THE EUROPEAN BOOK IN THE TWELFTH

CENTURY

The ‘Long Twelfth Century’ (1075–1225) was an era of seminal

importance in the development of the book in medieval Europe
and marked a high point in its construction and decoration. This
comprehensive study takes the cultural changes that occurred during
the ‘Twelfth-Century Renaissance’ as its point of departure to provide
an overview of manuscript culture encompassing the whole of
Western Europe. Written by senior scholars, chapters are divided
among three sections: the technical aspects of making books; the
processes and practices of reading and keeping books; and the trans-
mission of texts in the disciplines that saw significant change in the
period, including medicine, law, philosophy, liturgy and theology.
Richly illustrated, this volume provides the first in-depth account of
book production as a European phenomenon.

erik kwakkel is Professor at the School of Library, Archival and

Information Studies at the University of British Columbia,
Vancouver. His research is devoted to the relationship between the
physical appearance of manuscripts and the historical context in
which they were produced and used. His publications include
Turning Over a New Leaf: Change and Development in the Medieval
Book (2012), co-authored with Rosamond McKitterick and Rodney
Thomson; Manuscript of the Latin Classics 800–1200 (2015); Writing in
Context: Insular Manuscript Culture 500–1200 (2013) and Author,
Reader, Book: Medieval Authorship in Theory and Practice (2012), co-
edited with Stephen Partridge.
rodney thomson (faha) is Emeritus Professor of Medieval
History at the University of Tasmania. His publications include
Books and Learning in Twelfth-Century England: The Ending of
‘Alter Orbis’ (2006) and, as co-editor with Nigel Morgan, The
Cambridge History of the Book in Britain 2: The Manuscript Book
c.1100–1400 (2008). Professor Thomson has compiled descriptive
catalogues of manuscript collections held at Lincoln, Hereford and
Worcester Cathedrals, of Merton and Corpus Christi Colleges,
Oxford, and Peterhouse, Cambridge.
 cambridge studies in medieval literature
General Editor
Alastair Minnis, Yale University

Editorial Board
Zygmunt G. Barański, University of Cambridge
Christopher C. Baswell, Barnard College and Columbia University
Mary Carruthers, New York University
Rita Copeland, University of Pennsylvania
Roberta Frank, Yale University
Jocelyn Wogan-Browne, Fordham University

This series of critical books seeks to cover the whole area of literature written in the
major medieval languages – the main European vernaculars, and medieval Latin
and Greek – during the period c.1100–1500. Its chief aim is to publish and
stimulate fresh scholarship and criticism on medieval literature, special emphasis
being placed on understanding major works of poetry, prose, and drama in
relation to the contemporary culture and learning which fostered them.

Recent titles in the series

Lee Manion Narrating the Crusades: Loss and Recovery in Medieval and Early
Modern English Literature
Daniel Wakelin Scribal Correction and Literary Craft: English Manuscripts
1375–1510
Jon Whitman (ed.) Romance and History: Imagining Time from the Medieval to the
Early Modern Period
Virginie Greene Logical Fictions in Medieval Literature and Philosophy
Michael Johnston and Michael Van Dussen (eds.) The Medieval Manuscript Book:
Cultural Approaches
Tim William Machan (ed.) Imagining Medieval English: Language Structures and
Theories, 500–1500
Eric Weiskott English Alliterative Verse: Poetic Tradition and Literary History
Sarah Elliott Novacich Shaping the Archive in Late Medieval England: History,
Poetry, and Performance
Geoffrey Russom The Evolution of Verse Structure in Old and Middle English
Poetry: From the Earliest Alliterative Poems to Iambic Pentameter
Ian Cornelius Reconstructing Alliterative Verse: The Pursuit of a Medieval Meter
Sara Harris The Linguistic Past in Twelfth-Century Britain
Erik Kwakkel and Rodney Thomson (eds.) The European Book in the Twelfth
Century

A complete list of titles in the series can be found at the end of the volume.
THE EUROPEAN BOOK IN THE
TWELFTH CENTURY

edited by
ERIK KWAKKEL
University of British Columbia, Vancouver

RODNEY THOMSON
University of Tasmania
 University Printing House, Cambridge cb2 8bs, United Kingdom
One Liberty Plaza, 20th Floor, New York, ny 10006, USA
477 Williamstown Road, Port Melbourne, vic 3207, Australia
314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India
79 Anson Road, #06–04/06, Singapore 079906

Cambridge University Press is part of the University of Cambridge.

It furthers the University’s mission by disseminating knowledge in the pursuit of
education, learning, and research at the highest international levels of excellence.

www.cambridge.org
Information on this title: www.cambridge.org/9781107136984
doi: 10.1017/9781316480205
© Cambridge University Press 2018
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2018
Printed in the United Kingdom by TJ International Ltd. Padstow Cornwall
A catalogue record for this publication is available from the British Library.
isbn 978-1-107-13698-4 Hardback
Cambridge University Press has no responsibility for the persistence or accuracy of
URLs for external or third-party internet websites referred to in this publication
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
Dedicated to the memory of
Peter Gumbert
(1936–2016)
 Contents

List of Figures page ix

List of Contributors xii
Preface xvii
List of Abbreviations xviii

Introduction 1
Erik Kwakkel and Rodney Thomson

part i book production 7

1 Codicology 9
Erik Kwakkel and Rodney Thomson
2 Book Script 25
Erik Kwakkel
3 Decoration and Illustration 43
Martin Kauffmann
4 Scribes and Scriptoria 68
Rodney Thomson

part ii readers and their books 85

5 Scholars and Their Books 87
Constant J. Mews
6 The Libraries of Religious Houses 103
Teresa Webber
7 Modes of Reading 122
Jenny Weston

vii
viii Contents
8 Practices of Appropriation: Writing in the Margin 139
Mariken Teeuwen

part iii types of books 157

9 Hebrew Books 159
Judith Olszowy-Schlanger
10 Liturgical Books 175
Nicolas Bell
11 Books of Theology and Bible Study 192
Lesley Smith
12 Logic 215
John Marenbon and Caterina Tarlazzi
13 The Classical Revival 240
Irene O’Daly
14 Reading the Sciences 259
Charles Burnett
15 Medical Books 277
Monica H. Green
16 Law Books 293
Charles M. Radding
17 Vernacular Manuscripts I: Britain and France 311
Ian Short
18 Vernacular Manuscripts II: Germany 327
Nigel F. Palmer

Bibliography 345
Index of Manuscripts 394
General Index 400
 Figures

1.1 Page heights of 353 dated and datable manuscripts, 1075–1224. page 12
1.2 Number of columns in 353 dated and datable manuscripts,
1075–1224. 17
1.3 A representative group of twelfth-century bindings from Hereford
Cathedral (Mynors and Thomson, Hereford, pl. 21). 19
2.1 Pregothic book script, dated 1145–9. Leiden, Universiteitsbibliotheek,
BPL 196. Reproduced with permission. 27
2.2 Three examples of substituting letterforms. 29
2.3 Examples of increased popularity of placing feet on baseline. 31
2.4 Pregothic documentary script used for added glosses, eleventh
century, with twelfth-century marginal and interlinear glosses.
Leiden, Universiteitsbibliotheek, VLQ 51. Reproduced with
permission. 37
3.1 Frontispiece to the commentary of Jerome on the book of Isaiah,
illuminated by Hugo ‘pictor’, Normandy, late eleventh century.
Oxford, Bodleian Library, Bodley 717, ff. vv–vir. Reproduced with
permission of The Bodleian Libraries, The University of Oxford. 44
3.2 Frontispiece to volume 2 of the Floreffe Bible, showing scenes from
the book of Job, virtues and works of mercy, and Christ and the
Apostles, described by the tituli as a representation of the Active
Life; valley of the Meuse (modern Belgium), ca. 1153–6. London,
British Library, Add. 17738, f. 3v. Reproduced with permission. 47
3.3 Initial F(rater Ambrosius), letter of Jerome to Paulinus, from the
Bury Bible: England, ca. 1135. Cambridge, Corpus Christi College,
2, f. 1v. Reproduced with permission of the Master and Fellows
of Corpus Christi College, Cambridge. 50
3.4 Drawing of man as microcosm, from the Prüfening Miscellany,
Germany, 1158–65. Munich, Bayerische Staatsbibliothek,
Clm 13002, f. 7v. Reproduced with permission. 52

ix
x List of Figures
4.1 Handwriting of Diemut, inclusa of Wessobrunn. Munich,
Bayerische Staatsbibliothek, Clm 22009, f. 4v. Reproduced
with permission. 77
6.1 Recesses for books in the cloister at Escaladieu. Photograph by
the author. 107
6.2 Ground plan of Roche Abbey. Adapted from R. Gilyard-Beer,
Abbeys: An Introduction to the Religious Houses of England and
Wales (London, 1958). Reproduced with permission. 109
7.1 Chapter table. Rouen, Bibliothèque municipale, 1, f. 17r.
Reproduced by permission of Collections de la Bibliothèque
municipale de Rouen. 129
7.2 Chapter numbers and lection marks. Rouen, Bibliothèque
municipale, 1, f. 19r. Reproduced by permission of Collections
de la Bibliothèque municipale de Rouen. 130
8.1 Two hands entering commentary in the margins. Leiden,
Universiteitsbibliotheek, BPL 144, f. 10r. Photograph by the author. 144
8.2 Faces on the edges. Leiden, Universiteitsbibliotheek, BPL 189,
ff. 44r and 45r. Photograph by the author. 147
9.1 Bible, masora magna and parva, La Rochelle, 1215. Vatican City,
Bibliotheca Apostolica Vaticana, Vat. ebr. 468, f. 25r. Reproduced
with permission. 162
9.2 Pentateuch with interlinear Targum, masora magna and parva,
England or Normandy, late twelfth century. Jerusalem, National
Library of Israel, Heb. 4°5827, ff. 25v–26r. Reproduced with
permission. 166
10.1 Office and Mass of St Cuthbert, Durham, mid twelfth century.
Cambridge, Trinity College, O. 3. 55, f. 54r. Reproduced with
permission of the Master and Fellows of Trinity College,
Cambridge. 187
10.2 Pontifical, perhaps from Ely, ca. 1125–50. Cambridge, Trinity College,
B. 11. 10, f. 50r. Reproduced with permission of the Master and
Fellows of Trinity College, Cambridge. 188
11.1 Layout of the Glossa ordinaria. Oxford, Bodleian Library,
E. D. Clarke 35, ff. 53v–54r. Reproduced with permission
of The Bodleian Libraries, The University of Oxford. 196
11.2 Layout of the Glossa ordinaria. Oxford, Bodleian Library,
Auct. E. inf. 7, ff. 118v–119r. Reproduced with permission
of The Bodleian Libraries, The University of Oxford. 198
11.3 Peter Lombard’s Psalms commentary made for Thomas Becket.
Cambridge, Trinity College, B. 5. 4, f. 135v. Reproduced with
permission of the Master and Fellows of Trinity College, Cambridge. 202
 List of Figures xi
11.4 Hugh of St-Cher, Postilla on Revelation. Oxford, Bodleian Library,
Bodley 444, ff. 102v–103r. Reproduced with permission of The
Bodleian Libraries, The University of Oxford. 210
12.1 Depiction of Dialectica. Darmstadt, Universitäts- und
Landesbibliothek, 2282, f. 1v. Reproduced with permission. 218
12.2 Example of phytomorphic initial (commentary P14). Paris,
Bibliothèque nationale de France, lat. 17813, f. 14v. Reproduced
with permission. 227
13.1 Lucan’s Pharsalia, Egmond, just after 1050, with later additions
and glosses. Leiden, Universiteitsbibliotheek, BUR Q 1, ff. 1v–2r.
Reproduced with permission. 249
13.2 Cicero listens to the arguments of Cato and Caesar on the fate
of Catiline and his cohort. Oxford, Bodleian Library, Barlow 40,
f. 1r. Reproduced with permission of The Bodleian Libraries, The
University of Oxford. 253
14.1 A key to Arabic numerals with their names. Cambridge, Trinity
College, R. 15. 16, f. Av. Reproduced with permission of the
Master and Fellows of Trinity College, Cambridge. 270
15.1 Medical ‘bestsellers’ of the Long Twelfth Century. 281
16.1 The Walcausina. Paris, Bibliothèque nationale de France, lat.
9656, f. 13r. Reproduced with permission. 298
16.2 Justinian’s Code, two-column format with ample margins
to accommodate glosses. Leiden, Universiteitsbibliotheek,
d’Ablaing 1, f. 105r. Reproduced with permission. 303
17.1 The Oxford Psalter, St Albans, 1140s. Oxford, Bodleian Library,
Douce 320, f. 54r. Reproduced with permission of The Bodleian
Libraries, The University of Oxford. 318
18.1 Williram von Ebersberg, Commentary on the Song of Songs.
Written area: 198–203 × 146–54 mm. Vienna, Österreichische
Nationalbibliothek, Cod. 2686, f. 17v. Reproduced with permission. 330
18.2 Prayer Book from Muri. Written area: 70 × 50 mm. Sarnen,
Bibliothek des Benediktinerkollegiums, Cod. membr. 69, ff. 1v–2r.
Reproduced with permission. 331
18.3 Pfaffe Konrad, Rolandslied. Written area: 274–80 × 115–25 mm.
Heidelberg, Universitätsbibliothek, Cod. Pal. germ. 112, f. 41v.
Reproduced with permission. 333
18.4 Hartmann von Aue, Iwein. Written area: 126 × ca. 84 mm.
Gießen, Universitätsbibliothek, Hs. 89, ff. 1v–2r. Reproduced with
permission. 338
 Contributors

nicolas bell is Librarian and Fellow of Trinity College, Cambridge.

Until 2015 he was a curator in the music department of the British
Library. His publications include companion studies to facsimiles of two
medieval music manuscripts, as well as articles on medieval liturgy and
music. He is General Secretary of the Henry Bradshaw Society, founded
in 1890 for the publication of rare liturgical texts, and is a member of the
Council of the Plainsong and Medieval Music Society.
charles burnett is Professor of the History of Arabic/Islamic
Influences in Europe at the Warburg Institute, University of London.
His research centres on the transmission of texts, techniques and arte-
facts from the Arab world to the West, especially in the Middle Ages. He
has documented this transmission by editing and translating into
English several texts that were translated from Arabic into Latin, and
also by describing the historical and cultural context of these transla-
tions. Among his books are The Introduction of Arabic Learning into
England (1997), Arabic into Latin in the Middle Ages: The Translators and
Their Intellectual and Social Context (2009) and Numerals and Arithmetic
in the Middle Ages (2010).
monica h. green is a historian of medieval European medicine and
global health. An elected Fellow of the Medieval Academy of America,
she has published extensively on the history of women’s healthcare,
including her edition of The ‘Trotula’: A Medieval Compendium of
Women’s Medicine (2001) and her monograph, Making Women’s
Medicine Masculine: The Rise of Male Authority in Pre-Modern
Gynaecology (2008) both of which were based on surveys of hundreds
of Latin and vernacular manuscripts. She has also engaged with the fields
of palaeogenetics and palaeopathology as ways to help reconstruct the
history of humankind’s major infectious diseases. A focal point of her
continuing work is the radical transition in European medicine in the
xii
 List of Contributors xiii
eleventh and twelfth centuries, which was the first field to begin to
absorb the learned theories and practices recently coming out of the
Islamic world. She has several studies forthcoming on Constantine the
African and medicine at Monte Cassino.
martin kauffmann wrote his doctoral thesis at the Courtauld Institute
in London on thirteenth-century Anglo-Norman illustrated saints’
Lives. He is now Head of Early and Rare Collections and Tolkien
Curator of Medieval Manuscripts in the Bodleian Library, University
of Oxford, and a member of the Faculty of History. He has written
articles and contributed to exhibition catalogues on illuminated manu-
scripts ranging from Ottonian Germany to fifteenth-century England.
Most recently he has been working on the conjunction of images and
prayers in the prefatory cycle of an English twelfth-century psalter.
erik kwakkel is Professor at the School of Library, Archival and
Information Studies at the University of British Columbia,
Vancouver, where he teaches the history of the book. His research
interests are related to quantitative palaeography and the development
of the manuscript’s physical features over time. He has edited several
volumes on the production context of manuscripts for Studies in
Medieval and Renaissance Book Culture, a book series published by
Leiden University Press. At Leiden University he directed the NWO-
sponsored project ‘Turning Over a New Leaf’ (2010–2015), which
studied manuscript culture in the twelfth century. The project brought
together the scholars whose work is united in the present publication.
In 2015 Kwakkel was appointed to the Comité International de
Paléographie Latine (CIPL).
john marenbon is a Senior Research Fellow of Trinity College,
Cambridge and Honorary Professor of Medieval Philosophy in the
University of Cambridge. His two most recent books are Medieval
Philosophy, A Very Short Introduction (2016) and Pagans and
Philosophers: The Problem of Paganism from Augustine to Leibniz (2015).
constant j. mews gained his BA and MA from the University of
Auckland, New Zealand, and his DPhil from Oxford University. He
is Professor within the School of Philosophical, Historical and
International Studies, Monash University, where he is also Director of
the Centre for Religious Studies. He has published widely on medieval
thought, ethics and religious culture, with particular reference to the
writings of Abelard, Heloise, Hildegard of Bingen and their
xiv List of Contributors
contemporaries, including Abelard and Heloise (2005) and The Lost Love
Letters of Heloise and Abelard: Perceptions of Dialogue in Twelfth-Century
France, 2nd edn. (2008). His research interests range from the early
Middle Ages to late medieval religious and intellectual culture, as well as
the interface between various religious and ethical traditions.
irene o’daly is a Researcher at Huygens ING, a research institute of the
Royal Netherlands Academy of Arts and Sciences (KNAW). From 2011
to 2014 she was a postdoctoral researcher at Leiden University on the
project ‘Turning Over a New Leaf: Manuscript Innovation in the Twelfth
Century.’ Her principal research interests include medieval intellectual
history, the history of classical reception and the history of the book. She
has published on the influence of Roman philosophy on the political
thought of John of Salisbury, and on the use of diagrammatic annotation
in medieval Ciceronian rhetorical manuscripts. Her current research
concerns traditions of visualising knowledge and debate in medieval
theological manuscripts.
judith olszowy-schlanger is Professor of Hebrew Palaeography and
Manuscript Studies at the École pratique des hautes études, Institut de
recherche et d’histoire des textes, Paris. She holds a doctorate from the
University of Cambridge and is a Corresponding Fellow of the British
Academy. Her main research interests are in medieval Hebrew palaeo-
graphy, Cairo Genizah studies, Jewish–Christian intellectual relations
during the European Middle Ages and Hebrew diplomatics. Her pub-
lications include Les manuscrits hébreux dans l'Angleterre médiévale: étude
historique et paléographique (2003).
nigel f. palmer fba is Emeritus Professor of German Medieval and
Linguistic Studies at the University of Oxford and Emeritus Fellow of
St Edmund Hall. He is a corresponding fellow of the Medieval
Academy of America and of the Göttingen Academy of Sciences. His
main areas of research are medieval German and Latin religious litera-
ture, palaeography and codicology, and early printing. His doctoral
thesis, from 1976, was on the German and Dutch reception of the Visio
Tnugdali, and he has published extensively on blockbooks. His most
recent major publication, jointly with Jeffrey F. Hamburger, is the
Prayer Book of Ursula Begerin, 2 vols. (2015).
charles m. radding is Professor of History at Michigan State
University. He is interested in medieval cultural history, with
a specialization in the development of scholarly disciplines between
 List of Contributors xv
the tenth and thirteenth centuries. His books include A World Made by
Men: Cognition and Society 400–1200 (1985) and The Origins of Medieval
Jurisprudence: Pavia and Bologna, 850–1150 (1988).
ian short is Emeritus Professor of French at the University of London.
He also taught at the University of California, Berkeley and at Paris
Nanterre University. His principal fields of research are eleventh-
century literary culture in Britain, and Anglo-Norman language and
literature. He is president of the Anglo-Norman Text Society, and was
its secretary from 1974 until 2011. He has published widely and has
edited a number of medieval French texts. He is co-author (together
with Maria Careri and Christine Ruby) of Livres et écritures en français et
en occitan au XIIe siècle: Catalogue illustré (2011).
lesley smith is Professor of Medieval Intellectual History at Oxford
University. She studies the Bible and its commentaries, both as physical
and as intellectual objects. Recent work includes The Glossa Ordinaria:
The Making of a Medieval Bible Commentary (2009) and The Ten
Commandments: Interpreting the Bible in the Medieval World (2014).
caterina tarlazzi is British Academy Postdoctoral Fellow at the
Faculty of Philosophy, University of Cambridge, and a Research
Associate of St John’s College Cambridge. She works on a project
called ‘Logic in the Early Twelfth Century: A Manuscript-Based
Approach.’ She was awarded her PhD (Doctor Europaeus) in
Medieval Philosophy in 2013, in a co-tutelle programme between
Paris-Sorbonne University and Università degli Studi di Padova.
Her thesis (Individui universali. Il realismo di Gualtiero di Mortagne
nel XII secolo) was awarded the 2014 Ana María Aldama Roy Prize for
medieval studies. She works on logic in the time of Peter Abelard and
has a particular interest in the manuscripts transmitting twelfth-
century philosophical texts.
mariken teeuwen studied musicology and medieval studies at Utrecht
University. She is Senior Researcher at Huygens ING, a research insti-
tute of the Royal Netherlands Academy of Arts and Sciences (KNAW),
and Professor at the Department of History and Art History, University
of Utrecht, where she teaches the transmission of medieval Latin texts.
She is Principal Investigator of the NWO-funded research projects
‘Marginal Scholarship: The Practice of Learning in the Early Middle
Ages (c. 800–c. 1000)’ (2011–16), and ‘The Art of Reasoning: Practices of
Scientific Argumentation in the Middle Ages (ca. 400–1400)’ (2016–20).
xvi List of Contributors
Her publications focus on the medieval reception of the ancient learned
tradition, the vocabulary of intellectual life in the Middle Ages and
practices of annotating manuscripts.
rodney thomson is Emeritus Professor of Medieval History at the
University of Tasmania. He has published extensively on books and
learning in medieval Europe, with special attention to England from the
Norman Conquest to the early thirteenth century. In 2000–1 he was Lyell
Reader in Bibliography at Oxford University and in February 2017 he
delivered the Lowe Lectures in Palaeography at Corpus Christi College.
He has compiled descriptive catalogues of the medieval manuscripts held
by three English cathedrals, and by Merton and Corpus Christi Colleges,
Oxford and Peterhouse, Cambridge. He has also studied the English
Benedictine monk and scholar William of Malmesbury (d. 1143). He is
Fellow of the Australian Academy of the Humanities, of the Royal
Historical Society and of the Society of Antiquaries.
teresa webber is University Reader in Palaeography in the Faculty of
History, University of Cambridge and a Fellow of Trinity College,
Cambridge. She has published widely on the history of the production,
ownership and use of books in Britain in the twelfth century. She is the
author of Scribes and Scholars at Salisbury Cathedral c. 1075–c. 1125 (1992)
and co-editor of The Cambridge History of Libraries in Britain and
Ireland, Volume I: To 1640 (2006) and The Libraries of the Augustinian
Canons, Corpus of British Medieval Library Catalogue, vol. 6 (1998). She
is currently completing for publication a major study of public reading
and its books in monastic practice in England, ca. 1000–ca. 1300, which
was the subject of the lectures she gave as Oxford University’s
J. P. R. Lyell Reader in Bibliography, 2015–16.
jenny weston was awarded her PhD from Leiden University in 2015.
As a junior researcher for the NWO-funded Vidi Project ‘Turning
Over a New Leaf: Manuscript Innovation in the Twelfth-Century
Renaissance’ (Leiden University), Jenny specialised in manuscript stu-
dies, with a particular focus in reading and the material book. At present,
Jenny teaches for the English Department at Rutgers University in New
Jersey and is preparing a monograph on monastic reading practices in
the twelfth century.
 Preface

