measure, 186–87
Revenue Laws papyrus, 524, 530–32
Roman era, 533
sales tax, 384
salt tax, 382, 402, 408, 487–88n23
sitologos, 382–83, 389n65
survey of land, 525
tax farmers (telonai), 530
tax protest, 463
tax rolls, 383
temple administration, 566–67
Zenon archive, 234
Tebtynis. See also Fayyum
Arabic papyri, 460
astrology, 351

Page 58 of 64
Index

bilingual archives, 241

“Cantina dei papiri,” 227, 228
cartonnage, 45
crocodile mummies, 46, 221
dating, 43
(p. 686) demotic writing, 513n60

excavations, 647
family archive, 233–34
Greek language, 582
legal documents, 364
literary papyri, 236, 507
oracle questions, 579
ostraca, 40
Roman names, 514n77
scientific archaeology, 40, 55
scientific works, 566
Soknebtynis priests, 503–7, 513n55
Technê Grammatikê, 295–99
technical vocabulary
Arabization, 460
Aramaic texts, 407
bilingualism, 444
Fachwörter, 198
Greek, 194
Manichaeism, 639
translations, 208–9
Tell el-Masquta, 33
telonai (tax farmers), 530
temples
accounts, 371–72
administration of, 573–76
archives, 219, 236
astrology, 566
asylum, 575
Book of the Temple, 568
of the Carians, 404
decline of Egyptian, 430
dreams, 566, 584
Edfu temple, 562, 569
Egyptian religion, 561–64
Greek inscriptions, 581–82
Idumaeans, 405
Jewish, 398
and justice, 566–68
landowners, 565–66
medical practice, 572
oracles, 584
property ownership, 573–74

Page 59 of 64
Index

Terence, 286–87
Terentianus, Claudius, 162, 166
testa, 14
tetradrachm, 189–90
Tetragrammaton, 593
Tetrarchy, 183, 373
textual critics, 212–13
Theadelphia, 49, 383, 606
theagoi, 565, 565
Thebaid, 184, 523, 534
Thebes, 31, 241, 565, 571, 584
Theodosius, 376, 422
Theon of Alexandria, 332–33
thesauroi (granaries), 530
Thesaurus Linguae Graecae (TLG), 198, 290–91, 652
Third Syrian War, 372
Thmouis, 44, 92, 94
Thousand and One Nights, 457
Three Tales of Wonder, 283–84
Tiberius, 190
Timotheos papyrus, 44, 285, 405
title deeds, 219, 228
title tags, 264
tolls, 383
tombs, 44–45, 80, 497
tomos synkollêsimos, 228, 243, 244
topogrammateis, 527
topos/toparchy, 184–85, 526, 527, 529, 531, 534
Totoes, 223
Toubias, 474
trade, 648
trading licenses, 530–31
transcription, 74, 203
translations, 207–8, 445, 650
transliterations, 152–53, 167n5, 407, 427, 429, 431
transversa charta, 21, 22, 482, 485
trapezai (banks), 530
Traube, L., 592–93
tremissis, 191
trials, 377, 425, 427
trots, 298
tufa, 312
tuff, 92
Tura codices, 599–600
Turfan, 632
Turner, Eric G., 70, 266, 278n9, 596, 616n89
Tutu, 578, 583
Tyrwhitt, Thomas, 315n17

Page 60 of 64
Index

Uigur, 632
Ulpian, 18–19, 21
Umayyads, 459, 461, 463, 480–81
umbilicus, 19, 263–64
uncial writing, 142, 145, 241, 444
UNESCO, 69, 497, 510n11
University of California, 63, 198, 290
University of Chicago, 456
University of Cologne, 72
University of Michigan
Arabic papyri, 457
electronic databases, 648–49
Karanis excavations, 35, 38–39, 67
Michigan Papyri, 212
papyrus cartels, 66–67
Pauline epistles, 596
University of Wisconsin, 66
University of Zurich, 454
upright ogival majuscule, 132, 134, 136
Urkunden der Ptolemäerzeit, 212
Valerian, 606
van Groningen, B. A., 202
(p. 687) van Haelst, J., 591–92