This book was conceived in the context of ‘Turning Over a New Leaf:
Manuscript Innovation in the Twelfth-Century Renaissance’, a research
project directed by Erik Kwakkel and established at Leiden University
from 2010 to 2015. It fulfils the desire of both editors to make available
a synthesizing study devoted to all major subject areas of twelfth-century
manuscript culture in the West. In the autumn of 2014 a group of leading
scholars was invited to write chapters for this book. The aim was not only
to present the current state of research in the various subject areas, but also
to highlight the possibilities for new directions in the future. Moreover, the
chapters were intended to draw attention, insofar as possible within the
limitations of space, to the material features of twelfth-century manu-
scripts, an emphasis that was the driving force behind the ‘Turning Over
a New Leaf’ project. The contributors were subsequently invited to a three-
day workshop held in March 2015 at Leiden University, during which each
chapter was presented and discussed. The result of this phased production
process, in which the authors built upon the input and expertise of their
colleagues, is a book that analyzes, for the first time, the full scope of
twelfth-century book and text culture in the West.
The editors wish to thank the following people and institutions: the
libraries that granted permission for the reproduction of images; the
contributors for their enthusiasm and willingness to come to Leiden and
critique each other’s work; the Dutch funding agency Nederlandse
Organisatie voor Wetenschappelijk Onderzoek (NWO), which funded
the ‘Turning Over a New Leaf’ project and the workshop in 2015; Leiden
University Library for hosting the workshop; and Jenneka Janzen and Julie
Somers, who produced a discussion report that proved vital for writing the
final version of the chapters. This book is dedicated to the memory of the
Leiden codicologist Peter Gumbert (1936–2016), who was a respondent at
the workshop and an enthusiastic supporter of the ‘Turning Over a New
Leaf’ project.

xvii
 Abbreviations

AHDLMA Archives d’histoire doctrinale et littéraire du

moyen âge
AL Aristoteles Latinus: Codices, ed. G. Lacombe,
A. Birkenmajer, M. Dulong and
E. Francescini (2 vols., Rome, 1939, 1945),
with supplement, ed. L. Minio-Paluello
(Bruges/Paris, 1961)
Alexander, Medieval J. J. G. Alexander, Medieval Illuminators and
Illuminators Their Methods of Work (New Haven, CT,
1992)
Alexander, J. J. G. Alexander, Norman Illumination at
Mont St Michel Mont St Michel 966–1100 (Oxford, 1970)
ANS Anglo-Norman Studies
BAV Biblioteca Apostolica Vaticana
Beach, Manuscripts Manuscripts and Monastic Culture: Reform
and Monastic and Renewal in Twelfth-Century Germany,
Culture ed. A. I. Beach (Medieval Church Studies 13:
Turnhout, 2007)
Beach, Women A. I. Beach, Women as Scribes: Book
as Scribes Production and Monastic Reform in Twelfth-
Century Bavaria (Cambridge, 2004)
Becker, Catalogi G. Becker, Catalogi Bibliothecarum Antiqui
(Bonn, 1885; repr. Hildesheim, 1973)
Benson and Constable, R. L. Benson and G. Constable with
Renaissance C. D. Lanham, The Renaissance of the Twelfth
Century (Oxford, 1982)
Bibl. mun. Bibliothèque municipale

xviii
 List of Abbreviations xix
Bischoff, Latin B. Bischoff, Latin Palaeography: Antiquity
Palaeography and the Middle Ages, trans. D. Ó Cróinín
and D. Ganz (Cambridge, 1990)
BL London, The British Library
BnF Paris, Bibliothèque national de France
Bodl. Libr. Oxford, The Bodleian Library
BSB Munich, Bayerische Staatsbibliothek
Bumke, Mäzene J. Bumke, Mäzene im Mittelalter: die Gönner
und Auftraggeber der höfischen Literatur in
Deutschland, 1150–1300 (Munich, 1979)
Burnett and C. Burnett and D. Jacquart (eds.),
Jacquart, Constantine Constantine the African and ‘Alī ibn al-’Abbās
al-Mağūsī: The ‘Pantegni’ and Related Texts
(Leiden, 1994)
Cahn, Romanesque W. Cahn, Romanesque Manuscripts: The
Manuscripts Twelfth Century (a survey of manuscripts
illuminated in France: 2 vols., London, 1996)
Cavallo and Chartier, G. Cavallo and R. Chartier (eds.), A History of
Reading Reading in the West (Cambridge/Oxford,
1999)
CBMLC Corpus of British Medieval Library Catalogues
(The British Library, 1990–)
CCCC Cambridge, Corpus Christi College
CCCM Corpus Christianorum Continuatio
Mediaeualis
CCSL Corpus Christianorum Series Latina
Chavannes-Mazel and Medieval Manuscripts of the Latin Classics:
Smith, Latin Classics Production and Use, ed. C. A. Chavannes-
Mazel and M. M. Smith (Los Altos Hills, CA,
1996)
CHBB I–II Cambridge History of the Book in Britain I,
c. 400–1100, ed. R. Gameson; II, 1100–1400,
ed. N. J. Morgan and R. M. Thomson
(Cambridge, 2011, 2008)
CHL The Cambridge History of Libraries in Britain
and Ireland, Volume I: to 1640, ed.
E. Leedham-Green and T. Webber
(Cambridge, 2006)
CIMAGL Cahiers de l’Institut du moyen-âge grec et latin
xx List of Abbreviations
CUL Cambridge University Library
De Hamel, Glossed Books C. de Hamel, Glossed Books of the Bible and
the Origins of the Paris Booktrade
(Woodbridge, 1984)
Derolez, Gothic A. Derolez, The Palaeography of Gothic
Manuscript Books: From the Twelfth to the
Early Sixteenth Century (Cambridge, 2003)
Dodwell, Canterbury C. R. Dodwell, The Canterbury School of
Illumination (Cambridge, 1954)
Eadwine M. T. Gibson, T. A. Heslop, R. W. Pfaff
(eds.), The Eadwine Psalter:Text, Image and
Monastic Culture in Twelfth-Century
Canterbury (London, 1992)
Gullick, ‘Professional M. Gullick, ‘Professional Scribes in Eleventh-
Scribes’ and Twelfth-Century England’, English
Manuscript Studies 1100–1700 7 (1998), 1–24
Haskins, Renaissance C. H. Haskins, The Renaissance of the Twelfth
Century (Cambridge, MA, 1927)
HBF Histoire des bibliothèques françaises, I: Les
bibliothèques médiévales du vie siècle à 1530,
ed. A. Vernet (Paris, 1989)
Hunt Essays Medieval Learning and Literature: Essays
Presented to R. W. Hunt, ed.
J. J. G. Alexander and M. T. Gibson (Oxford,
1976)
Illich, Vineyard I. Illich, In the Vineyard of the Text: A
Commentary to Hugh’s Didascalicon
(Chicago, IL, 1993)
Jeauneau, ‘Prologue’ E. Jeauneau, ‘Le Prologus in Eptateuchon de
Thierry de Chartres’, Mediaeval Studies 14
(1954), 171–5, repr. in his ‘Lectio philoso-
phorum’: Recherches sur l’école de Chartres
(Amsterdam, 1973), 87–91
JWCI Journal of the Warburg and Courtauld
Institutes
Ker, BCL N. R. Ker, Books, Collectors and Libraries:
Studies in the Medieval Heritage, ed.
A. G. Watson (London and Ronceverte,
1985)
 List of Abbreviations xxi
Ker, English MSS N. R. Ker, English Manuscripts in the Century
after the Norman Conquest (Oxford, 1960)
Ker Essays Medieval Scribes, Manuscripts and Libraries:
Essays Presented to N. R. Ker, ed. M. B. Parkes
and A. G. Watson (London, 1978)
Kwakkel, Latin E. Kwakkel (ed.), Manuscripts of the Latin
Classics Classics (Leiden, 2015)
Liber Eliensis Liber Eliensis, ed. E. O. Blake (RHS Camden
3rd ser. 92, 1962)
MBKDS Mittelalterliche Bibliothekskataloge
Deutschlands und der Schweiz, ed.
P. Lehmann et al. (4 vols., Munich, 1918–79)
Medieval Book Medieval Book Production: Assessing the
Production Evidence, ed. L. L. Brownrigg (Los Altos
Hills, CA, 1990)
MGH Monumenta Germanica Historica
srg scriptores rerum germanicarum
Scriptores Scriptores in usum scholarum
MJ Mittellateinisches Jahrbuch
MMBL N. R. Ker, Medieval Manuscripts in British
Libraries (5 vols., Oxford, 1969–2002)
Munk Olsen, B. Munk Olsen, L’étude des auteurs classiques
e e
Auteurs classiques latins aux XI et XII siècles (4 vols., Paris,
1982–2014)
Mynors, Cassiodorus Cassiodorus, Institutiones, ed.
R. A. B. Mynors (Oxford, 1937)
Mynors, Durham R. A. B. Mynors, Durham Cathedral
Manuscripts to the End of the Twelfth Century
(Oxford, 1939)
Mynors and Thomson, R. A. B. Mynors and R. M. Thomson,
Hereford Catalogue of the Manuscripts in Hereford
Cathedral Library (Cambridge, 1993)
ÖNB Vienna, Österreichische Nationalbibliothek
Orderic Orderic Vitalis, Historia Ecclesiastica, ed. and
trans. M. Chibnall (6 vols., Oxford, 1969–80)
Parkes, Pause and M. B. Parkes, Pause and Effect:
Effect An Introduction to the History of Punctuation
in the West (Aldershot, 1992)
xxii List of Abbreviations
PL Patrologia Latina
RB Revue Bénédictine
Reynolds, Texts and Texts and Transmission: A Survey of the Latin
Transmission Classics, ed. L. G. Reynolds (Oxford, 1984)
RHS Royal Historical Society
Riedmayer, Lambeth J. Riedmayer, Die ‘Lambeth Bibel’: Struktur
Bibel und Bildaussage einer englischen
Bibelhandschrift des 12. Jahrhunderts
(Frankfurt, 1994)
Rouse and Rouse, R. H. Rouse and M. A. Rouse, Authentic
Authentic Witnesses Witnesses: Approaches to Medieval Texts and
Manuscripts (Notre Dame, IN, 1991)
RS Rolls Series
RTAM Recherches de théologie ancienne et médiévale
Shepard, Lambeth D. M. Shepard, Introducing the Lambeth
Bible: A Study of Texts and Imagery
(Turnhout, 2007)
Thomson, Bury Bible R. M. Thomson, The Bury Bible
(Woodbridge and Tokyo, 2001)
Thomson, St Albans R. M. Thomson, Manuscripts from Saint
Albans Abbey 1066–1235 (2nd edn., 2 vols.,
Woodbridge, 1985)
Thomson, Worcester R. M. Thomson, A Descriptive Catalogue of
the Medieval Manuscripts in Worcester
Cathedral Library (Cambridge, 2001)
Turning over a E. Kwakkel, R. McKitterick and
New Leaf R. M. Thomson, Turning Over a New Leaf:
Change and Development in the Medieval Book
(Leiden, 2012)
The World of John The World of John of Salisbury, ed. M. Wilks
of Salisbury (Oxford, 1984)
 Introduction
Erik Kwakkel and Rodney Thomson

With its roots in the fourth century CE, the medieval codex is an old soul
which pushed its established competitors, the papyrus book and parch-
ment scroll, into the margins of history with surprising speed and ease.1
Christian culture in late antiquity needed a medium that could both hold
longer texts and help readers access them more easily than was possible
with the scroll.2 The success of the codex reflects its ability to meet this
requirement. However, the first surviving codices from the Latin West
looked very different from manuscripts made in the later Middle Ages.
The physical format of manuscripts developed significantly throughout the
medieval period, because new physical traits, production methods and
scribal practices were introduced, either to improve established practices
or for no apparent reason at all. The number and speed of such develop-
ments, it appears, increased during periods of cultural and intellectual
change; for example, Carolingian rulers presented educated society with
a new script, the product of deliberate design rather than spontaneous
evolution.3 When culture shifts, arguably the practices of reading and
writing do as well.
Bearing in mind this potential relationship between general cultural
change and transformations in written culture, the present collection of
essays focuses on the production and use of manuscripts in the Long
Twelfth Century – that is, the period stretching from the late eleventh
through the early thirteenth century – taking the cultural changes that
occurred during the so-called Twelfth-Century Renaissance as its point of
departure.4 This period is significant for manuscript culture, which saw
a substantial increase in the production of books as well as the appearance
of important new physical features. This ‘Renaissance’ also gathers under
one umbrella a number of important and interrelated historical events,
such as monastic reform, the establishment of universities, the birth of
scholasticism, a revival of jurisprudence and the introduction of Greek and
Arabic science and philosophy. The ‘Awakening of Western Europe’, to
1
2 erik kwakkel and rodney thomson
use David Knowles’ term,5 was characterized by a boost in energy and
optimism within educated society, whose members sensed that they were
living in a time different from the immediate past and who contemplated,
often explicitly, their role in the course of history and the new – or at least
changing – present.6
The term ‘renaissance of letters’ is sometimes used to emphasize that this
cultural movement was primarily driven by intellectuals: scholars who read
and wrote texts with renewed passion and interest.7 First there were those in
northern France, England and northern Italy, followed by kindred spirits in
southern Italy (the Kingdom of Sicily), the Germanic lands, the Low
Countries and Christian Spain. These intellectuals – who lacked cohesion
beyond a shared background (a ‘career’, perhaps) in higher education, a deep
yearning for knowledge and the sense that classical ideas ought to be revived
in their lifetime – exchanged ideas through texts and letters which were
disseminated through the main intellectual centres of medieval Europe:
monasteries, cathedral schools and proto-universities. Here the new voices,
presenting new ideas in Latin, the language of eloquence in the West, were
read and heard, contradicted and expanded upon, by a broad range of
intellectuals from St Bernard of Clairvaux and William of Malmesbury to
Peter Abelard and John of Salisbury.
The Long Twelfth Century was a seminal period for the development of
the handmade book in medieval Europe: not only did production increase
dramatically but it was also a high point in material terms such as con-
struction and legible writing, as well as in decoration, usually modest
though sometimes of great splendour and elegance. What Neil Ker said
of English books of the period can be generalized (with appropriate
nuances) to Europe as a whole:
The period is the greatest in the history of English book production.
Manuscripts were perhaps better written in the eighth century and in
the tenth, but they are not numerous. It is no exaggeration to say that
a well-written English twelfth-century manuscript is something we have
a good chance of being able to see in many of our towns. [. . .] In London,
Oxford, and Cambridge there are hundreds of them. They are the
considerable remains of the large number of books produced by the scribes
of this period; accurately copied, competently and often beautifully
written and decorated, well spaced, fully punctuated, and neatly corrected.8
The fundamental texts of both polytheist and Christian antiquity were
disseminated more widely than before, and some indeed were revived for
the first time since the fall of the Roman Empire. Scribes now took greater
care with word separation, punctuation and the articulation of texts. Books
 Introduction 3
catered to new readerships: new styles of monasticism, those with means
but ‘illiterate’ (i.e. not literate in Latin) and the populations of the fluid
educational communities termed collectively ‘schools’ by contemporaries.
Different styles of formatting were developed for the different disciplines
of law, theology and biblical studies. Each of these developments has
received considerable scholarly attention, but they have not been inte-
grated into a ‘big picture’ overview.
Providing such an overview, encompassing the whole of Western
Europe and based upon expert analysis of each subject area, is the aim of
this book. Certain elements of the manuscripts from the Long Twelfth
Century have already received scholarly attention, such as script, decora-
tion, binding and the glossing they frequently contain.9 Furthermore, the
contexts of production and use have been illuminated for some individual
copies,10 regional centres11 or monastic houses.12 However, the manuscript
as a whole and as a developing European book format has not yet received
significant attention, nor has the historical backdrop of its creation as
a pan-European intellectual movement.
The present volume, then, investigates how readers in the twelfth
century interacted with books and texts. It aims to show how a changing
literary taste, a shift in the use of texts and a new outlook on the world
among intellectuals affected the practices of book production and reading
in varying degrees. In an age defined by the introduction of an unusually
high number of new authors (both foreign and home-grown), texts (ori-
ginal Latin works and translations) and genres (natural philosophy, ency-
clopaedias), as well as a new approach to reading and evaluating the written
word – through, for example, the scholastic method – it became important
for readers to own manuscripts that presented texts in formats different
from those they inherited from the Carolingians and their successors.
By gauging the physical features of manuscripts and their dispersal, the
chapters in this book assess, for example, what features were developed for
manuscripts holding different text genres, what the manuscripts of these
genres tell us about how the texts were used and how the different genres
‘sit’ in the book culture of the period. How did their numbers increase or
their physical aspects evolve over the course of the century? To address
these and other textual and ‘bookish’ questions, the essays in this volume
are presented in three sections.
The first section, Book Production, is devoted to the production of
manuscripts. The four chapters in this section discuss how twelfth-century
manuscripts were produced and who was involved in their production.
The focus is on the objects’ main material aspects, from the execution of
4 erik kwakkel and rodney thomson
script (Chapter 1) and the physical construction (Chapter 2), to techniques
of decoration (Chapter 3). These chapters attempt to emphasize how the
twelfth-century book reflected the shifting interests of readers: they show
how the objects were, to a large extent, customized for use in a new age,
while maintaining certain features inherited from earlier centuries.
The final discussion in this first section (Chapter 4) focuses on the
individuals or teams who made these manuscripts that survive in their
thousands, as well as the locations in which they produced, stored and used
their books: shelved cupboards (armaria) in the monastery, perhaps even
a ‘scriptorium’ that was a physical entity.
The second section, Readers and Their Books, assesses the ways in which
readers interacted with texts and the physical books that contained them.
The essays in this segment focus on the owners of manuscripts, both
individual scholars (Chapter 5) and institutions, most importantly mon-
asteries (Chapter 6). Other chapters in this section deal with the consump-
tion of texts, highlighting new modes of reading (Chapter 7) and new
practices in annotating books (Chapter 8). These chapters show how
different groups of readers practised different ways of reading and inter-
acting with texts (monks, scholars and students), while also highlighting
the networks through which texts were disseminated and how the texts
were made available to readers, for example in libraries.
The main thread running through the third section, Types of Books, is
the contents of manuscripts produced in the Long Twelfth Century. After
an assessment of Hebrew manuscripts in the West (Chapter 9), the focus
shifts to the seven main text genres that were popular in the period: Liturgy
(Chapter 10), Theology (Chapter 11), Philosophy (Chapter 12), the Classics
(Chapter 13), the Sciences (Chapter 14), Medicine (Chapter 15) and Law
(Chapter 16). The final two chapters deal with a new genre of texts, those
written in or translated into the vernacular languages of England and
France (Chapter 17), and Germany (Chapter 18). The chapters in this
third section enquire how these genres advanced in the century and a
half between 1075 and 1225, how manuscripts accommodated genre-specific
elements such as diagrams, glosses, complex thoughts or debates, or how
information not previously available in Europe (such as that found in
scientific texts) was presented to the reader. Ultimately, the authors aim to
place the contents of manuscripts within their historical context alongside
evidence of use and observations related to the book as a physical object.
We might have included a chapter on the legacy of the twelfth century.
Some of the types of book mentioned earlier in this introduction, and some
of the characteristics of their construction, formatting and decoration
 Introduction 5
continued to appear through later centuries, and indeed still influence
book culture today. Other features died quickly. In the course of the
thirteenth century Benedictine monasteries largely ceased to produce
their own books, and the grand monastic books that characterize the first
half of the twelfth century gave way to products of commercial workshops
and extra-monastic milieus such as university towns. Speed and expense
were now significant factors influencing book production, aimed at new
readers such as students and friars. Splendid luxury books continued to be
made in small numbers up until the era of print. Some specific types of
book disappeared altogether: glossed books of the Bible, produced in great
numbers throughout the second half of the twelfth century, were no longer
made after the late thirteenth century.13
Some of this process may be reasonably described as ‘decline’ or at least
as deliberate change, whether due to fashion (especially as applied to
decoration) or economic factors. But the legacy of the twelfth century is
greater than that appears at first sight – and this is often forgotten – simply
because of the lasting qualities of the books made then. Glossed books
ceased to be made because the existing stock was sufficient and would
remain usable over the centuries.14 When Europe revived again in the wake
of the Black Death, and another ‘renaissance’ fostered renewed interest in
the classics and the Fathers, twelfth-century volumes were reread.
The marginal annotation and ownership inscriptions of fifteenth-century
humanists demonstrate this to have been so. An inadvertent compliment
was paid to the earlier age through the development of ‘humanist’ script
and decoration, which was influenced by exemplars in twelfth-century
script.15 When printers started to use roman type, which was developed
from this humanist script, the artisans gave, indirectly, new life to the
venerated Pregothic script. This renewed, now mechanized script appeared
on pages that already featured other traits developed during the Long
Twelfth Century, such as running titles bearing the name of the text,
foliation and the relative proportions of the page itself. In many ways the
legacy of the handmade books studied in this volume endures in printed
books today.

Notes
1. Hall 2004; Roberts and Skeat 1983.
2. Searching through pages of a physical book is still today easier than ‘scrolling’
through an electronic text.
3. Ganz 1995.
6 erik kwakkel and rodney thomson
4. Benson and Constable, Renaissance; Haskins, Renaissance; Swanson 1999.
5. Knowles 1962, 79.
6. Abulafia 2006; Jaeger 1994.
7. Damian-Grint 1999; Luscombe 2004; Verger 1999.
8. Ker, English MSS, 1.
9. For example, Derolez, Gothic; Cahn, Romanesque MSS; Sheppard 1995; De
Hamel, Glossed Books, respectively.
10. For example, Donovan 1993; Eadwine; Gullick 1990.
11. For example, Kauffmann 1975; Ker, English MSS; Thomson 1998, 2006.
12. For example, Palmer 1998; Thomson, St Albans.
13. De Hamel, Glossed Books, p. xiii; more nuanced, L. Smith, The Glossa
Ordinaria (Leiden/Boston, 2009), 181–7.
14. The most convenient evidence for this is the editions of many late medieval
booklists in CBMLC, which include identifications of surviving books.
15. Ullman, for example, argues that late fourteenth-century humanists preferred
manuscripts written in large twelfth-century script, especially Italian: Ullman
1960, 12, 14 n. 10.
 part i
Book Production
 chapter 1

Codicology
Erik Kwakkel and Rodney Thomson

Parchment
The pages of all books made in Western Europe between the eighth and
early thirteenth centuries were made of ‘membrane’, that is, animal skin
prepared in a particular way, to produce what is commonly known today as
parchment or vellum.1 Paper made from rags was used in Muslim Spain2
and its existence was known to Western Europeans, who despised it as an
inferior material; this probably explains why it was not made or widely
used there before the late thirteenth century. In his polemic against the
Jews, completed in or after 1143, Peter the Venerable, abbot of Cluny,
asked:
God, you say, reads the book of Talmud in Heaven. But what kind of a
book? Is it the kind we have in daily use, made from the skins of rams, goats,
or calves? Or from reeds and rushes out of eastern swamps, or from old rags,
or from some other even more vile material?3
Peter had strong connections with Spain, with Toledo in particular, and
his reference to papyrus may be to contemporary usage, as his mention of
rags certainly is. Paper, then, is not part of our story, except as exemplifying
the need for cheaper books that increased throughout the next centuries as
a result of the expansion of the universities and growth in lay readership.
The earliest surviving document from Western Europe made of paper is
dated 1216–22; the earliest datable paper books come from the late thir-
teenth century.4
Peter’s reference to the types of skin used for book-making in the Europe
of his time is interesting; it is no surprise to find sheep and calf mentioned,
and goat is also known to have been used, though to an extent that has
never been quantified. In fact the treatment of skins to produce high-grade
parchment obliterates most traces of the species of animal from which it
came.5 Goatskin can sometimes be distinguished by its tawny colour and
prominent hair follicles, and the evidence of the surviving books is that it

9
10 erik kwakkel and rodney thomson
was more commonly used in Italy than elsewhere. What we know nothing
of is how the flocks and herds were managed to enable the manufacture of
so many books so quickly. Clearly there must have been a connection
between the provision of skins and of meat, with meat having the priority.
It would seem to follow that the skins of nearly all animals slaughtered for
their meat must have been used to make books. Meat was needed on a
regular basis for consumption by high-status laypersons; it was not con-
sumed by the religious or the peasantry. Parchment, on the other hand, was
needed, mainly by religious communities, in batches big enough to make
at least one whole book in a single operation.6 A new or reformed monastic
house might stock its library with, say, 150 volumes over a period of thirty
years. Assuming an average of about as many leaves as this per volume, this
project might need about 5,500 skins, involving the slaughter of about 183
animals a year.
The elaborate process by which skin was made into parchment suitable for
writing on has been described many times, recently with excellent
photographs and line-drawings, and so need not be repeated here.7 What
is less well recognized is the generally high quality of twelfth-century
parchment, especially that employed to make monastic books. It is
off-white but not yellow, comparatively thick, each side slightly furry, holds
the ink well, supple yet provides a flat writing-surface, not easy to crease. By
contrast the parchment of books from the late thirteenth until the late
fourteenth century is often yellow or even brownish, thin, smooth but easily
creased or crinkled. The fifteenth century saw a return to twelfth-century
quality. One imagines that the decline in quality was due to the omission of
some of the stages of the production process, for reasons of expense. The
background to the making of this excellent parchment probably lies back in
the eighth century and in the British Isles.8
Clients wishing to produce a grand book might conduct a search for the
best-quality, blemish-free parchment, which was not necessarily available
locally. A famous case is the Bury Bible (Figure 3.3), made at and for Bury-
St Edmunds Abbey ca. 1130, for which the parchment was sought ‘in
partibus Scotie’, which by the thirteenth century (when this fact was
recorded) ought to mean Scotland rather than Ireland.9 The difference
matters little: as Brown pointed out, the significance of the monks’ quest is
that the Celtic areas of the British Isles were known to be places where the
ancient techniques for making high-quality parchment were still intact. In
the Bury Bible, and in other grand books, high-quality parchment was
pasted onto the existing pages to take decoration in saturated colours that
 Codicology 11
would otherwise show through as unsightly dark blotches on the other side
of the leaf.10
The process of turning skin into parchment was a highly skilled one: what
sort of people did it? The famous full-page miniature in Bamberg,
Staatsbibl., Patr. 5, f. 1v, suggests that it might sometimes have been done
by the members of a monastic community. The miniature is bordered by a
series of roundels showing the process of book-making from preparation of
the parchment on, all the persons engaged being tonsured and in monastic
robes. But there is far more evidence for the existence of commercial
parchmenters and for their supplying religious houses in their area with
skins at various stages of preparation. The English evidence, which is
particularly abundant, can probably stand for all. Here follows a sample
with no pretensions to completeness. In north Lincolnshire ca. 1170
‘Gilebertus perkamenarius’ witnessed a grant of property to the
Premonstratensian abbey of Newhouse.11 The Pipe Roll for 1176–7 records
a substantial debt to the Exchequer of one mark owed by ‘Gille parchemi-
narius’, probably from Shrewsbury.12 Late in the century a customary note in
the Register of Winchcombe Abbey states that ‘When the parchmenters . . .
come to the [monks’] parlour with their parchment on the first day of
Advent, they should receive the caritas of the house, namely bread and
beer from the abbot’s cellar, at the hand of the gatekeeper.’13 This is
particularly illuminating: it tells us that even a small town such as
Winchcombe could be home to more than one parchmenter, and that
they regularly came to the abbey parlour (where the monks were permitted
conversation) to sell their wares. About 1205 ‘Hernaldus parchiminarius’
figures as a tenant of Worcester Cathedral Priory.14 A document of the early
thirteenth century from Hereford Cathedral was witnessed by a Theobald
‘parcamenarius’.15 In the city of Lincoln ca. 1157 there was even a ‘uicus
pargamenariorum’,16 suggestive of something like a local industry.
Documentary evidence indicates that the parchmenter could deliver the
parchment in untrimmed sheets, usually by the dozen, or else trimmed
and folded. Pricking and ruling were part of the next stage of preparation,
executed by the scribe, but first he had to decide what dimensions he would
give the book.

Dimensions
Significant changes observed in other domains of twelfth-century book
production, for example in script and decoration,17 prompt the question of
whether page dimensions likewise show developments over time. Dated
12 erik kwakkel and rodney thomson
and datable manuscripts provide some insights in this respect: from the
period between 1075 and 1224, 353 surviving manuscripts can be drafted
into service (Figure 1.1).18 To create categories that are represented by a fair
number of manuscripts, the spectrum is broken up into segments that
increase with ten-centimetre increments. The discussion is limited, for
now, to the height of the page, considering that the width usually follows
automatically from it: quantitative studies show that after the tenth cen-
tury the width of the page equals 70 per cent of its height.19 This, of course,
is in part dictated by the shape of the skin.
Figure 1.1 shows, first of all, that there is limited change over time. It is
striking, for example, that at no point does one line of development
overtake another: their relative order remains the same. Moreover, none
of the developments shows a sharp upward or downward turn. Only
manuscripts with a height of between 151 mm and 200 mm are ultimately
represented in somewhat different numbers: 38 per cent in 1200–24 com-
pared to 23 per cent in 1075–99 – an increase. In addition, Figure 1.1
suggests that very large manuscripts of more than 450 mm in page height
may have somewhat decreased in popularity: from 12 per cent in the late
eleventh century to 2 per cent in the early thirteenth. This decrease may be

100%

90%

80%

70%

60%

50%

40%

30%

20%

10%

0%
1075–99 1100–24 1125–49 1150–74 1175–99 1200–24
<150 0% 0% 0% 0% 0% 2%
151–250 23% 42% 35% 24% 33% 38%
251–350 50% 50% 47% 49% 38% 52%
351–450 15% 5% 13% 14% 19% 6%
>450 12% 3% 5% 12% 11% 2%

Figure 1.1 Page heights of 353 dated and datable manuscripts, 1075–1224.
 Codicology 13
the result of the fact that outsize manuscripts (such as gospel books
and bibles) had a longer working life and may not have often needed
recopying. Most manuscripts in the sample can be placed in the categories
151–250 mm and 251–350 mm. The latter is the most popular size range
throughout the period: it consistently represents around 50 per cent of the
corpus, except for a small dip in the late twelfth century.
The smallest manuscripts are a mixed bag. Specimens up to 160 mm in
height (an arbitrary break-off point) include a breviary for Cistercian use
(Bodl. Libr. Lat. liturg. f. 1, 1219, 128 mm), a calendar (Cambridge,
Fitzwilliam Museum 24, 1204, 153 mm), a poem in praise of the city of
Chester (Bodl. Libr. Bodley 672, 1194, 154 mm), Bernardus Silvestris’
Cosmographia (BnF lat. 2770, 1198–1205, 155 mm), a New Testament
(Cambridge, St John’s Coll. G. 15 [183], 1167–83, 158 mm) and
Cassiodorus’ Epistolae (Cambridge, Trinity Coll. O. 7. 13 [1341], 1167–83,
159 mm). In spite of the limited size of these codices, scribes wrote a relatively
large number of lines on their pages. To do so, they opted for a notably small
script size: the line height of the script is, in increasing order: 3.3 mm
(Fitzwilliam 24), 3.6 mm (Trinity O. 7. 13), 3.8 mm (St John’s Coll. G. 15
as well as Bodl. Libr. Lat. liturg. f. 1), 4.3 mm (BnF lat. 2770) and 4.7 mm
(Bodl. 672).20 Considering that scribes aimed to maximize the amount of
information contained in these small books, it is fair to conclude that they
were probably designed for portability.
At the high end of the spectrum we encounter very different manu-
scripts. The tallest category of books, those of 450 mm or taller, encom-
passes a high number of bibles (twelve of the twenty-eight manuscripts in
this category), including many so-called Giant Bibles: the Stavelot Bible
(BL Add. 28106 and 28107, 1094–7, 575 mm) being the largest manuscript
in the corpus, the Arnstein Bible (BL Harley 2798 and 2799, 1172,
545 mm), the Worms Bible (BL Harl. 2803 and 2804, 527 mm), the
Bury Bible (CCCC 2, 1121–38, 520 mm) and the Floreffe Bible (BL
Add. 17737 and 17738, ca. 1155, 475 mm).21 Other manuscripts in this
category are commentaries, such as on the Epistles of St Paul (BnF lat.
11575 and 11576, 1164, 477 mm), Lombard on the Psalms (Bodl. Libr.
Auct. E. inf. 6, 1173–6, 467 mm) and Jerome on the Minor Prophets (BnF
lat. 1835, ca. 1202, 455 mm). There are also two manuscripts of canon law
more than 450 mm high: BnF lat. 3853 (ca. 1154, 510 mm) and BnF lat.
14314 (ca. 1152, 490 mm). The largest manuscripts in our corpus were
written in much larger script: the line height varies between 7 mm and
10 mm, which is often twice the height as in the small manuscripts
discussed before.
14 erik kwakkel and rodney thomson
A comparison of the dimensions of these books with the kind of works
they contain reveals significant differences between text genres. In descend-
ing order of their height, the averages for the major genres are as follows:22

Biblical (31 mss): 395 mm

Commentaries (32 mss): 345 mm
Patristic (32 mss): 305 mm
Historical (89 mss): 289 mm
Theology (15 mss): 279 mm
Liturgical/Devotional (77 mss): 275 mm
Administrative (17 mss): 272 mm
Hagiography (19 mss): 268 mm

While these numbers are merely averages, they do hint at a correlation

between a manuscript’s dimensions and the text it holds. As expected, bibles
and commentaries (including Glossed Bibles) top the list, as was already
suggested by the examples provided earlier in this chapter. In contrast, genres
that were generally written in manuscripts of smaller dimensions appear to
be administrative in nature (cartularies, in many cases) or hagiographical
(mostly Lives of individual saints, and some passionals), closely followed by
liturgical and devotional books, and those filled with theological works.
The rationale behind the chosen height for a manuscript is probably
related, at least in large part, to the use for which the object was intended.
For example, Giant Bibles, a major component of the ‘tallest’ text genre,
were commonly used for reading out loud in a monastery’s refectory. The
large size minimized the frequency with which pages needed to be turned,
while it also accommodated large script. A fair share of liturgical and
devotional manuscripts was made for handheld use, such as breviaries,
missals and psalters, which may explain their modest size. The same
rationale – handheld use – lies at the heart of a type of manuscript that
breaks with the common norm of relative dimensions (which stipulates that
the width of the page is about 70 per cent of its height, as discussed). The
corpus contains a number of ‘holster books’, which feature a particularly tall
page with a width that equals 60 per cent or less of the page’s height. This
narrow format was helpful for a reader who held up the book to read or sing
from. It was thus favoured by teachers in the monastic classroom (a striking
number of classical manuscripts from the eleventh and twelfth centuries are
in holster format) and soloists in the Mass. Notably, nearly all tropers and
cantatoria made before ca. 1200 are in holster format.23
 Codicology 15
Layout
Layout encompasses a variety of codicological features, such as columns
(their number and width, and how many lines they contain); the use of
headings, paragraphs, running titles and other aids to the reader; the size of
the margins; and the inclusion of interlinear and marginal glosses. While
the application of these and other design features varied greatly, the
majority of twelfth-century manuscripts do not have a particularly com-
plicated layout. Certain locations in the book, however, demanded special
attention. Capitula lists were often written in a smaller version of the script,
or even in a different script entirely (usually Pregothic cursive), which
impacted the line spacing and sometimes also pricking and ruling pat-
terns.24 Moreover, in single-column manuscripts the capitula lists may be
presented in two columns, perhaps as a consequence of the smaller script
that was used, which also demanded special attention during page design.
Similarly, liturgical manuscripts included not only different script sizes but
also musical notation: this, too, complicated page design, especially if the
music was presented on regular text pages. The insertion of decoration and
illustrations, finally, also presented reasons for pause. Larger initials were
not yet placed in generic square- or rectangular-shaped spaces, as in the
thirteenth century; instead the letters were ‘pushed’ into the textblock, as it
were. The scribe had to be mindful of their shape when he copied the text
(writing preceded decoration): he had to leave both the correct amount and
the right shape of empty space.25
Two important considerations for the ultimate design of the page were
the contents of the manuscript and how it would be used. The impact of
these is shown by twelfth-century commentaries, in particular the Glossa
ordinaria, a key text of the period, which was principally compiled by
Anselm of Laon (d. 1117) with the help of his brother Ralph.26 The Gloss is
encountered in two characteristic layouts.27 The oldest was ‘simple’ in that
it presented a central text column with a fixed width. The commentary was
placed around the text in large margins: usually one column on each side.
The main text and the marginal commentary form a single entity that takes
up much of the page, leaving little room for additional notes. Ruling in the
commentary columns was only executed when glosses were added (termed
‘ruling on demand’ by Peter Gumbert). A more refined design appeared in
the 1160s in Paris. It presented greater flexibility – and new design chal-
lenges. The most distinct development is the variable width of the text
column, which now depended on the amount of commentary. Generally
the page still presented three columns – centre column for text, two
16 erik kwakkel and rodney thomson
flanking columns for commentary – but this division could be adjusted:
text or commentary could switch to a presentation in two columns, if need
be, even on the same page. While more challenging to execute, also
considering that text and gloss were pricked and ruled independently,28
the new presentation saved space.
Another correlation between text and page design is encountered in
classical texts copied for classroom use. Throughout the Long Twelfth
Century such copies were frequently presented in holster format.29 Holster
books with classical texts, which are nearly always copied in a single
column, were commonly used handheld during the teaching of grammar,
rhetoric and dialectic. The scribe, usually the teacher himself, planned
plenty of empty space on the page, which was then filled with notes, short
explanations or synonyms.30 A peculiar design feature of these manuscripts
is that the text column is usually not placed in a central position on the
page, which would have allowed readers to add notes on both sides of the
text – as in Glossa ordinaria manuscripts. Although there are exceptions,
the single text column is commonly positioned close to the inner margin,
so that the extra space planned for annotations all ended up on the outer
side of the page (the right side on a recto).
There were also external factors at play with respect to page design. The
reader probably had a voice in the ultimate choice of codicological features,
including those pertaining to page layout. Near the end of the twelfth
century, an increasing number of manuscripts was produced commer-
cially.31 Like their peers in monastic houses, commercial scribes knew
how to shape the page when copying a certain text, but patrons may still
have steered the design as well.32 The milieu of the scribe, lastly, may also
have been a factor of influence. Scribes who were members of religious
houses produced books according to common practices in their order, for
example with respect to abbreviations and decoration, as seen in the
Cistercian Order.33 It is not unlikely that page layout was also governed
by order-specific practices, although such research remains yet to be
undertaken.
Then there is, finally, the matter of change and development, which has
emerged at several occasions in this chapter. As discussed, a shift to a more
flexible layout is observed in the Glossa ordinaria. Moreover, over the
course of the twelfth century the margins of manuscripts began to include
elements that changed the presentation of the text, such as running titles
displaying the full title, marginal enumerations that reflected the structure
of the argument and the alternation of blue and red paragraphs, which all
helped the reader. It is hard to pin down when these new features of layout
 Codicology 17
emerged, even with dated manuscripts, but they appear to be standard by
the early thirteenth century.34 However, core codicological features in page
design, such as the size of the margins and the number of text columns,
remained remarkably stable throughout the period. Observing the first of
these (again through the lens of dated and datable manuscripts), an almost
even balance is seen in the number of columns on the page: at the end of
the century the numbers are not so very different than at the outset (Figure
1.2).35 Taking the numbers at face value would suggest that a single-column
layout remained in favour throughout the period.36
Similarly, dated and datable manuscripts suggest that the ratio of text to
marginal space did not evolve much either. Throughout the century and a
half between 1075 and 1224 the marginal space hovered between 47 per cent
and 50 per cent and shows no consistent increase or decrease.37
The margins did change in one respect: they started to accommodate
glosses in a more efficient manner. When producing manuscripts that were
anticipated to take on a considerable number of annotations during use,
scribes placed extra horizontal and vertical ruling in the blank space
surrounding the main text columns. A grid of square and rectangular
‘comment boxes’ thus appeared, as many as thirty or more on a single