van Hoesen, H. B., 346–47

van Minnen, Peter
AIP papers, 644
gymnasium, 513n66
historiography, 496
house-to-house-approach, 48, 226–27
Karanis, 511n30
Soknebtynis dossier, 504
ten commandments, 652
Vandorpe, Katelijn, 230, 499
Vasunia, Phiroze, 498
Vatican, 624
vellum, 11
verbs, 155–56
Vergilian codices, 141
Verhoogt, A. M. F. W., 227
Verschleifungen (writing slurs), 177
verses, 259, 352
vessels, 353
Vesuvius, 59, 303
Vettius Valens, 346
Viereck, Paul, 63
Villa dei Papiri, 303–5
Villa dei Pisoni, 92
Vindolanda writing tablets, 14, 23, 137–38, 141, 149, 385–86

Page 61 of 64
Index

Virgin Mary, 586

Vitelli, Girolamo, 71
vocabulary, 163, 165, 407
volumen, 25
vowels, 152, 154, 174, 455
Vulgar Latin, 162–63, 166
Wadi Daliyeh, 476
Wadi Natrun, 5
wall paintings, 305, 314n5
“war of scepters”, 219, 372
war report, 372
waxed tablets
birth certificates, 380
concatenated, 23
Dura-Europos, 477
Latin handwriting development, 137
mathematics, 340
school exercises, 323
as writing material, 4, 11–12
weak ties, 506
weather prediction, 339, 344
Weber, Dieter, 419
Weber, Karl, 305, 312
weights, 175, 188–89
Weintraub, Jock, 509
Welles, C. Bradford, 73–74, 595
Wenger, Leopold, 69–70
Wessely, Carl, 70
West, M. L., 293
Westermann, William Linn, 71
wet-nursing contracts, 655
wheat, 382
Wilamowitz-Moellendorff, Ulrich von, 63
Wilcken, Ulrich
abbreviations, 170, 173n1
Archiv für Papyrusforschung, 63–65
Herakleopolis, 34
historiography, 496
Karnak, 43–44
papyrus editing, 212
pre-WWI excavations, 36
wills
Abraham of Hermonthis, 439, 445
agoranomic documents, 558n14
Coptic, 439
double documents, 485
language for, 422
Latin language, 553

Page 62 of 64
Index

law of the Egyptians, 550

legal documents, 367, 542
Wilson, N. G., 612n26
Winckelmann, J. J., 304, 309–10
wisdom literature, 413, 414, 583
Wiseman, D. J., 27n42
witch of Endor, Origen's homily on, 599
women
Arabic commerce, 465
Babatha archive, 476
contracts, 445
Gnostics, 629
guardianship, 555
ius trium liberorum, 366, 555
Near Eastern papyri, 483
property ownership, 366
right of three children, 366, 555
wills, 367
wooden tablets, 12, 13
abbreviations, 170
archives, 245
concatenated, 23
frequency of use, 4
Manichaeism, 639
mathematics, 340
mummy labels, 371–72
school exercises, 322–23, 326
wooden leaf tablets, 49
as writing material, 3
word division, 158, 164, 262, 266, 273, 277n4
World War II, 211
Wörterbuch, 65, 198–99
Xusro, 419
xystos, 187
Yadin, Yigael, 49–50
Yale University, 66–67, 648–49
years, 182–83
Youtie, Herbert C., 72–74, 496
(p. 688) Zeitschrift für Papyrologie und Epigraphik, 72

Zeno, 311
Zenon archive
Alexandrian chancery, 103
antiquities trade, 224, 225, 226
bilingualism, 408–9
biography, 218
business correspondence, 234–35, 243
chancery style, 105
clothing, 371

Page 63 of 64
Index

Guide to the Zenon Archive, 70, 226, 230, 655, 658

Near Eastern papyri, 474
petitions, 406
Préaux presentation, 68
private archives, 48
private documents, 361–62
Zereteli, Grigorij F., 70–71, 74
zero-grade hand, 323–24
zetêmata, 331–32
Zeugma, 487–88n23
zodiacal signs, 344, 346–52, 580–81
Zoroastrianism, 636
Zuckerman, Constantin, 329, 500–501, 512n41
Zuntz, G., 329

Page 64 of 64