100%

90%

80%

70%

60%

50%

40%

30%

20%

10%

0%
1075–99 1100–24 1125–49 1150–74 1175–99 1200–24
1 Col 58% 66% 55% 44% 47% 48%
2 Cols 38% 26% 43% 46% 42% 40%
Variable 4% 5% 3% 8% 11% 12%
2+ Cols 0% 5% 0% 2% 0% 0%

Figure 1.2 Number of columns in 353 dated and datable manuscripts, 1075–1224.
18 erik kwakkel and rodney thomson
page, which helped the user deal in an efficient manner with annotations.
Because the grid consisted of three to five or more vertical glossing
‘columns’, clearly outlined in plummet, the reader was also given the
chance to annotate the same line multiple times while positioning all
remarks at the precise height of the line they pertain to. This system
appeared on the grid late in the second half of the twelfth century and
was finally fully embraced in the thirteenth-century university.38

Binding
The last stage in the process of making a book was binding. This did not
necessarily happen immediately; a book might be kept in quires tacketed
together with or without a limp wrapper for a long time.39 Some books
never passed beyond this stage.40 Nonetheless most books were bound,
especially in cathedrals and monasteries, and the so-called romanesque
bindings are distinctive.
It is only recently that a small group of scholars has realized, and begun
to publicize, the fact that in the twelfth century European bookbinding
reached a pinnacle of excellence which it would arguably never reach
again.41 That this perception has been long in coming is due primarily to
the small number of surviving medieval bindings, aggravated by further
destruction even in relatively recent times. It is probably also due to the fact
that modern binding scholarship has concentrated on the more beautiful
and elaborate items, eschewing the majority of medieval bindings, which
were plain and utilitarian. Until very recently, descriptive catalogues of
medieval manuscripts paid scant attention to their bindings and, if they
described them at all, did so perfunctorily. For instance, M. R. James
regularly described surviving medieval bindings in something like the
following terms: ‘Old binding, (sheep/white) skin over wooden boards,
clasps gone.’42 He never attempted to date these bindings. Such a negative
impression did his perfunctory descriptions make that librarians often had
the books so described rebound soon after the publication of James’
catalogues.
The standard twelfth-century binding was of tawed skin over wooden
boards (Figure 1.3). The skin was applied as a single sheet, pasted to the
exterior of the boards, but not to the spine, folded round the edge of the
boards to form ‘turn-ins’ which were also pasted, and sometimes mitred
and sewn. At each end of the spine the sheet was sometimes extended out
in a semicircle, into which was sewn a lining also of skin, the other end of
which was sewn into the spinefolds of the quires. The purpose of these
 Codicology 19

Figure 1.3 A representative group of twelfth-century bindings from Hereford

Cathedral (Mynors and Thomson, Hereford, pl. 21).

so-called tabs is not altogether clear. It has been suggested that they were
used to pull out a book stored in a chest foredge-down, but this is
guesswork, unprovable in the absence of any illustrations or other
physical evidence of such a process. The ‘tabs’ were often shorn off at a
later date, when books were stored on desk-lecterns or, later still, upright
in presses. The boards were thicker than was usual in earlier times, of
quarter-sewn oak in England and sometimes on the Continent, though
beech was more usual in Germany. An important innovation, the sewing
frame, seems to have first been devised early in the century. It is
illustrated in one of the roundels of the famous miniature in Bamberg,
Staatsbibl. Patr. 5, f. 1v, of ca. 1150. These roundels suggest that, as in the
case of the parchment, binding was done in-house, and this impression is
reinforced by documentary evidence such as the Liber ordinis of
St-Victor, which makes binding of books one of the precentor’s responsibil-
ities.43 It was certainly the case that binding was carried out in the precentory
at Worcester Cathedral Priory, though the documentary evidence is no earlier
than the fourteenth century.44
Few complete bindings have survived from before ca. 1100; the earliest is
on the famous St Cuthbert (Stonyhurst) Gospels (BL Add. 89000), made
20 erik kwakkel and rodney thomson
at Wearmouth-Jarrow in the early eighth century,45 followed by a large
group of Carolingian bindings from the ninth century.46 A very few
bindings that might be pre-Conquest survive on English books.47
So far this section has described and discussed what seem to have been
the most common forms of twelfth-century binding. But there were other
types, rare now but perhaps more common than we think: limp bindings
or wrappers, blind-stamped and treasure bindings.48
The quires of a finished book intended to be used but not given a
stiff-board binding immediately had to be kept together somehow. This
could be achieved, even for books with no binding at all, by means of
quire-tacketing49 or the application of a limp wrapper. Tacketing was
achieved by piercing slits or holes, singly or in pairs, through the spinefold
of each quire, near the head and foot. Through these ‘stations’ was passed a
length of thread or thin strip of twisted parchment. These ‘tackets’ were
knotted or twisted together on the spinefold exterior. The purpose of the
procedure was to keep together the leaves of an individual quire, not to join
the quires to each other. Durham Cath. A. IV. 34 is a famous example of a
twelfth-century book that has never been bound, with the remains of
tackets still in place.
Limp bindings, in occasional use since at least the eighth century and
well beyond the twelfth, have not survived well and are hard to date.50
Worcester Cath. Q. 44 (ca. 1200) is in an undamaged wrapper of thick,
tawed skin lined with fabric.51
At least 139 books in blind-stamped bindings survive from the twelfth
century,52 after which, for unknown reasons, there is a long hiatus until the
late fifteenth century, after which they become common. They look quite
different from the standard tawed bindings described earlier in this chap-
ter. The wooden boards were covered with ‘thinner, shiny tanned leather,
usually red-brown, impressed all over with rows of very small stamps, of
many different designs on a single binding, arranged in patterns of grids,
circles and other geometric shapes’.53 About ninety are so similar to each
other that they have been thought to come from the same place, assumed
by Hobson and confirmed by De Hamel, to be Paris. Sixty-three of the
ninety cover glossed books of the Bible, the bindings being more or less
contemporary with the books they covered. Here, then, is an instance of
binding being co-ordinated with the manufacture of the unbound book.
But Paris was not the only venue for the production of such bindings.
Three blind-stamped romanesque bindings were made at Winchester,
commercially no doubt but for the Cathedral Priory there, and the con-
temporary binding of the late twelfth-century Puiset Bible was made at
 Codicology 21
Durham. Other such bindings have been associated with London. All seem
to be datable to the second half of the century.
Treasure bindings were applied almost exclusively to great liturgical
books used in the Church, gospel books and the like, generally kept
securely in the treasury rather than with the library books. They were
made in some abundance between late antiquity and the eleventh century,
so that probably not many new ones were needed in the twelfth century,
except for newly made books. They consisted of wooden boards, attached
to the main book in the same way as any other binding. But instead of a
skin cover, to the boards were nailed sheets or strips of metal, on which
were mounted files of gemstones en cabochon, enamels or antique cameos,
usually flanking a centrepiece of gold or ivory depicting a scene such as
Christ in Majesty. Always a very small proportion of all bindings, their
survival rate, for obvious reasons, is appallingly low.54 From England, not a
single medieval treasure binding survives intact, thanks to the depredations
of Henry VIII’s officials. Glimpses of what has been lost are provided by
church inventories. For instance, a mid-twelfth-century inventory from
the abbey of Ely lists seventeen gospel books, all in treasure bindings which
are described in detail; at least six, perhaps as many as nine of these were
made in the tenth and eleventh centuries; only one was definitely made in
the twelfth.55 Another such inventory, drawn up in 1295 at St Paul’s,
London, lists eleven gospel books in treasure bindings, none datable except
for three which were given in the late twelfth century by the evidently
wealthy Henry of Northampton, canon before 1174 and probably dying on
2 April 1192.56 The remains of such bindings, the boards only without the
ornament, survive on two twelfth-century books from Britain, the
Sherborne ‘Cartulary’ (really a gospel book) and the Book of Llandaff
(a register of documents).57 The extreme discrepancy between the figures
supplied by medieval inventories and what survives today holds good for
France, but less so for the regions of the old German Empire. From
northern Italy to the Netherlands, but especially within modern
Germany, dozens of such bindings survive, most of them modified in
the later Middle Ages or badly damaged; on the other hand, many
essentially twelfth-century treasure bindings cannibalized much older
components such as ivory plaques, enamels and cameos. It appears that
from the eleventh-century Liège and the Mosan region, then from the mid-
twelfth-century Limoges, were centres where such bindings, or their metallic
elements, might have been made commercially.
22 erik kwakkel and rodney thomson
Notes
The sections of this chapter on parchment and binding are contributed by
R. Thomson, and those on dimensions and layout by E. Kwakkel.
1. Gullick 1991, 145, on terminology. At least in the Anglophone world the
current preference is to use the term ‘parchment’ to describe indifferently
writing material prepared from the skin of either sheep or calves.
2. See Chapter 14.
3. ‘Legit, inquis, Deus in caelis librum Thelmuth. Sed cuiusmodi librum? Si
talem quales alios cotidie in usu legendi habemus, utique ex pellibus
arietum, hircorum uel uitulorum, siue ex biblis uel iuncis orientalium
paludum, aut ex rasuris ueterum pannorum seu ex alia qualibet forte uiliori
materia compactos.’ Peter the Venerable, Adversus Iudeorum Inveteratam
Duritiem, ed. Y. Friedman, CCCM 58 (1985), 130 ll. 191–7. The earliest
surviving Western paper manuscript may be Tours, Bibl. mun. 927, ff. 1–46,
dated late twelfth or early thirteenth century. The earliest datable manu-
script is Padua, Bibl. Antoniana 550, from 1287.
4. The classic account of paper manufacture is Dard Hunter, esp. 170–202; also
CHBB II. 48 n. 49; Chapter 14 of the present volume; Kwakkel 2003, at
220–7.
5. MMBL 1, ix n. 3; Gullick 1991, 145 n. 1.
6. The Account-Book of Beaulieu Abbey, ed. S. F. Hockey (RHS Camden 4th ser.
16 [1975], 195–8); Gullick 1991, 147–8, 153.
7. Thompson 1936, 24–30; De Hamel 1992, 8–16; Clemens and Graham, 2007,
9–13.
8. Brown 1993.
9. Cf. Brown 1993, 126, who translates ‘Scotia’ as Ireland.
10. Thomson, Bury Bible, 6–7, 26.
11. F. M. Stenton, Documents Illustrative of the Social and Economic History of the
Danelaw (London, 1920), 198.
12. Pipe Roll Soc. 26 (1905), 39.
13. ‘Pargamenarii . . . cum ad locutorium cum pargameno uenerint prima die
aduentus sui caritatem domus scilicet panem et ceruisiam de cellario abbatis,
per manum ianitoris habere debent’: Landboc sive Registrum monasterii . . . de
Winchelcumba . . ., ed. D. Royce (2 vols., Exeter, 1892), 1. 56.
14. The Cartulary of Worcester Cathedral Priory (Pipe Roll Soc., n.s. 38 [1968],
no. 341).
15. Hereford Cathedral, Dean and Chapter 990.
16. A. Saltman, Theobald Archbishop of Canterbury (London, 1956), 382,
charter 157.
17. See Chapters 2 and 3 of this volume.
18. The 353 manuscripts at the basis of this graph are listed in Kwakkel 2012, 112–
25 (Table 3); conventions for inclusion at 112. Number of manuscripts per
quarter-century: twenty-six (1075–99), thirty-eight (1100–24), seventy-five
 Codicology 23
(1125–49), ninety-eight (1150–74), sixty-four (1175–99) and fifty-two
(1200–24).
19. Bozzolo and Ornato, 1983, 287–310, esp. at 287.
20. For this calculation, I divided the height of the textblock by the number of
lines. I was unable to calculate this number for Bodl. Libr. Rawl. Q. f. 8.
21. De Hamel 2001, 64–91; Ayres 1994.
22. The list totals 312 manuscripts. I have removed the following genres from the
enumeration because of the small number of manuscripts they contain:
sermons (twelve MSS, 319 mm), classical texts (nine MSS, average
249 mm), law manuscripts (five MSS, 365 mm) and encyclopaedic works
(four MSS, 338 mm). The remaining eleven manuscripts are of miscellaneous
contents.
23. Robinson 2008, 54 (type of book), Kwakkel 2012a (use), Huglo 2001, 89–104,
esp. Tables 3.1a at 96, 3.1b at 97 and 3.2 at 99 (cantatoria and tropers).
24. See Chapter 2 of the present volume.
25. See Chapter 3 of the present volume.
26. Smith 2009, 17–38.
27. Smith 2009, 92–105 and 105–9, respectively. For layout of the Gloss, see also
Gumbert 1999.
28. See also De Hamel, Glossed Books, 23–7.
29. Kwakkel 2012a. Examples are the three base manuscripts in Reynolds 1996
(all actual teaching copies): BnF lat. 8216 (1150–1200); Cambridge, Peterhouse
229 (ca. 1150); BL Harl. 3534 (1100–25).
30. Reynolds 1996a, 103–5.
31. Rouse and Rouse 2000, 17–50 (concerning Paris, 1175–1225).
32. Kwakkel 2015, 65–70.
33. Palmer 2010 and Reilly 2012, 129, respectively.
34. Parkes 1991, and Rouse and Rouse 1991. The latter, at 207: ‘One cannot give a
precise terminus ante quem for general acceptance of the individual elements,
save to say that by about 1220 they were all standard; most can be seen on the
pages of any late twelfth-century glossed Bible or manuscript of the Sentences.’
35. ‘Variable’ in this graph represents manuscripts that mix the number of
columns, mostly showing both one and two columns (eight MSS) or two
and three (four MSS).
36. This contradicts earlier assessments, for example in Parkes 2008, 55: ‘During
the twelfth century scribes preferred two-column layouts for “library” copies
of texts, especially patristic works.’
37. Counted are columns with main text only, not those filled with glosses. The
breakdown: 49 per cent (1075–99), 47 per cent (1100–24), 47 per cent, (1125–
49), 49 per cent (1150–74), 47 per cent (1175–99) and 50 per cent (1200–24).
38. Examples: BL Royal 10 C. iv (Gratian, Decretum, 1198–1202), Royal 4 D. vii
(Comestor, Historia scholastica, 1195–1214).
39. Gumbert 2011.
40. Hobson 1929, 56. A famous example is Durham Cath. A. IV. 34: Mynors,
Durham, 57, no. 74; Wormald and Wright 1958, 38–9 and pl. 7; Doyle 1972,
24 erik kwakkel and rodney thomson
35–47. Michael Gullick informs me that parchment tackets remain in place in
five of the book’s eight quires.
41. Szirmai 1999, ch. 8; Gullick in Mynors and Thomson, Hereford, xxvi–xxxii;
Gullick in Thomson, Worcester, xl, xliii–xliv; Gullick 2008, 95–103 and fig. 1;
Clarkson 2013.
42. For instance, in James 1907–14, describing Cambridge, Gonv. and Caius Coll.
MSS 2/2, 3/3, 6/6, 7/7, 10/10, 12/128, 14/130, 15/131, 16/132, 17/133, 18/134, 19/135
&c., all replaced within a few years of the publication of James’ catalogue.
43. Gullick 1996, 249–50.
44. Thomson, Worcester, xxxi, xlvi–xlvii.
45. Powell and Waters 1969.
46. Szirmai, ch. 7, listing 230-odd surviving bindings.
47. Pollard 1975.
48. De Hamel, Glossed Books, ch. 6.
49. Gullick 1996, Gumbert 2011.
50. Scholla 2002.
51. Thomson, Worcester, xlv, 143 and pl. 5(a–b). Another limp binding with fabric
lining is The Hague, Koninklijke Bibl. 73 J 7 (late twelfth century).
52. Schmidt-Künsemüller lists 138 examples, to which another has been added by
Gullick 2000. Of earlier literature on such bindings, see particularly Hobson
1988, papers I–IV.
53. De Hamel 1984, 64.
54. Ganz 2014 discusses treasure bindings across the whole of the Middle Ages,
giving the impression that the twelfth century was not one of its high
points. For a bibliography of earlier surveys, see his 26 nn. 4–6; funda-
mental is Loubier; also Fingernagel, ‘Der romanische Bucheinband’, in id.
2007, 355–408, and many German exhibition catalogues.
55. Liber Eliensis III. 50.
56. Ker, BCL, 230–1; J. Neve 1968, 37.
57. Borrie; Huws, 146–8.
 chapter 2

Book Script
Erik Kwakkel

The palaeography of twelfth-century manuscripts is defined by transfor-

mation. Book scripts used by European scribes show considerable change
over the course of the century, whereby either a substantial number of
letterforms or an entire script system was replaced. For example, Visigothic
disappeared over the course of the twelfth century, while Beneventan
shifted into a period of decline from around 1200, the first signs of which
are already observed in the twelfth century.1 An important factor in these
shifts is the emergence of Gothic script, or rather, the coming of features
that would ultimately culminate in the script we call Gothic Textualis.2 For
example, what Lowe identifies as markers of decline in Beneventan are, in
fact, Gothic traits that become woven into the script, including key
features such as the angular appearance of round strokes (‘angularity’)
and biting in adjacent letters with contrary curved strokes (‘biting’),
which Lowe calls ‘unions’.3 However, rather than focusing on geographi-
cally confined ‘national’ styles of handwriting, this chapter assesses the
scripts used for books across Europe in general. Ultimately this under-
taking brings to the foreground one particular kind of script, Caroline
minuscule, albeit that the twelfth-century version looks quite different
from what is encountered in the Carolingian age. The challenge of this
chapter is to assess how and to what extent twelfth-century script deviates
from ‘pure’ Caroline minuscule, how the influx of new features occurred
over time and whether something can be said about the geographical
spread of these novelties.

Caroline, Gothic and Pregothic

As in many other respects, the twelfth century is a transitional period from
a palaeographical point of view. As the period progresses, we witness the
waning of Caroline minuscule, the dominant book script from ca. 800 to ca.
1100, as well as the proliferation of letterforms that are usually regarded as
25
26 erik kwakkel
features of Gothic Textualis, a script whose life spans from the early thirteenth
to the sixteenth century. While there is considerable agreement about what
constitutes Caroline minuscule and Gothic Textualis, whose traits are fairly
well defined,4 palaeographers show less conformity about the script encoun-
tered in the twelfth century, a script that shows a mixture of Caroline and
Gothic features (Figure 2.1). Some define this style of handwriting as a late
expression of Caroline minuscule, addressing it with such terms as ‘Late
Caroline’ or ‘Post Caroline’, while others regard it as an early form of
Gothic Textualis, labelling it ‘Primitive Gothic’, ‘Proto-Gothic’, ‘Littera
praegothica’ or ‘Pregothic’. Emphasizing the script’s hybridity, a third
group calls it ‘Carolino-Gothica’, ‘Caroline gothicisante’, ‘Minuscola di
transizione’, ‘Übergangsschrift’ or ‘Transitional script’.5
There is also disparity with respect to the start and longevity of this
‘Pregothic script’, as it is addressed here for convenience. While most
palaeographical handbooks place the start in the late eleventh or early
twelfth century, the verdict of when the script reaches maturity varies
from the late twelfth century to ca. 1275,6 which may be reflective of the
geographical variation in the adoption of the script, as is discussed later in
this chapter. In the same handbooks Pregothic is treated as an entity of its
own, meaning that it is discussed in a separate chapter or section. There are
reasons not to do so. Both Caroline and Gothic are stable script systems
with core features that remained unchanged even as regional peculiarities
emerged. In contrast, Pregothic script is, at its heart, defined by continuous
change, given that it represents a moving point on the sliding scale from
Caroline to Gothic. Starting in the eleventh century, the script develops
from an almost pure form of Caroline minuscule with a modest number of
features we tend to define as Gothic to, in the thirteenth century, a script
that can almost be called Gothic Textualis, were it not for some remaining
traces of Caroline. By definition Pregothic script is never Caroline or
Gothic, but it represents a collection of stages in between the two script
systems.
This blend makes Pregothic problematic to define, study and under-
stand. A major problem is that of identity. It is unclear when Caroline
minuscule has acquired sufficient change to be called something else: how
many Gothic features does a Caroline bookhand need to take on before it
can be called Pregothic? The variety of terms used to address the transi-
tional script of the Long Twelfth Century is telling of just how differently
scholars are inclined to answer this query. Some observations in this
chapter further aggravate this problem, in particular that some of the
features defined as ‘Gothic’ in palaeographical handbooks are in fact
 Book Script 27

Figure 2.1 Pregothic book script, dated 1145–9. Leiden, Universiteitsbibliotheek,

BPL 196.
28 erik kwakkel
encountered as early as the second quarter of the eleventh century.
A second problem is that of description. While Pregothic is evidently
a different beast from the scripts that flank it, we have no vocabulary at
our disposal to explain precisely in what way. Because Pregothic script is
Caroline minuscule that includes, to a varying extent, Gothic features, one
is forced to describe Pregothic by referring to features of two other scripts.
There are, at any rate, no apparent script features that are unique to
Pregothic in that they were not included in Caroline or would not become
part of Gothic.

Assessing Pregothic Book Script

Given that the presence of Gothic features is ultimately what differentiates
Pregothic from Caroline, a way out of the first problem – when to call
a style of handwriting Pregothic – may be to focus on the emergence and
development of Gothic traits, as is done in this chapter. In an attempt to
define Pregothic and assess its development, this chapter drafts into service
all dated and datable manuscripts contained in the Catalogue des manuscrits
datés that were produced between 1075 and 1225, a total of 353 manuscripts.
For each of these, twenty-eight palaeographical changes are examined and
keyed into a database, noting whether a letter shape’s execution is in
Caroline or Gothic style, or whether a manuscript features both forms at
the same time (a mixed use of traditional and new forms).7 This quanti-
tative approach ultimately shows how the execution of letterforms changed
over time, allowing us to measure the waning of Caroline, the emergence
and attainment of Gothic features, the speed of their adoption and the
regional variety in their application. Before turning to these key issues,
the second problem – that of description – needs to be addressed. How
does one describe evolving letter shapes in such a way that they can become
part of quantitative research?
The answer to this question lies in a particular mode of change encoun-
tered in the developing script of the twelfth century. As I argue elsewhere in
more detail, medieval scribes in the process of adopting a new script
changed their scribal mannerisms in two ways.8 The first is through
a process that may be called ‘substitution’, whereby one graph (letterform)
was replaced by another. Pregothic script encompasses only a modest
number of these: the introduction of two uncial letterforms (round
d and s) and the emergence of the orum abbreviation for r (‘round r’).9
These new forms were introduced relatively late in the period and they
gained ground very slowly (Figure 2.2). For example, the oldest dated
 Book Script 29
100%

90%

80%

70%

60%

50%

40%

30%

20%

10%

0%
1075–99 1100–24 1125–49 1150–74 1175–99 1200–24
Round r 24% 17% 43% 63% 79% 64%
Uncial s 0% 0% 0% 2% 12% 20%
Uncial d 0% 0% 1% 6% 8% 37%
Manuscripts 26 38 75 98 64 52

Figure 2.2 Three examples of substituting letterforms.

manuscripts in which d and s are consistently presented in uncial form

were made in 1125–49.10 At the close of the Long Twelfth Century, in
1200–24, less than half of the manuscripts consistently present the new
forms (d: 37 per cent, s: 19 per cent). Round r, which was exclusively used
in or-ligature, gained a much firmer foothold: in the last quarter of our
period it is used in 64 per cent of manuscripts.11
A second mode of change, which can be called ‘modification’, was less
invasive and much more common. With modification, a graph was not
replaced by something entirely new, but the existing appearance was altered,
usually modestly. As in other medieval scripts, it concerned alterations on
the level of the stroke, the individual trace of the pen. Such modifications
were produced through a change in the stroke’s length (a reduction in some
cases, an extension in others), direction (the ‘vanishing point’ of the pen,
which can be quantified by comparing it to the dial of a clock) and shape
(e.g. straight, curved or forming a bowl). Additionally, scribes modified the
number of strokes that were used to produce a graph: either strokes were
added to the Caroline presentation of a letter or they were cut.12 This second
mode of change is crucial for this chapter, because the notions of ‘reduction’,
‘extension’, ‘direction’ and ‘number’ enable us to assess the hybrid script of
30 erik kwakkel
the Long Twelfth Century in a quantifiable manner. Approaching Pregothic
script in this fashion offers three important insights, which are addressed in
the remainder of this chapter: script development in the twelfth century
lacks cohesion, is much less innovative than traditionally assumed and shows
great regional variety.

Lack of Cohesion
The adoption rate of Gothic features over the course of the twelfth century
was uneven. One might perhaps have expected that a distinct form of
Pregothic emerged in the handwriting of a small number of scribes, such as
inhabitants of a certain intellectual centre or monks affiliated to a certain
order, after which that specific style gained popularity among a growing
number of scribes and geographical locations. However, the pattern of
development is very different. Notably, at a moment when a sharp increase
is seen in the adoption of one Gothic letterform, the popularity of another
increased only modestly or not at all. For example, at one point the
application of Gothic feet (which turn to the right) gains popularity with
surprising speed: in 1090–1104 only 20 per cent of manuscripts show this
feature systematically, while in 1105–19 the feature has jumped to 70
per cent.13 Typically, however, Gothic features show very little growth in
these two decades.
The sharp increase in both the adoption of angularity and Gothic feet
draws attention to something else. The two sharp increases occur in the
same manuscripts: fifteen of the seventeen manuscripts with consistent
Gothic feet also feature angularity. This suggests that a significant number
of scribes adopted two particular Gothic features in a short period of time,
while not showing interest in a great deal of other new traits. Notably, the
adoption of several new features by the same scribe is not a common
occurrence in the twelfth century and certainly not for a large group of
palaeographical shifts. It is only in the early thirteenth century that it
became common for scribes to adopt new Gothic features in larger
numbers.
The varying speed with which new features gained popularity and the
varying moments at which their popularity increased attest to a lack of
cohesion in the book script of the Long Twelfth Century, the development
of which appears uncoordinated and random. Other observations under-
score this assessment, such as the occurrence of apparently opposing
trends. For example, while one letter development entailed an extension
of a stroke (the second leg of h and x will ultimately be placed below
 Book Script 31
baseline), in other cases the same stroke was retracted (f, r and s).
Ultimately the development of book script between 1075 and 1225 can
perhaps best be understood as a collection of individual developments
culminating into a style of writing that no longer underwent significant
change, thus marking the birth of Gothic Textualis.

Innovation
While the chapters in this volume show how the century and a half
between 1075 and 1225 represents an age of renewal, it is necessary to
temper the traditional verdict that the period is innovative from
a palaeographical point of view. The main reason for moderation is the
important observation that the roots of many script innovations tied to the
twelfth century are, in fact, encountered much earlier. The extent to which
Gothic features are present in the last quarter of the eleventh century (when
Pregothic script is traditionally regarded as being in its infancy, as dis-
cussed) is striking. An example is the Pregothic trend whereby the stem of f,
straight r and long s was reduced in size: while Carolingian scribes placed
the feet of the stems below baseline, their peers writing Gothic Textualis
would ultimately place them on baseline (Figure 2.3). Dated manuscripts
suggest that a small portion of scribes in Europe (10 per cent) already

100%

90%

80%

70%

60%

50%

40%

30%

20%

10%

0%
1075–99 1100–24 1125–49 1150–74 1175–99 1200–24
f 12% 34% 68% 82% 78% 80%
r 38% 68% 79% 89% 83% 79%
Long s 35% 37% 51% 70% 72% 73%
Manuscripts 26 38 74 98 64 52

Figure 2.3 Examples of increased popularity of placing feet on baseline.

32 erik kwakkel
placed the f consistently on baseline in 1075–99. More scribes already did so
with long s (35 per cent) and straight r (38 per cent). Obviously the
beginning of this particular palaeographical trend predates 1075.
Many other Gothic features appear in such high numbers in the last
quarter of the eleventh century. For example, fifteen of the twenty-six
dated manuscripts made in that period (71 per cent) extend the second leg
of x below baseline (on baseline in Caroline); thirteen (50 per cent) present
their a in the Gothic fashion, with the vertical stroke in an upright position
(slanted in Caroline); eight manuscripts (31 per cent) present g with
a closed lobe (open in Caroline); the same number use uncial
d complementary to straight d; in seven manuscripts (27 per cent) the
‘tongue’ stroke of e is traced in the direction of two o’clock on the dial
(three o’clock in Caroline); and in four (15 per cent) the minims are in the
Gothic style, meaning their feet turn to the right (to the left or down in
Caroline). Notably, these examples all concern manuscripts in which
a scribe consistently presents a given letter shape in the Gothic style.
If we also include cases in which the Gothic presentation is used from
time to time in an individual manuscript that also presents these same
forms in the Caroline manner, the numbers are even higher. To give one
example, in 1075–99 an additional nine manuscripts show a mix of
Caroline and Gothic feet at minims, meaning that half of the twenty-six
manuscripts from that quarter-century show some degree of ‘Gothicness’
in the formation of their feet (either consistent or sporadic).
These examples highlight a key feature of script development in the
twelfth century: many of the palaeographical shifts measured in that
century have, in fact, older roots. For a number of script features these
grow deep into the eleventh century. For example, the sixty-five dated
manuscripts produced between 1000 and 1074 show that the practice of
placing the stems of f, r and s on baseline was not uncommon in the second
quarter of the century (Figure 2.3). Among the seventeen dated manu-
scripts from this quarter-century two place long s consistently on baseline,
three do so for f and five for straight r. Given the absence of these traits in
dated manuscripts from the first quarter, it is tempting to infer that the
novelty of placing feet on baseline emerged in the second quarter of the
eleventh century.14 Some other Gothic traits appear to be even older. For
example, among the twenty-three dated manuscripts from the first quarter
of the eleventh century there are four in which the stem of t pricks through
the bar consistently, while eight present g with closed lobe.
While the numbers in these examples are still relatively low, they do
highlight how features considered typical for Littera Textualis were
 Book Script 33
practised by scribes in the first half of the eleventh century, well before the
Gothic or even Pregothic period. In fact, while more scribes in the twelfth
century began to favour Gothic traits as the century progressed, only
a limited number of features are actually innovations of the century itself.
Dated manuscripts suggest that the few real palaeographical novelties are:
the five different kinds of biting (involving the letters b, d, h, o and p), the
extension of the second i in ij, the use of uncial s in final position and the
tailed e (e-caudata) that loses its tail.15
There is another reason why we may need to temper a claim that the
twelfth century is an innovative period as far as script is concerned. Dated
manuscripts show how the process of adopting Gothic features had by no
means been completed by the early thirteenth century. In fact, during the
first quarter of the thirteenth century few palaeographical features are
consistently executed in the Gothic fashion in all dated manuscripts.
Notably, ten Gothic features are encountered in fewer than 50 per cent of
the manuscripts in the corpus (which, in the period 1200–24, consists of
fifty-two manuscripts): uncial d (45 per cent), biting in ‘de’/’do’
(39 per cent), diacritical ij (38 per cent), biting following
p (26 per cent), biting following b (24 per cent), biting following
h (19 per cent), uncial s in final position (17 per cent), biting following
o (15 per cent), use of round r after b or p (13 per cent), ct-ligature
(9 per cent) and diacritical single i (7 per cent). In other words, the script
developments occurring over the course of the twelfth century do not
culminate in a script that has finished developing. It appears that,
palaeographically speaking, the period is part of a much longer conti-
nuum, as is also suggested by the presence of Gothic features in the first
half of the twelfth century.

Regional Variety
Dated manuscripts also highlight, lastly, a lack of cohesion in the geogra-
phical spread of Gothic features, which varied significantly in terms of both
speed and execution (the manner in which the letters were formed).
The extent to which the regional acceptance of Gothic traits varied
becomes clear when we compare England, France and the Germanic
countries (nowadays Austria, Germany, The Netherlands and
Switzerland, and perhaps Flanders as well), which formed a separate
Kulturraum.16 The general trend is that during the twelfth century far
fewer Germanic scribes favoured Gothic traits in comparison to their peers
in England and France. The placement of f, s and r on baseline may serve as
34 erik kwakkel
an example: this trend is much less popular among Germanic scribes.
In most quarter-centuries more than twice as many scribes in England
and France execute their r in the Gothic fashion than do their counterparts
in Germanic countries.17
Another regional peculiarity of Germanic countries is that some Gothic
features are never used there, even when they are well established elsewhere.
This phenomenon is witnessed most clearly in fusion or ‘biting’, an
important Gothic feature whereby two adjacent contrary curved letter-
forms started to overlap.18 The feature whereby uncial d consistently
merges with round letterforms in an adjacent position (‘de’, ‘do’) first
appears in 1150–74 in England and France, albeit in a very low number of
manuscripts (4 per cent of surviving dated manuscripts). From there it
grows in popularity to 13 per cent (England) and 12 per cent (France) in
1175–99, and subsequently to 60 per cent (England) and 30 per cent
(France) at the end of our period, in 1200–24. Notably, this particular
type of fusion is encountered in none of the dated manuscripts from
Germanic countries in these same periods. The same goes for fusion
involving h (‘he’, ‘ho’), o (‘od’, ‘oe’, ‘oq’) and p (‘pe’, ‘po’). Scribes in
England and France, in contrast, did use these forms, although not all of
them did so in significant numbers. In the last quarter-century of our
period relatively few cases of biting involving h are observed (France:
17 per cent, England: 13 per cent), as are those involving o (France:
15 per cent, England: 7 per cent). More frequent is fusion with the letter
p (France: 22 per cent and England: 31 per cent). Here the contrast with the
mannerism of Germanic scribes, who do not fuse letters with p at all, is
most profound.
Observations like these underscore the importance of studying script in
the Long Twelfth Century on a regional as well as a broader European
level. Moreover, they also identify regional differences as yet another
variable in the development of book script – that is, in addition to the
precise moment at which Gothic features were introduced and the speed
with which they became more popular. Within these large geographical
spaces smaller regions may be identified with their own palaeographical
peculiarities (e.g. southern France versus France as a whole).19 Within such
smaller regions two kinds of palaeographical idiosyncrasies are observed.
The first is related to the adoption of Gothic features, and it reflects the
pattern witnessed on a supra-regional level: a Gothic feature may be
introduced at a different moment or develop at a different speed. For
example, scribes in southern France tended not to execute the feet at the
minims of m and n in the Gothic fashion (with sharp flicks to the right),
 Book Script 35
but commonly directed them straight down. This happened as late as
the second half of the twelfth century, when even the majority of
Germanic scribes had embraced this feature.20
The second manner in which a smaller geographical space could branch
off in a palaeographical respect does not concern the introduction of
Gothic features as such, but the manner in which the new features were
executed. For example, while scribes in southern France were at par with
their peers in other French regions in the adoption of the seven-shaped
Tironian note, which first supplemented and later fully replaced the
ampersand, they actually shaped this symbol in a uniquely southern
French manner: it is characterized by its upright appearance and by its
very long and straight horizontal stroke. Similarly, scribes in the region had
their own way of shaping, for example a, i, ta and the con-abbreviation.21
These distinct southern features show that confined geographical areas
could develop their own ‘brand’ or ‘interpretation’ of Pregothic.
Among all this geographical variation one region appears to stand out in
terms of advancement. As the examples given here show, across the board it
is France that most frequently comes in first place regarding the introduc-
tion moment of new features and their rate of adoption. It is possible,
however, to home in on a region within France where Gothic features are
encountered notably early and in high numbers: Normandy. For example,
Norman scribes are very early adopters of Gothic angularity and the
Gothic fashioning of feet.22 There is more evidence for the advanced
position of Normandy. It turns out that Norman scribes also take
a prominent position at the head of the column when we observe all
(twenty-eight) palaeographical traits that underwent change. When we
place the dated manuscripts from 1075 to 1099 in the order of the number
of features that have consistently been copied in the Gothic fashion, the
first four turn out to have been made in Norman houses, while the fifth
was produced in Christ Church, Canterbury, a community that included
a large contingent of Norman monks.23

Other Modes of Writing

Given the focus of this volume, namely manuscript books and their con-
tents, the script used for documentary texts has so far been excluded from
discussion.24 While full manuscripts are not usually copied in documentary
script, it deserves a place in this chapter because it was, from time to time,
used for copying a segment of a manuscript. Pregothic documentary
script is closely related to Pregothic book script.25 Pronounced differences
36 erik kwakkel
are the high head of a (which could be extended significantly), the
extension of ascenders and descenders, the extension of f and long
s below baseline, the sharp turning to the left of the feet at f and s (and
sometimes p) and the near-formation of loops at the ascender of uncial
d and sometimes f, though this curl is never completed into a real loop
(true cursive elements, such as connections between letters, are absent
in the script). The top of ascenders is sometimes accentuated with
a decorative motif, such as a superfluous curl. All of these features are
most clearly observed when Pregothic documentary script is written with
a thinner and more flexible pen (as was customary for the production of
documents). They are also encountered, though less pronounced, in
specimens written with the broad nib used for producing books, although
the ascender of d and f do not usually curl.26
When used in manuscripts, Pregothic documentary script is mainly
employed as a contrast script, although for this purpose scribes in the
Long Twelfth Century preferred to use a smaller version of Caroline or
Pregothic, which is sometimes, for this purpose, written with a thinner pen
than the main text. Documentary script plays a role in the hierarchy of
scripts in that it expresses that the text in question is not part of the actual
main text but is somehow standing apart from it. Pregothic documentary
script was most often used for glosses, both interlinear and in the margin
(Figure 2.4),27 although scribes in the Long Twelfth Century clearly pre-
ferred a smaller version of Caroline or Pregothic for this purpose. Bischoff
therefore called the script in question ‘Glossenschrift’, although others prefer
the term ‘notula script’.28 When used for glosses, the script is usually notably
smaller than the version used in charters, which probably results from the
confined space of the margin and in between two lines.
A larger version was used for adding text within the actual text columns,
although this appears to have happened infrequently. In such cases the
similarity with Pregothic book script is much more striking than in the
minuscule gloss version: the central part of the letters (ascenders and descen-
ders excluded) is effectively the same. This larger version of the documentary
script is nearly always used for writing down short texts, such as ex libris
inscriptions,29 colophons,30 notes,31 tables of contents,32 short introductions,33
capitula lists,34 calendar or obituary entries,35 enumerations36 or segments of
manuscripts.37 Full texts are rare, although there are exceptions, such as
the autograph of Nigel Witeker, presumably a draft text, which was
written in 1193–4 (Cambridge, Gonville and Caius Coll. 427/427), and
a Computus written in England or Wales in 1164–8 (Bodl. Libr. Digby 56).
Here scribes resorted to documentary script not so much because of the
 Book Script 37

Figure 2.4 Pregothic documentary script used for added glosses, eleventh century,
with twelfth-century marginal and interlinear glosses. Leiden,
Universiteitsbibliotheek, VLQ 51.

contrast it provided with the Pregothic book script (which is not even found
in the manuscripts), but probably because it provided a faster means to copy
the text. The ‘utilitarian’ nature of the works may have invited a less formal
writing style because it meant the copying could be done with less effort.
From time to time one encounters a manuscript from the Long Twelfth
Century copied in a bookhand that was influenced by Pregothic docu-
mentary script. Such influence is often shown by an extension of f and
s well below baseline, perhaps even with the foot bending sharply to the
left. A notably early example of such influence is a manuscript made in
Freising in 1022 (BSB Cgm 5248/7), with extended f and s and decorative
curls at the top of ascenders.

Making a Script
The observations presented so far prompt several important queries:
What motivated scribes to seek new ways of executing letters? How did
Gothic features become established among individual scribes or groups
38 erik kwakkel
of scribes sharing a scriptorium? How does a palaeographical feature
turn from idiosyncrasy into norm? A key notion at the heart of these
issues must be training, which is where the acquisition of any script
started. While little is known about scribal training, in most monas-
teries of the eleventh and twelfth centuries there will probably have
been a person assigned the task of teaching novices to write. Some have
argued that the cantor played this role, given that he was responsible
for running the school and supplying scribes with the materials for
producing manuscripts.38
Cohen-Mushlin’s study of the scriptorium at Frankenthal in
the second half of the twelfth century suggests that students learned
to write a script by studying writing samples of their master and
attempting to imitate his style. This was done, she states, to ensure
the production of palaeographically ‘homogenous and uniform
manuscripts’.39 Surviving Frankenthal manuscripts show how this was
achieved. First the master wrote out a few lines, after which the student
took over and wrote a few lines of his own. Then the master took over
again, writing a few lines, after which the pupil wrote some more.40
‘Taming’ pupils in this fashion (as Cohen-Mushlin calls it) – which
implies that master and student actually sat next to one another – may
have been a much broader practice. It is also encountered, for example,
in manuscripts copied by Orderic Vitalis (d. 1142) in the Norman
house of St-Évroult.41
The teaching method whereby the teacher’s handwriting is used as
a model and where the teacher closely monitors how well the student is
following his example demonstrates the importance of the monastic writ-
ing master to the process of script change, at least within individual
communities. After all, a conservative teacher could arguably hold back,
palaeographically speaking, several generations of new monks in his vici-
nity, while one who was willing to weave new letterforms in his script had
the ability to advance script around him. Particularly important is how
closely the script of the pupil could match that of his teacher, which is
evidenced both by the manuscripts from Frankenthal and by those pro-
duced by Orderic Vitalis. In fact, in books from St-Évroult a hand is
encountered that looks so similar to Orderic’s style of writing that the
writer is dubbed his alter ego. Chibnall concluded that the scribe ‘had learnt
to write under [Orderic’s] guidance, and had modeled himself remarkably
closely to his master’.42 Such observations suggest that if the writing master
included Gothic features, these subsequently had a good chance of spread-
ing through the community.
 Book Script 39
Still, training cannot be the whole story. The history of Pregothic script
is one of continuous – if inconsistent – change, which could only have
occurred if monastic writing masters were introduced to new script features
on a regular basis. In other words, another key notion in the development
of book script in the Long Twelfth Century must be travel, either by
members of religious houses or by their books. The Norman Conquest
shows just how profound the influence of travelling mannerisms could be
on the development of a book script. Norman scribes, whose handwriting
was heavy on Gothic traits, spread an advanced form of Pregothic script
throughout England as they entered religious communities there.43 This
may help explain why the scripts of England and France take a similar path
of development in the late eleventh and early twelfth centuries, as the
figures shown earlier suggest. Moreover, the scripts include palaeographical
traits not seen elsewhere in Europe, such as the macron in the shape of
a bowl that is slightly slanted.
In the wake of the Conquest two houses in Kent adopted a writing
style that was modelled on script from the Norman abbey of Bec.
The script, which is known as ‘prickly’ because of the split tops of
ascenders, the use of hairlines and the pointy top or back of round
letters (c, e, o, t), was first developed in Christ Church, Canterbury
and then brought to Rochester, where it was used in its own distinctive
style.44 Rochester monks all came from Normandy, but some were
trained in other Continental regions, such as Germany, the Low
Countries and Italy. In the first quarter of the twelfth century, they
all abandoned their native styles and switched to the new prickly script,
which may have been modelled on the handwriting of Ralph, the
community’s prior until 1107, whose manuscripts are early and pro-
nounced examples of prickly script.45 The new script was executed so
perfectly within the community that the Continental origins of their
users became hidden: only when the scribes tested their pens on
flyleaves did they reveal their native script, as if lowering their guard
for a few moments.46
The case of Rochester not only underscores the importance of modelling
and travel but also how an entire community could quickly and perfectly
switch palaeographical register and acquire a new script that was heavy on
Gothic traits. Given these observations, the lack of speed and consistency
in the development of book script during the twelfth century in general is
all the more striking: if individual communities and regions could adopt
Gothic features so quickly and consistently, why does the Europe-wide
process of adoption lack speed and uniformity?
40 erik kwakkel
Notes
1. Bischoff 1979, 122–9, esp. 129 (Visigothic); Lowe 1999, 125–6 (Beneventan).
2. Derolez, Gothic, 72–122.
3. Other examples: the extension of x below baseline, the use of round r in or
combination, the formation of ct-ligature. See for these and other Gothic
traits Lowe 1999, 126–49 (angularity at 125–6, biting at 149).
4. See the enumeration of features in Derolez, Gothic, 47–57 (Caroline) and
56–71 (Pregothic). Also see Bischoff 1979, 154–62 (Pregothic).
5. Derolez, Gothic, 57. The term ‘Übergangsschrift’ is used in Schneider 1999,
30–1, while I have used the term ‘Transitional script’ in Kwakkel 2012, 85.
6. For example, Bately, Brown and Roberts 1993, 55 (late eleventh to late twelfth
century); Brown 2002, 73 (late eleventh century to middle of thirteenth
century); Derolez, Gothic, 72 (Gothic is completed ca. 1200); Roberts 2005,
104 (from 1100 to 1150 to the early thirteenth century); Schneider 1999, 28 (late
eleventh or early twelfth century to mid-thirteenth century or ca. 1275).
7. This assessment leans heavily on palaeographical research undertaken within
my NWO-sponsored research project ‘Turning Over a New Leaf’ (2010–5).
See Kwakkel 2012, table 3 at 112–25 for the 353 manuscripts in question and the
criteria for their inclusion; at 86–7 twenty-one of the twenty-eight graphs are
listed. I have since added seven, five of which are types of ‘biting’ (as noted
further in this chapter).
8. Kwakkel, in press.
9. There are others, such as the exchange of ampersand for seven-shaped
Tironian note and the changing appearance of the macron. However, this
chapter is confined to letterforms.
10. This chapter uses twenty-five-year increments, with the exception of instances
where the start and end of developments are discussed, in which case fifteen-
year increments are used.
11. The actual data to support these and the following statistics are too elaborate
to include in this chapter. They will be made part of the monograph I am
presently preparing on the birth of Gothic script.
12. For a more detailed discussion, see Kwakkel, in press.
13. The percentages generated by the data at the heart of this chapter should be
taken as approximate estimations, not absolutes. The percentages mentioned
in this chapter are important for establishing whether a certain period showed
significant palaeographical change and in which direction it moved, as well as
for determining when a new feature was approximately introduced. See
Kwakkel 2012, 206, graph 1.
14. There is one exception from near the end of the first quarter: BSB Cgm 5248/7
(Freising, ca. 1022) has the f consistently on baseline.
15. See n. 11.
16. See Kwakkel 2012, 91 and 102. General studies of Pregothic in these regions are
Parkes 2008; Parkes 2008a, 93–100; Roberts 2005, 104–7 (England);
Schneider 1987 (Germany); Bischoff 1979, 157–8 (France).
 Book Script 41
17. Differences between Germanic countries and France in the six quarter-
centuries: 44 per cent, 58 per cent, 56 per cent, 35 per cent, 69 per cent
and 17 per cent; differences between Germanic countries and England:
60 per cent, 62 per cent, 63 per cent, 39 per cent, 69 per cent and
31 per cent.
18. For this feature, see Derolez, Gothic, 57–8, and Kwakkel 2012, 96–102,
including the identification of different phases of development.
19. See also Chapter 4 of the present volume.
20. The same observation is made by Derolez, Gothic, 117 (‘a lack of precision in
treatment of the feet of m and n’).
21. On southern French features, see Derolez, Gothic, 116–7.
22. Kwakkel 2012, 94 and 96.
23. These are Bodl. Libr. Lat. th. d. 20; Rouen, Bibl. mun. 1406 (St-Ouen,
1072–92); Rouen, Bibl. mun. 1409 (Jumièges, 1078–95); CUL Ii. 3. 33
(Christ Church, Canterbury, 1079–1101); Rouen, Bibl. mun. 477 (Fécamp,
ca. 1075).
24. Book and documentary scripts of the same age are ideally studied together;
Derolez, Gothic, 4–6.
25. For features of emerging Gothic cursive script, see Derolez, Gothic, 125–8.
26. See, for example, Brown 2002, plate 27 at 79 (compare specimens a and b,
written with flexible pen, to c, which is done in a thicker pen).
27. Leiden, Universiteitsbibl. VLQ 51. Other examples: Bodl. Libr. Canon. Class.
lat. 41 (Juvenal, eleventh century); Rouen, Bibl. mun. 57 (Glossed Bible,
twelfth century); BL Burney 161 (Cicero, ca. 1150–1200).
28. Bischoff 1953, 8; Schneider 1999, 26–8.
29. For example, Leiden, Universiteitsbibl. BPL 196 (1145–9).
30. For example, Leiden, Universiteitsbibl. VUL 46, dated 1176–7; Avranches,
Bibl. mun. 91, f. 175v (twelfth century).
31. For example, BnF nouv. acq. lat. 214, f. 194r, ca. 1151 (donations?); Leiden,
Universiteitsbibl. VLQ 12, f. 67v, dated 1190 (description and signature of
individuals); BL Egerton 3661, f. 15v, of 1216 (note following explicit).
32. The Hague, Koninklijke Bibl. 76 E 15, dated 1173–83; Dijon, Bibl. mun. 114,
f. 1v, written 1183–8.
33. For example, Brussels, Bibl. royale II 2425, dated 1132–5 (prologue to
the Bible).
34. For example, Bodl. Libr. Canon. Pat. lat. 148, f. 99r, dated 1145; Leiden,
Universiteitsbibl. BPL 20, f. 5r, the famous copy of the Dukes of Normandy,
1136–7.
35. For example, Soissons, Bibl. mun. 9, f. 122v, copied 1178–9 (calendar); Graz,
Universitätsbibl. 1703/137, f. 1r, copied 1225 (obituary).
36. Leiden, Universiteitsbibl. BPL 20, f. 10r (list of abbots).
37. Admont, Stiftsbibl. 434 (1166–9), containing Gerhoch of Reichersberg’s
Epistolae with corrections in autograph.
38. Steinmann 2010, esp. 31–2.
39. Cohen-Mushlin 2010, 64.
42 erik kwakkel
40. As seen in ÖNB 1568, discussed in Cohen-Mushlin 2010, 64–5.
41. Such prompting is seen in, for example, Rouen, Bibl. mun. 31 and in a new
Orderic Vitalis manuscript introduced in Weston 2016. Steinmann 2010, 31,
discusses how the twelfth-century Consuetudines of Fruttuaria mention how
student and master sit next to one another, reflecting perhaps a similar
practice.
42. Orderic, 2. xxxix–xl, citation at xl.
43. Ker, English MSS, 22–32. For the opposition of English and Norman scribes
against European scribes in general, see also Parkes 2008, 111.
44. Ker, English MSS, 26–8 (Bec influence on 27). See for the script also Webber
1995 and 2011, 214 (Norman origins).
45. For example, BL Royal 12 C. i; Waller 1984, 240.
46. On Continental scribes in Rochester, see Kwakkel 2013.
 chapter 3

Decoration and Illustration

Martin Kauffmann

The position of a chapter on decoration and illustration in a volume

devoted to the twelfth-century book is in some ways an uncertain one.
The simplest decoration may be considered merely part of the scribal
design and structuring of the text, something to be studied by the palaeo-
grapher. In catalogues of manuscripts, the line to be drawn between
rubrication (often treated, together with script, in a section devoted to
codicological description) and decoration is obscure. At the other end of
the scale, it could be said that the pictorial or illustrative elements in
a manuscript, which accompany a text, whether a bible, liturgical book,
chronicle, or book of natural science, should be treated as a constituent part
of the relevant textual genre: historians should include in their treatment of
the text the artistic production related to it. Yet it is the perception that
these two ends of the scale, the ‘lowest’ decoration and the ‘highest’
illustration, are connected, that may constitute the distinctive field of the
art historian. The elements of decoration and illustration exist in a state of
hierarchical relation. They are aware of each other and do not usurp each
other’s roles. There is almost nothing which can usefully be said about
either the humblest initial or the greatest full-page miniature if one forgets
that in each case these elements form part of the entirety of the illumina-
tion in the same book (Figure 3.1). It follows that the different elements all
participate in versions of a single stylistic vocabulary which could be
applied at different levels. That vocabulary is often labelled ‘romanesque’,
and we shall have to confront the term in our survey of the different
questions asked of and the approaches taken to decoration and illustration
in the twelfth-century book.

By Whom and for Whom?

First, however, we should touch on the evidence for the patronage and
production of illuminated books: artistic styles are not self-generated,
43
44 martin kauffmann

Figure 3.1 Frontispiece to the commentary of Jerome on the book of Isaiah,

illuminated by Hugo ‘pictor’, Normandy, late eleventh century. Oxford, Bodleian
Library, Bodley 717, ff. vv–vir.
Decoration and Illustration 45

Figure 3.1 (cont.)

46 martin kauffmann
organic forms, but exist in books made by and for particular people at
particular times and places. Whereas the greatest illuminated manuscripts
of the Carolingian and Ottonian periods were produced under the patron-
age of emperors, or of the imperial bishops who were often their relatives,
the greatest patrons of illuminated manuscripts in the twelfth century were
the monastic houses. Luxury books were still produced in monasteries
either for the emperor himself, such as the gospel book made either for
Henry IV or for Henry V at the abbey of St Emmeram, Regensburg,
between 1099 and 1111 (Cracow Cath. 208),1 or for other rulers, such as the
Gospels of Henry the Lion, duke of Saxony and Bavaria, produced ca.
1185–8 at the abbey of Helmarshausen and containing two complex ded-
ication miniatures extolling Henry’s power (Wolfenbüttel, Herzog August
Bibl. Guelf. 105 noviss. 2° and BSB Clm 30055).2 But the main patrons of
illuminated books in the twelfth century were the monasteries themselves,
and the production was carried out within their walls. It was not only
the old-established Benedictine houses which were involved in this
activity; the houses of the reformed Orders also contributed. The new
Premonstratensian houses, for instance, belonging to the Order founded at
Prémontré in 1120 by Norbert of Xanten, played a particularly active part
in the production of illuminated books – witness the Bible from the abbey
of Floreffe (near Namur) of ca. 1160 (BL Add. 17737–8) (Figure 3.2).3
Monasteries sometimes produced illuminated books for others: the
recipients could include dependent houses, such as Dover Priory (founded
1139), which probably received its Bible (CCCC 3–4) from its mother
house, Christ Church Cathedral Priory, Canterbury.4 Books are portable,
and the carrying of them from one house to another, whether in the
possession of an itinerant owner or as gifts, must have been an important
means for the transmission of stylistic innovation. Among the gifts to
Durham Cathedral Priory donated by Bishop William of St-Calais (some-
times called Carilef) were books written and illuminated in Normandy,
where William had spent time in exile between 1088 and 1091;5 and it was
probably the ‘Second Winchester’ (or ‘Auct.’) Bible (Bodl. Libr. Auct.
E. inf. 1–2) that was given by King Henry II of England to his new
foundation, the Carthusian house at Witham, where Hugh (later bishop
of Lincoln) was abbot.6 By contrast, there is comparatively little evidence
for the production of illuminated manuscripts by or for the secular (non-
monastic) clergy, and an illuminated book made for one of their number
might still have been produced in a monastic context: the Lectionary of
Archbishop Frederick of Cologne (Cologne, Dombibl. 59) probably origi-
nated in one of the monastic houses in the city.7 There is even less evidence
 Decoration and Illustration 47

Figure 3.2 Frontispiece to volume 2 of the Floreffe Bible, showing scenes from the
book of Job, virtues and works of mercy, and Christ and the Apostles, described by
the tituli as a representation of the Active Life; valley of the Meuse (modern
Belgium), ca. 1153–6. London, British Library, Add. 17738, f. 3v.
48 martin kauffmann
for the lay patronage of illuminated books; not until the thirteenth century
were wealthy literate laity sufficiently numerous to form an important
source of patronage.
Most artists in the twelfth century worked anonymously in the sense
that they did not append their names to their works. The signatures of
artists are rarer than those of scribes: the Rule of St Benedict enjoins
humility on monastic craftsmen. This complicates the model of patron-
age commonly applied to art historical studies of later periods in
European art. Instead of named individual artists and patrons, whose
relationship is commercial but who may in some sense be regarded as co-
creators of the works of art, we must imagine artists who were themselves
members of a corporate body which was the patron: this makes it
especially difficult to explore the creative balance between individual
and institution. But a few artists do identify themselves, either verbally
or visually, as monks. In the case of the sumptuous Gospels of Henry the
Lion, one individual, Herimann, monk of Helmarshausen, seems to have
been responsible for both the script and the illumination. Hugo ‘pictor’,
whose work is found in manuscripts from more than one Norman house,
painted and labelled a self-portrait at the end of a manuscript which
was subsequently taken to Exeter Cathedral (Bodl. Libr. Bodl. 717)
(Figure 3.1).8 The Premonstratensian canon Rufillus is one of several to
have painted himself in the act of painting one of the initials in the book
he illuminated (Cologny-Genève, Bibl. Bodmeriana 127, f. 244r).9
The production of illuminated manuscripts was always collaborative,
and the circumstances of that collaboration were evidently various.
The artist-priest Sintram, an Augustinian canon of Marbach, depicted
himself in adoration of the Virgin Mary together with the scribe Guta,
a nun of the convent at Schwarzenthann, in a manuscript dated 1154
(Strasbourg, Bibl. du Grand Séminaire 37, f. 4r).10
Although most illuminated books were made for monastic patrons, they
were not necessarily illuminated by monks: there is evidence of lay profes-
sional artists working for (and sometimes in) monasteries. Studies of the
illuminated manuscripts produced by individual religious houses have
been made, and the word ‘school’ has even been applied to them in
conformity with the model applied to Renaissance and later European
art;11 but the most outstanding (and thus best known) artistic productions
from those houses have sometimes been found to accord least well with the
house style identified in more modest productions. Whereas the skill of
writing in this period was largely limited to those in clerical orders of
varying sorts, the skill of painting could be acquired by a layman.12 The lay
 Decoration and Illustration 49
illuminator of the Dover Bible is another to have painted himself in the act
of painting an initial, while his assistant prepares the colours (CCCC 4,
f. 241v). In several cases there is evidence for lay artists working in associa-
tion with monastic scribes: the lay painter Felix depicted himself in an
initial (albeit in separate roundels) with the scribe John, the one-eyed
monk of Corbie (BnF lat. 11575, f. 1r).13 The Bible from Bury St
Edmunds of ca. 1135 (CCCC 2) was illuminated by a lay artist known as
‘Master Hugo’, who was also a sculptor and metalworker (Figure 3.3).14
Royal patronage had made Bury one of the richest foundations in the
country, so it could also afford to attract the artist known by modern
scholars as the Alexis Master, previously active at St Albans, to illustrate the
Life of its patron saint (New York, Pierpont Morgan Libr. M. 736).15 Thus
the style of a lay artist was not necessarily the property of a particular
monastic house, and the mobility of professionals, no less than the mobility
of books, must have contributed to the spreading of stylistic innovations.
In the middle and second half of the twelfth century there is increasing
evidence of itinerant artists. The master of the Lambeth Bible (London,
Lambeth Palace 3 + Maidstone Museum P. 5) illuminated the Gospels of
Abbot Wedric of Liessies in Hainault in 1146 (of which two leaves survive:
Avesnes, Société Archéologique).16 A series of manuscripts with illumina-
tion in the ‘Transitional’ style of the last quarter of the century is associated
with a group of artists who evidently worked on both sides of the Channel:
the artist of the initials in the last volume of the Capucins Bible, produced
perhaps in Troyes (BnF lat. 16743–6), had previously been employed at St
Albans under Abbot Simon (1167–83).17 Nevertheless, from the twelfth
century we have almost none of the apparatus associated with professional
artists which survives from the thirteenth, such as contracts, property
records, and verbal and visual instructions. Lay artists clearly participated
in monastic production and may well have received boarding and lodging
in a monastic house while active there, but there is little evidence for the
establishment of independent lay professional workshops of the kind
found in later centuries.

Styles and Techniques

Whether monastic or lay, artists (like those involved in other aspects of book
production) must usually have learnt their craft by practical instruction.
The technical manuals from this period include treatises by practising artists,
extracts from earlier sources (sometimes with added original material), and
isolated recipes or instructions. They vary in the extent of their usefulness,
50 martin kauffmann

Figure 3.3 Initial F(rater Ambrosius), letter of Jerome to Paulinus, from the Bury
Bible: England, ca. 1135. Cambridge, Corpus Christi College, 2, f. 1v.
 Decoration and Illustration 51
and seem never to have achieved a wide circulation.18 The evidence for
techniques must derive mostly from the manuscripts themselves, occasion-
ally supplemented by references in documentary sources.19 Illumination was
a collaborative activity, carried out almost always before the book was
bound. By leaving spaces to be filled, it was the scribe who decided (or at
any rate whose work determined, on the instruction of others) the distribu-
tion of decoration and illustration. Preliminary drawings survive in hard
point, (greyish) lead point or (brownish) crayon, sometimes gone over in
ink, but there are few of the marginal instructions for subject-matter found
in later manuscripts. Though drawings can be of high aesthetic quality, they
were not usually regarded as finished works of art, though (as we shall see)
they were considered suitable for certain kinds of subject-matter, and strong
graphic traditions are evident in manuscripts from particular parts of Europe
(Figure 3.4).20 The Anglo-Saxon technique of tinted drawing makes
a reappearance in a manuscript of Bede’s Life of St Cuthbert (Oxford,
University Coll. 165), produced at Durham Cathedral Priory in the early
twelfth century and probably the earliest surviving manuscript containing
a cycle of religious narrative illustrations to have been produced in England
after the Norman Conquest.21 Unfinished manuscripts, and manuscripts in
which a change of plan is identifiable, are a particularly valuable resource.
In the Winchester Bible (Winchester Cath. 1) we have both.22 We can see
that initials drawn by one set of artists were painted over and completed by
subsequent artists, who respected the subject-matter but updated the style.
Full-page miniatures were added to the book, which had originally been
conceived as containing only initials. We can study the various stages of
finish, observing for instance that if gold was involved, it was inserted onto
the preliminary drawing before the application of other colours, perhaps
because the act of burnishing threatened the surrounding area. Colour
washes were overlaid with stronger or lighter tones to provide shadows and
highlights before the final outlining of the contours of figures and drapery
folds. In twelfth-century manuscripts the choice of colours is sometimes
specified in instructional notes, words, or abbreviations situated within the
margin or within the artistic space and designed to be covered.23 This has
been taken as evidence of a division of labour, though it is not impossible to
imagine a scribe/artist writing notes to himself. The reliable analysis of
pigments has been hampered by a reluctance to take samples, though the
possibilities of identification by new imaging techniques such as Raman
spectroscopy are just beginning to be realized.
The style in which artists worked in the twelfth century is often
described as romanesque, a concept which has been subjected to scrutiny
52 martin kauffmann

Figure 3.4 Drawing of man as microcosm, from the Prüfening Miscellany,

Germany, 1158–65. Munich, Bayerische Staatsbibliothek, Clm 13002, f. 7v.
 Decoration and Illustration 53
in recent writing.24 This is a term applied to all the arts. It is the inter-
nationalism of the arts which justifies the use of a single descriptive term;
indeed, the romanesque style is often characterized as the first international
style in Western Europe since the fall of the Roman Empire. Even with
a manuscript from the middle of the eleventh century, it is often possible at
a glance to distinguish the style of the illumination in national terms such
as Ottonian or Anglo-Saxon. But by the middle of the twelfth century that
sense of national identity is less emphatic, though stronger regional tradi-
tions persisted in France and Germany than in the politically more cen-
tralized kingdom of England (and many of the most important modern
catalogues and studies of twelfth-century illuminated manuscripts are still
arranged by country).25 Some features considered characteristically roman-
esque, such as the stern figures relating to one another by intense stares and
expressive rhetorical gestures, were descended from their Ottonian pre-
decessors. Art historians have begun to explore the performative implica-
tions of gestures such as the raised and curving index finger, which has its
roots in the classical declamatio.26 But there are important differences
between the Ottonian and romanesque aesthetic. Gone are the remnants
of classical space, the pink and blue skies and clearly defined ground which
had survived into the Ottonian period; they are mostly replaced by flat
backgrounds of different coloured panels. The depiction of drapery – the
discussion of which has traditionally occupied a central place in the study
of romanesque art – delineates the individual body but at the same time
forms a linear pattern which welds individual figures into a unified whole.
The ordered balance and high seriousness of many of the figural composi-
tions contrasts with the exuberant invention of the decoration, something
to which we shall return.
The identification of a Byzantine element, of ‘Byzantine influence’, is also
intrinsic to the characterization of romanesque. The new internationalism of
romanesque art was partly a function of the respect for the ancient traditions
of Constantinople (and admiration for its wealth) among the kingdoms of
the West, fuelled by new contacts in the age of the Crusades and the Norman
kingdom of Sicily.27 Facial modelling with ochre or greenish shadows and
white highlights was derived from Byzantium. The ‘damp fold’, the
pear-shaped panel of drapery which delineates the body beneath, was
a formula derived from Byzantine art which has been thought to give
expression to new forms of self-confidence and self-consciousness. Art
historians have identified two ‘waves’ of the concentration on Byzantine
art, one early in the century and another distinct phase around 1170 in
which the expressive characteristics of late Comnenian art make an
54 martin kauffmann
appearance. All this of course does not explain the means or routes by which
such ideas travelled. Once a stylistic trait had been adopted in one Western
centre, it presumably did not require further direct contact with Byzantium
for it to be picked up in another. Italy has usually been seen as the main
channel – Roman and south Italian painting in the early part of the century,
Sicilian and Venetian mosaics in the later part. It is striking that two early
model books or sheets show the copying of Byzantine models by Western
artists: the early thirteenth-century model book at Wolfenbüttel (Herzog
August Bibl. Guelf. 61. 2 Aug. 8°) and the single leaf of ca. 1200 in Freiburg
(Augustinermuseum, G 23/1a).28 But Byzantine illuminated books were also
portable – witness the Greek psalter from St Gereon, Cologne (ÖNB theol.
gr. 336) – even if most literate people in the West would not have been able to
read a Greek text.29 While there was no ‘reform style’ as such, the aban-
donment of the Ottonian tradition, which was closely identified with
Imperial patronage, and the adoption of a more Italianate style in the
newly reformed monasteries, has been seen as the outcome of the
sociocultural change brought about by the conflict between Empire and
Papacy. The concept of influence, so prevalent in art historical writing, has
been subjected in this generation to more rigorous analysis.30 Often repre-
sented as an emanation from one work of art to a successor, influence should
really be viewed from the other direction, as the choice of a successor or
inheritor which makes them an active agent rather than a passive recipient.
Thus Western artists, at the same time that they were choosing to adopt
some of the stylistic principles and formulae of Byzantium, seem generally
not to have chosen to adopt their compositions, and to have remained
faithful to Western traditions in subject-matter. There was no Western
equivalent of the Byzantine Painter’s Manual, which codified the represen-
tation of Christian subjects; attempts such as the Cistercian Pictor in
Carmine of ca. 1200 never achieved widespread authority.31 In recent years
the whole discussion of Byzantine influence has been recast. Following Hans
Belting’s conception of the ‘history of the image before the age of art’,
attention has been paid to the origins and transformation of the cult image
in its passage from Byzantium to the West.32 The switch of attention from
production to reception has enlarged the scope of enquiry. But the study of
the transformation from Hellenistic panel painting to Byzantine icons and
subsequently to the Western altarpiece has tended to diminish the attention
paid to miniature painting, and a concentration on the cult image also fails
to do justice to the range of imagery, especially narrative imagery, contained
in books.
 Decoration and Illustration 55
The renewed influence of Byzantine forms in the period from 1170 was
accompanied, first of all in England and France, by a change in the
depiction of drapery, in which the abstract patterns of the damp fold
were replaced by cascading multiple folds. The use of intense single
colours, such as vermilion or ultramarine, is now balanced by more subtle
combinations of mauve, fawn, and grey. Smaller and neater figures accord
with the reduced size of many books and the fine white outlines which are
usually considered a feature of Gothic illumination make their first appear-
ance. Towards the end of the century overt Byzantine modelling becomes
less apparent in the calm classicizing spirit of the style sometimes referred
to as ‘Transitional’ – a style which in fact differs as radically from roman-
esque abstraction as it does from the elegance of courtly Gothic.33

Embellishing the Word

It was in the twelfth century that the decorative and illustrative potential of
the initial was fully exploited. The practice of decorating initial letters had
not been characteristic of the ancient world. The emergence of this manner
of embellishing the word in the early Middle Ages has been linked both to
the growth of textual reception through private reading rather than listen-
ing to a text read aloud and to the Christian emphasis on truth revealed
through the written word of the gospels. Alexander has stressed the conflict
inherent in the decoration of initials.34 On the one hand, the legibility of
script depends upon the repetition of clearly defined forms which carry
a consistent significance: scribes were trained to write letters regularly and
distinctly. On the other hand, the decorated initial thrives on ambiguity
and variation: letters can become symbols, conveying the mystery of the
Divine Word by their very complexity.35
The most basic function of decorated initials, as of tituli, rubrics, and
punctuation, was to structure a text, aiding memorization and cueing the
process of recollection by means of which a reader engaged with a text.36
To achieve this it was important to maintain a decorative hierarchy, so that
the size and elaboration of the initial could act as a guide, communicating
the position of major and minor textual divisions even before a word was
read. The clarity of the decorative hierarchy, with the parts subjugated to
the interests of the whole, together with the endless invention displayed by
the romanesque decorated initial, can create an exquisite combination of
function and ornament even in a modestly decorated book. The initials
used to mark secondary textual divisions, or even primary divisions in less
luxurious manuscripts, are sometimes called ‘arabesque’, following the
56 martin kauffmann
term proposed by Alexander.37 They include foliage and tendrils but are
inorganic and two-dimensional: they are the ancestors of Gothic fleuronnée
initials. It has been suggested that the arabesque initial derived from the
initials of ninth-century manuscripts from Tours. Anglo-Saxon manu-
scripts, despite the strong Carolingian influence on their script and dec-
oration, did not develop this form: the main components of their
decorated initials are interlace, and bird and animal forms. It is in
Norman manuscripts of the first half of the eleventh century that we can
observe the appearance of simple scrolls, which become more complex later
in the century, and it was Norman influence which gave impetus to the
development in England.
These arabesque initials are intimately connected with the script of the
book. They appear in greatest profusion in the patristic volumes with
which twelfth-century libraries were stocked. The scribe leaves a space
for the initial to fill, whether that initial is executed by the same person or
by someone else. Sometimes a guide letter is still visible in the margin or
within the initial space. The fact that there seem to be fewer missing initials
in twelfth-century manuscripts than in those from later centuries suggests
that the initials were inserted within the monastic scriptorium as part of the
main production process. Much of their liveliness stems from the colour
contrasts, especially of red, green, blue, and violet, which often appear very
similar to those of the accompanying display headings. Under-drawing is
sometimes visible, but the final form of the initial does not always follow it.
The distinction between what is done with the pen and what is done with
the brush can be difficult to ascertain. House styles – that is, particular
features observable in the arabesque initials of manuscripts from individual
religious houses or orders – can in some cases be identified; but the limitless
variety and inventive power of these initials produced for the most part by
those living a communal religious life subject to a rule has been contrasted
with the relative uniformity of later initials produced in commercial
circumstances. Nevertheless there is some evidence of an effort on the
part of the reformed monastic orders to extend their austerity into the area
of book production, and specifically to restrain the exuberance of the
initial. A Cistercian statute of 1131 forbade the illumination of initials in
more than one colour: ‘litterae unius coloris fiant et non depictae’. This
rule was not rigorously followed in all Cistercian houses, and even when
they stuck to the letter of the rule, single-colour initials could still be very
elaborate.38
What does this interweaving of script and initials imply about the
distinction between scribes and artists? Hugo ‘pictor’, the late eleventh-
 Decoration and Illustration 57
century Norman monk already mentioned, refers to himself as an illumi-
nator but depicts himself as a scribe. This may just reflect the strength of
the tradition of the scribal portrait. But the self-portrait of the nun Guda
(Frankfurt, Stadt- und Universitätsbibl. Barth. 42, f. 110v) is accompanied
by the inscription: ‘Guda peccatrix mulier scripsit quae pinxit hunc
librum.’39 The lay artist Hildebertus is depicted in one manuscript as an
artist (Stockholm, Kungliga Bibl. A. 144, f. 34r) and in another (Prague,
Metropolitan Libr. A. 21. 1, f. 153v) as a scribe in the act of throwing his
sponge at a mouse which is eating his lunch, while his assistant Everwinus
paints an arabesque initial.40 The four principal twelfth-century scribes at
Corbie, near Amiens – Nevelo, Ingelrannus (Enguerrand), John of Amiens
the ‘one-eyed’, and Elias (Hélie) – are all recorded as having ruled, written,
rubricated, and illuminated their manuscripts.41 In general it seems fair to
assume that the closer the relation of script and decoration, the more likely
they are to have been executed by the same hand. In the later Middle Ages,
different individuals, who might be working at different times and in
different places, specialized in the execution of script, rubrics, penwork
initials, painted initials, borders, and miniatures.
Moving up the decorative hierarchy, the foliate forms of the
twelfth-century painted initial are so ubiquitous that it becomes easy not
to notice this incorporation of the natural world into the sign language of
writing. The foliage of a twelfth-century initial is different from that of
earlier examples. In Anglo-Saxon and Ottonian manuscripts, foliage stems
were arranged in loose spirals with shoots growing in all directions, termi-
nating in a turned-over trefoil or quatrefoil. From the beginning of the
twelfth century, starting in England and the region of the Meuse, increasing
emphasis was placed on individualized flowers. Typically the stem termi-
nates in a small circle from which emerges a central petal surrounded by two
smaller ones. Other petals or leaves and blossoms were added until flowers of
this kind, usually placed in the centre of foliate spirals, became the dominant
motif. In the second quarter of the century they develop into large, luxuriant
plant formations which extend in all directions like a fleshy octopus. This
type of decoration reached its maturity in England in the Bury Bible of ca.
1135,42 and became universal in Northern Europe in the third quarter of the
century (Figure 3.3). Only Italy and to some extent southern France stand
apart from this development. Art historians have struggled to explain, as
opposed merely to describe, this vocabulary. Vine scrolls and other types of
arboreal ornament have been interpreted as symbols of renewal and
regeneration in the context of Church reform. One might see the taming
of the natural world as a traditional function of ornament. The desire to
58 martin kauffmann
impose order on a chaotic universe is expressed in twelfth-century literature:
the Cosmographia of Bernard Silvester opens with Nature’s complaint of the
state of chaos in the universe. Recently Weinryb has taken this idea further
by focusing on the account of creation in Plato’s Timaeus, in which the
divine artifex creates the four elements not ex nihilo but out of primordial
matter that was in a chaotic, amorphous state.43 In the Latin translation of
Calcidius, the Greek term hyle for the primordial matter was rendered as
substantia in the text but as silva in the commentary. Thus the raw material
of the foliate initial may be understood as the primordial matter of Creation,
shaped by the human artifex into accordance with the Word of God.
The twelfth century also saw the introduction, enmeshed in such foliage
scrolls, of animals and of clambering elongated human figures, often
engaged in combat with dragons or hybrid monsters. Plant scrolls inhab-
ited by birds or beasts had originally been a classical motif which
reappeared in the margins and frames of manuscripts from the
Carolingian period. The animals or hybrid creatures themselves could be
derived from Near-Eastern textiles, or from astrological illustrations, calen-
dar scenes, or bestiaries. Such inhabited scrolls were adapted to the decora-
tion of initials already in the tenth century and more commonly in northern
French manuscripts of the early eleventh century. What was new was the
regular introduction of human figures into the foliage decoration. An early
example appears in a manuscript from Mont-St-Michel, of the third quarter
of the eleventh century (Avranches, Bibl. mun. 72, f. 151r), in which a man
with an axe is attacking a lion within the body of an initial P,44 and it is seen
at its most inventive in the Giant Bible from Jumièges of the last quarter
(Rouen, Bibl. mun. 8).45 It was in England that this type of initial was to
reach its peak, though there are examples from all over Northern Europe,
especially Flanders, the Meuse, and the Rhineland. The distinction made by
modern scholars between decorated and historiated initials is not always
clear. Many of the initials which feature human figures do not seem to
represent an identifiable narrative or history – hence the use of the additional
term ‘figured’ or ‘figurative’. Some of these initials may have been intended
to represent the eternal struggle of human beings, trapped in a hostile world
of sin, searching for salvation. Speaking of a different medium, that of stone
carving, the Cistercian Bernard of Clairvaux, writing to William of
St-Thierry in ca. 1125–6, castigated the distraction caused by hybrid creatures
from the claustral concentration of the religious life.46 His rhetorical ques-
tion, ‘What signifies these ridiculous monsters?’, suggests that to
a contemporary eye they lacked symbolic meaning – though the alignment
of dragons with the Devil is a feature of Christian writing from biblical
 Decoration and Illustration 59
times, and the fact that Bernard was exercised to condemn the capitals which
tempted monks to read in the marble may be taken as proof that there were
some who did precisely that. Monks, who were used to reading long
expositions of biblical texts, meditating on the implications of the text and
using it as a stimulus for prayer, might surely have searched enigmatic images
in their books for spiritual edification.47
From about 1170 a new form of decorated initial developed in France
and England. There was clearly a growing taste for minuteness of form.
The foliage becomes thinner and more stringy, the spirals more regularly
and tightly wound, the ground now usually of burnished gold. Many of the
old floral forms have been retained, including the ‘octopus’ flower, but
they have generally become smaller and are often surrounded by groups of
white dots. The principal animal presence now consists of small white lions
which inhabit the scrolls in profusion. The initial itself is placed on
a framing panel, usually of blue, red, or gold, often itself decorated with
small circles or dots. Such initials, neater and more refined but perhaps less
vigorous and inventive than their predecessors, feature in the glossed
biblical books commercially produced in considerable numbers in Paris
and exported throughout Europe.48

Illustrating the Text

In surveying the illustration of texts in the twelfth century, we encounter
on the one hand persistent traditions, and on the other an expansion and
enrichment in arrangement and subject-matter. Though the detailed study
of iconographical prototypes – tracing the descent of individual motifs and
of whole compositions – is no longer considered sufficient explanation of
the meaning of each new pictorial event, this is nevertheless an art with
strong patterns of expectation, in which context the invention of new
forms, or even small departures from tradition, may be deliberate and
meaningful.49 In recent years the emphasis has shifted from the study of
iconography as a self-contained system to its role in the rhetoric and
theology of images.50 In fact iconography has proved a resilient concept
and has begun to be applied to new fields such as colour.51 Heslop has
pointed to the establishment of a visual vocabulary of class stereotypes,
based on classical paradigms, in English manuscripts.52
To begin with the book greatest in authority (and also in size): the
twelfth century saw the production of magnificently illustrated monastic
Giant Bibles, reviving a Carolingian tradition.53 These lectern books, often
in two volumes, were read in both church and in refectory, and must have
60 martin kauffmann
been impressive symbols of the status of the communities which possessed
them. They may have originated in Italy, but very soon they were also
being produced in the valleys of the Meuse and the Rhine. The one
surviving volume of the Lobbes Bible of 1084 (Tournai, Bibl. du
Séminaire 1) is perhaps the first to feature a full set of historiated initials,
in many cases excerpted from larger cycles such as that of the Carolingian
bible in S. Paolo fuori le Mura, Rome, but simplified, compressed, and
ingeniously fitted into the confines of the letters.54 Many of these scenes,
such as King David and the Amalekite for II Samuel, are subsequently
found in bibles throughout the twelfth century. From about the 1130s,
these Giant Bibles are found all over Europe, with several of the most
outstanding examples produced in England. Some or all of the biblical
books were provided with a miniature or a historiated initial, whose subject
was designed to summarize the content of the book (such as the Crossing of
the Red Sea for Exodus) but was often drawn from the events occurring
near its beginning. Certain Old Testament scenes, such as the Ascension of
Elijah for II Kings, were chosen for their typological significance as
a prefiguration of the New. Some manuscripts also have a full-page frontis-
piece to some of the biblical books. As in the case of the Tree of Jesse
illustration for the Book of Isaiah in the Lambeth Bible (f. 198r), these
frontispieces can go beyond the illustration of the biblical narrative to
become vehicles of complex typological or theological doctrines. Likewise
the miniatures of the Floreffe Bible encapsulate complex biblical exegesis
and are closely related to the pictorial diagrams of contemporary Mosan
metalwork (Figure 3.2).55
But in England especially, the book which most commonly received
extensive illustration was not the whole Bible but the Psalter. Recited both
by monks and by the laity, the Psalms were the prime devotional texts of
the earlier Middle Ages. The text of the Psalms is often preceded by
a liturgical calendar which can contain pictorial cycles of the occupations
of the months and the signs of the zodiac, either within the KL monogram
itself (for the Latin kalends), which stands at the head of each month, or in
separate roundels. The monastic text of the Psalter was divided first into
eight sections, marking the beginnings of the parts to be read at matins
each day and at vespers on Sunday: that is, at Psalms 1, 26, 38, 52, 68, 80, 97,
and 109 (Vulgate numbering). At the same time the threefold formal
division at Psalms 1, 51, and 101, which had Insular origins, was also
retained, the two systems combining to make a tenfold division. These
divisions were marked by large decorated or historiated initials.
A consistent choice of subjects begins to appear in the initials of some
 Decoration and Illustration 61
twelfth-century psalters, but it is not until the following century that
a regular series becomes standard. The subjects themselves are usually
not literal illustrations of the text, but were suggested either by the
Psalm’s titulus or by its opening verse. The small group of manuscripts
containing a literal illustration to every Psalm was directly inspired by the
presence at Canterbury of an extraordinary exemplar, the Carolingian
Utrecht Psalter (Utrecht, Bibliotheek der Rijksuniversiteit 32).56
The initials in the St Albans Psalter (Hildesheim, Dombibl.
St. Godehard 1) are exceptional in responding to the allegorical, tropolo-
gical, and eschatological exegesis of the Psalms.57 Several psalters also have
a cycle of full-page miniatures prefacing the text. This practice had begun
in eleventh-century England, but only became common in the twelfth.
As well as narrative subjects drawn from the Old and New Testaments,
especially from the life of Christ, these cycles sometimes also include
images of the Tree of Jesse, King David, the Virgin and Child, and
Christ in Majesty. The scenes do not illustrate the text of the Psalms
directly, but can be related to the Christian typological reading of the
Psalms as messianic prophecies with King David, their presumed author, as
the precursor and ancestor of Christ. Thus in England it was the Psalter,
not the complete Bible, which carried the largest cycles of New Testament
scenes. This is less universally true on the Continent, though there are
examples such as the early thirteenth-century Psalter of the Landgrave
Hermann of Thuringia (Stuttgart, Württembergische Landesbibl., HB II
24).58
In gospel books, each gospel may begin with an Evangelist portrait and
a large decorated initial; decorated canon tables preceding the biblical text
are found only occasionally after 1100. Only very rarely, as in the excep-
tional typological scheme of the Gospels of Henry the Lion, were gospel
books furnished with extensive picture cycles. Gospel lectionaries contain-
ing the passages to be sung at Mass through the liturgical year occasionally
contain New Testament miniatures. The tradition of illustrating the
Commentary on the Apocalypse by Beatus of Liébana was peculiar to
Spain, where the gradual assimilation of romanesque style to Mozarabic
forms may be traced in a series beginning with the manuscript from Santo
Domingo de Silos of the late eleventh century (BL Add. 1695).59 Among
liturgical books, only the Mass books (sacramentaries and missals) were
regularly illustrated, usually with a large miniature of the Crucifixion
before the Canon of the Mass; but the abbey of St Michael at
Hildesheim produced two outstanding missals with complex typological
schemes, the Ratmann Missal of 1159 (Hildesheim, Dom-Museum DS 37)
62 martin kauffmann
and the Stammheim Missal of the 1170s (Los Angeles, J. Paul Getty
Museum 64).60 At the same time we can observe, sometimes in a female
context, the beginning of the illustration of Office books such as the
breviary.61 Few illustrated manuscripts announce the manner of their use
as clearly as the Exultet Roll. These large-format manuscripts from south-
ern Italy contain the prayer for the blessing of the Paschal candle sung by
the deacon on Holy Saturday. Uniquely, its illustrations face the opposite
way to the text so that it could be unrolled over a lectern and hang down
with its pictures to be seen the right way up by the congregation. As well as
religious and liturgical scenes such as the Crucifixion and the Passage of the
Red Sea, these rolls contain a representation of bees and a beehive in
honour of the providers of the wax for the candle.62
The handsome copies of the biblical commentaries and other works of
the Church Fathers, with which monastic and cathedral libraries were
stocked, generally contain only decorated initials. Illustrative traditions
were established only for one or two patristic works, such as the Moralia in
Job of Gregory the Great and the De civitate Dei of Augustine. Similarly,
the works of medieval theologians and commentators on Scripture, if
illustrated at all, usually attracted only an author portrait and one or two
historiated initials. The Prayers and Meditations of Anselm of Bec and
Canterbury provides a rare example of such a text attracting an extensive
scheme of illustration.63 Copies of the text had been sent to Anselm’s
monastic and lay friends during his lifetime, but it is uncertain whether
these early copies already contained pictures. From about 1070, all over
Europe, illustrated saints’ Lives were produced by communities which thus
proclaimed the virtuous lives, miracle-working powers, and continuing
protection of their patrons; they had seldom been illustrated before.64
The picture cycles of these manuscripts are closely related, just as the
authors of their texts sought to bring the lives of their subjects into
conformity with the pattern set by Christ. These manuscripts are all
dedicated to single saints or to a small group of related ones. By contrast
some passionals, which contain shorter saints’ Lives arranged in the order
of the liturgical year, contain a historiated initial depicting each saint, or
even an episode from the saint’s life.65 In canon law a tradition of illustrat-
ing Gratian’s Decretum spread from Bologna to Paris.66
Perhaps no text composed in the twelfth century can rightly
be described as secular. The illustrated twelfth-century encyclopaedias,
for example, the Liber Floridus written by Lambert, canon of St-Omer
(Ghent, Universiteitsbibl. 92), and the Hortus Deliciarum of Herrad
of Hohenbourg in Alsace (destroyed in 1870, but partially
 Decoration and Illustration 63
reconstructable), attempt to assimilate all aspects of monastic learning into
a Christian world picture.67 Monastic chronicles and world histories are
most often illustrated (if at all) by ruler portraits, sometimes in genealogical
trees to underline the importance of dynastic pedigree. The History of the
Two Cities by Otto of Freising (Jena, Universitätsbibl. Bos. 9. 6) and the
Liber ad honorem Augusti by Peter of Eboli, celebrating the conquest of
Sicily by the Emperor Henry VI (Bern, Burgerbibl. 120), are rare examples
at this period of the invention of narrative cycles of illustration for
historical texts.68 But most illustrated twelfth-century secular manuscripts
contain texts which had their origins in the classical world, either as Latin
works or as late antique Latin translations of Greek works. Sometimes
continuities between antique and medieval picture cycles can be identified,
though the pictorial tradition in most cases seems to extend only as far back
as the fourth or fifth century CE. In such instances the pictures in the
twelfth-century manuscripts usually seem to have been adapted not
directly from late antique exemplars, but from Carolingian or Anglo-
Saxon intermediaries. Of literary, historical, and philosophical works,
only the comedies of Terence and the Psychomachia of Prudentius were
regularly provided with cycles of pictures. The battle of virtues and vices,
composed by Prudentius in the fifth century CE, depicts allegorical figures
of classical origin, but the text is Christian.69 The illustrated Terence
manuscripts, with their scene-by-scene illustrations reflecting the masks
and even perhaps the gestures of the Roman stage, represent the end of
a monastic antiquarian tradition – though the original sketchy three-
dimensionality has been replaced in the twelfth-century manuscript prob-
ably produced at St Albans (Bodl. Libr. Auct. F. 2. 13) by the firm outlines
and stylized draperies of romanesque art.70 In later centuries the illustra-
tions are clothed in knightly Gothic garb to cater to the tastes of a secular
readership. The classics are present even in the vernacular: one of the few
examples of a romanesque cycle of illustrations to a vernacular text is the
early thirteenth-century manuscript from Regensburg of the Middle High
German version of the Aeneid by Heinrich von Veldeke (Berlin, Staatsbibl.
germ. fol. 282).71
The other illustrated ‘secular’ texts are mostly technical and scientific,
including herbals, bestiaries, and treatises on astronomy and astrology.
Of these, the bestiary (the popularity of which was to reach its height only
in the thirteenth century) had travelled furthest from its late antique roots;
the animal lore in its text was by now almost entirely in the service of
Christian moralizing themes, and the picture cycles are medieval in
inspiration.72 The main purpose of the herbal was to identify plants and
64 martin kauffmann
to describe their medicinal properties; illustrations were integral to the
identifications, though in most cases artists did little more than copy the
conventionalized pictures they found in their exemplars.73 Illustrations
accompanying medical tracts, such as drawings of cautery figures, were also
traditional.74 The illustrations were designed to show the points on the
body where the cautery iron should be applied to cure specific ills; the
purpose was to readjust the imbalance of bodily humours, not merely to
prevent the spread of infection. The occasional representation of opera-
tions for the removal of cataracts and nasal polyps, found in conjunction
with the herbal, are not, however, detailed enough to have been of any
practical application. The monastic study of the natural sciences is attested
by the survival of illustrated astronomical manuscripts. The texts, mostly
based on Cicero’s verse version of the Greek of Aratus with later accretions,
were illustrated by pictures of the planets, constellations, and signs of the
zodiac. The twelfth-century versions were derived from Carolingian mod-
els which themselves were based on late antique exemplars. By these means
images of classical gods and mythological figures entered the consciousness
of the medieval monastery. In one instance (Bodl. Libr. Bodl. 614)
illustrated astronomical and astrological texts are found in conjunction
with the illustrated Marvels of the East, descriptions of natural wonders and
monstrous races going back ultimately to ancient Greek descriptions of the
fabulous peoples of India.75
The twelfth-century love of classification found its pictorial expression in
diagrams illustrating books of natural science, the texts of which were often
excerpted from the works of earlier medieval authors such as Isidore of
Seville and Bede.76 The balanced and elegant construction of these figures
could be used to depict the harmonious relations between microcosm and
macrocosm in the constitution of the universe according to Christian
cosmology: the four seasons, elements, humours, and ages of man, for
instance, or the seven planets, sacraments, and liberal arts.77 Such diagrams,
often in ink outline, drew on the formulations of twelfth-century texts such
as the Cosmographia of Bernard Silvester and the Elucidarium of Honorius
Augustodunensis: they combine the biblical idea that God created man in
His own image with the ancient teaching (found in Plato’s Timaeus) that
mankind consists of the same elements as the material world. Some of the
monastic houses of southern Germany, such as Hirsau and Zwiefalten in
Swabia or Prüfening and Regensburg in Bavaria, specialized in the produc-
tion of such drawings (Figure 3.4).78 The diagrams were designed to impress
the relations between different parts of a subject on the memory of the
reader. Indeed, the organization of space in many romanesque illustrations
 Decoration and Illustration 65
may reflect the same purpose, with geometric or architectural forms provid-
ing a grid within which images, often identified by inscriptions, are grouped
hierarchically around a central motif or figure. Nordenfalk’s classic defini-
tion of romanesque style contrasted the miniature of the ascension of
St Amand in the first illustrated Life of the saint of ca. 1070–80
(Valenciennes, Bibl. mun. 502), still essentially Carolingian in character,
with the same scene in the Life of the mid twelfth century (Valenciennes,
Bibl. mun. 501), now characterized by strict symmetry, frontality, and
a compartmentalized layout.79 Caviness has explored the way in which
such governing principles created a visual syntax for the representation of
divine order in romanesque art more widely.80 Thus the word ‘illustration’
does not begin to do justice to the variety of relationships we can observe
between the texts and images which were juxtaposed with each other in the
twelfth century.

Notes
1. Mütherich and Dachs 1987, no. 26.
2. Kötzsche 1989.
3. Cahn 1982, cat. no. 46.
4. Ibid., cat. no. 28.
5. Mynors, Durham, 32–45.
6. Oakeshott 1981.
7. Dodwell 1993, 278–9 and pl. 276.
8. Gameson 2001.
9. Gullick 2006.
10. Alexander 1992, figure 28.
11. For example, Dodwell 1993.
12. Clanchy 1993, esp. 224–52.
13. Alexander 1992, figure 17.
14. Thomson, The Bury Bible.
15. Kauffmann 1975, no. 34.
16. Riedmaier 1994; Shepard 2007.
17. Cahn 1996, no. 79.
18. Clarke 2001.
19. Alexander 1992; Barral i Altet 1986–90; De Hamel 1992.
20. Evans 1969; Holcomb 2009.
21. Lawrence-Mathers 2003, 89–108.
22. Oakeshott 1981.
23. Petzold 1990.
24. Hourihane 2008.
66 martin kauffmann
25. Kauffmann 1975 (England); Cahn 1996 (France); Klemm 1980–8 and Butz/von
Borries-Schulten 1987 (Germany); Murano and Saggese 2005. Fingernagel
2007 includes a series of essays arranged by country.
26. Camille 1985.
27. Demus 1970.
28. Buchthal 1979; Scheller 1995.
29. Legner 1985, vol. 2, cat. E 41.
30. Lowden and Bovey 2007.
31. Wirth 2006.
32. Belting 1994.
33. Hoffmann 1970.
34. Alexander 1978, 87.
35. Hamburger 2014.
36. Carruthers 2008.
37. Alexander 1978.
38. Glorieux-De Gand 1990; Lawrence 1995; Reinecke, Reinecke and Tivig 1998.
39. Smith 1997, figure 6.
40. Alexander 1992, figs. 18–9.
41. De Mérindol 1976, 1. 431–45.
42. Thomson, The Bury Bible.
43. Weinryb 2013.
44. Alexander, Mont St Michel, pl. 12a.
45. Cahn 1982, cat. no. 106.
46. Rudolph 1990.
47. Heslop 1986.
48. De Hamel, Glossed Books, esp. ch. 4.
49. Cassidy 1993.
50. Hamburger and Bouché 2006.
51. Petzold 1999.
52. Heslop 1990.
53. Cahn 1982; Kauffmann 2003, 73–104.
54. Cahn 1982, cat. no. 48.
55. Ibid., cat. no. 46.
56. Van der Horst, Noel, and Wüstefeld 1996; Eadwine.
57. Bepler, Kidd, and Geddes 2008; Pächt, Dodwell, and Wormald 1960.
58. Heinzer 1992.
59. Williams 2002–3.
60. Teviotdale 2001.
61. Seeberg 2002.
62. Cavallo 1973.
63. Pächt 1956.
64. Abou-El-Haj 1994; Hahn 2001.
65. Michon 1990.
66. Melnikas 1975.
67. Derolez 1968; Green et al. 1979.
 Decoration and Illustration 67
68. Kölzer and Stähli 1994.
69. Katzenellenbogen 1939; Norman 1988; Stettiner 1895–1905.
70. Jones, Webber, and Morey 1931.
71. Henkel and Fingernagel 1992.
72. Clark 2006.
73. Collins 2000.
74. MacKinney 1965.
75. James 1929.
76. Murdoch 1984.
77. Saxl 1957.
78. Boeckler 1924; Mütherich and Dachs 1987.
79. Grabar and Nordenfalk 1958, 182–9.
80. Caviness 1983.
 chapter 4

Scribes and Scriptoria

Rodney Thomson

This chapter focuses on a single broad, general question: what, if

anything, is new, characteristic and significant about the activities
and organization of scribes and scriptoria across Western Europe in
the Long Twelfth Century? I believe that there was something char-
acteristic and significant about these things, and that it can be
described under three heads: first, there was a sizeable increase in the
numbers of both scribes and scriptoria; secondly, there was a high
average level in the quality of scribal work; and finally, there was
a gradual movement from local and regional variety of scripts towards
international uniformity, in other words the emergence of what we call
Gothic, the first pan-European script.

Scribes
But first of all, in keeping with the aims of this project, we need a working
definition of ‘scribe’. That definition has to encompass more than just
‘anyone who could write’. In the first place, we need to consider those
persons who wrote in books, not, or not just, documents. In parentheses
and speaking very broadly, at the beginning of our period the same persons
usually wrote both, and in the same sort of script; by the end, due to the
huge proliferation in the creation of administrative documents in both
Church and State, both personnel and script-type had sundered. Secondly,
it is appropriate to focus on those persons who wrote substantial amounts
of text and in more than one book, in other words, persons who were, at
least to some extent, dedicated copyists, active over a stretch of time. And
finally we need to focus on those who were trained to write well and who
sustained the standard and style they were taught through reasonably
constant practice. At first sight one might think that these criteria are too
restrictive, allowing us to discuss only a small minority or elite among the
copyists of the day. But this is not so. After taking these criteria into
68
 Scribes and Scriptoria 69
account we are still able to talk about the sort of copyists who are most
heavily represented in the surviving books. The point can be made by
observing the differing levels of competence revealed in the writing found
in surviving mortuary rolls, for instance the most famous of them all, that
of Vitalis, abbot of Savigny (d. 1122), with its tituli from more than 200
houses in France and England.1 Among these entries we find hands that
exemplify the ‘house style’ recognizable from books produced in the local
scriptoria, but we also find hands that do not, and that can only be
described as unpractised or even uncalligraphic.2 It is possible to find
‘bad’ hands in books of the period, but they are relatively scarce.3
Numbers and Locations: As is well known, a, perhaps the, primary motor
for much of the creativity within the period was religious reform, defined
in terms of a particular interpretation of the monastic ideal. The revival of
traditional Benedictine monasticism and the appearance of new religious
Orders led to a considerable increase in the total monastic population of
Western Europe, and in the number and size of individual religious
communities.4 This increase in population and communities was accom-
panied by a concomitant increase in the production of books, as new
libraries had to be stocked, and old ones enlarged and refurbished.
Secular cathedrals are not centre stage in this process because they were
not nearly so numerous, and few were new foundations. Most scribes were
monks (or regular canons), and therefore worked within and for their local
communities. It follows that the number of scribes, which perhaps peaked
around mid-century, bears an almost direct relationship to the number of
new, enlarged and reformed convents. A baseline figure would be one
scribe per community, but this is almost certainly too low. On the one
hand, it would allow for the fact that small communities, especially cells of
larger ones, might not make their own books at all but commission them
from the mother house or from a community that was nearby and wealthy.
On the other hand, we know of large houses where a dozen or so scribes
were active over half a century or more, so maybe a multiplier of two or
three might be thought more reasonable. About the considerable number
of new female convents, we are mostly very poorly informed. It used to be
assumed that their libraries were small and that nuns did little
(non-liturgical) reading or copying; however, recent work has revealed
a substantial amount of book-making at some convents,5 so perhaps the
multiplier for female scribes should be about the same as for males.
In parenthesis, we should note that neither secular authorities (emperor,
kings, magnates) nor the papacy play a significant role in this story.
70 rodney thomson
It seems to me one of the puzzles of the twelfth century, little commented
on, that emperors and kings in particular failed to emulate their
Carolingian, Anglo-Saxon and Ottonian forebears by maintaining
a court library or commissioning great liturgical books.6 In sum, by the
third quarter of the century there were probably several thousand scribes
operating contemporaneously across Western Europe.
Levels of Professionalism: One of the reasons for the high average level of
proficiency among twelfth-century scribes was presumably the monastic
ethos: writing was sometimes explicitly described and prescribed as
a spiritual exercise;7 books were to be carefully written because they were
the bearers of texts conveying eternal truths. A particularly poignant
example is the prefatory miniature in a copy of Isidore’s Etymologiae
expertly written, shortly before 1165, by the monk Swicher for his house
of Prüfening (near Regensburg). It shows the dead Swicher’s successful
entry into Heaven after his book has been weighed in a scales in the
presence of Christ – a result helped by the fact that his book is depicted
in a heavy treasure binding.8 Such a high level of proficiency implies
difficulty in distinguishing the well-trained monastic scribe from the
paid (and doubtless also trained) professional, and indeed the relative
numbers are impossible to calculate.9 Some scribes who wrote monastic
books were fully professional, making a living from their work, often
itinerant and prepared to cross oceans and borders. For example, the
writing (and artwork) of Eadwine, monk of Christ Church, Canterbury,
can hardly be judged inferior to that of the younger Manerius, the son of
a Canterbury mercer who was not a monk and made a living as a scribe and
an artist on both sides of the Channel.10 The scribes ‘sought from afar’ by
Abbot Paul of St Albans (d. 1093) to copy fine liturgical books were clearly
not monks, and would have been paid for their work.11 Such scribes might
be used to supplement the labours of members of the house, as they were at
Abingdon under Abbot Faricius (d. 1117).12 The Liber ordinis of
the Augustinian canons of St-Victor in Paris includes instructions for the
operations of the scriptorium, including the hiring of paid scribes from
without. The brothers who were active in writing could be excused by the
abbot from participation in the Divine Office.13
During the second half of the century, as is well known, the balance, as
far as we can perceive it, begins to change. For the first time, the phrase
‘book trade’ can be used, and when used of this period is focused on
Paris.14 Through the rest of our period the balance continued to tilt in
favour of lay scribes and stationers working in the larger towns, especially
 Scribes and Scriptoria 71
those associated with universities. In the earliest stages of this process the
professional usually only shows up, if at all, when writing for an institution,
but towards the end of our period we can find ourselves a long way from
the cloister or the monastic ethos. Consider the professional scribe
Raulinus of Fremington, an Englishman who worked in Paris and
Bologna. The astonishingly frank and quite improper monologues which
he inserted, unsignalled, into a great bible that he copied led to him being
described by M. A. and R. H. Rouse, in their entertaining study of him, as
‘lustful, coarse, and self-absorbed’.15 One such passage, inserted after the
prologue to Proverbs, will serve for all: ‘O you whore Meldina, you have
always deceived me. When my purse feels full, you embrace me and kiss
me. But when it lies empty, then there is neither a kind look nor love.
In you there never was, is, nor shall be loyalty or truthfulness.’
Status: What level of society did professional scribes occupy, and what
measure of esteem did all scribes enjoy? Can anything be inferred about
this from the notorious anonymity of most of them? Was it an instance of
monastic humility? Does it signify that neither they nor others thought
their highly skilled work valuable? Probably the general anonymity means
none of these things. We do not know the names of most monks in any
given community in our period; we do not know the names of most artists,
even major ones. In fact, instances of praise and appreciation for scribes,
and even some instances of self-esteem, are comparatively numerous: one
might think, for example, of the inscription accompanying the famous self-
portrait in the Eadwine Psalter, astonishing if it is indeed autograph:
‘Eadwine princeps scriptorum ego’.16 The historian Orderic Vitalis, writ-
ing ca. 1114/15, describes a certain William, oblate of the Norman house of
St-Évroult, as ‘a distinguished scribe and illuminator of books. The works
executed by his own hands for reading and singing are still models which
encourage us to put away idleness and follow his example.’17 The house
chronicle of St Albans Abbey talks of Abbot Paul seeking from afar
‘electissimos scriptores’ (‘the choicest scribes’).18 Among the scribes of the
books written at Mont-St-Michel in the course of the eleventh century are
no fewer than fifteen who name themselves in their colophons.19 Even
more remarkable are the sixteen surviving books made at and for
the Augustinian canonry of Cirencester (founded in 1131), in the second
and third quarters of the twelfth century, bearing contemporary inscrip-
tions naming the canon scribes, and the man who was abbot at the time of
writing.20 Most striking of all, moving even, is the repeated praise, both
contemporary and over several successive centuries, by the monks and nuns
72 rodney thomson
of the Benedictine double house of Wessobrunn, seventy-seven kilometres
south-west of Munich, for Diemut, the inclusa who made so many books
for the house in the first half of the twelfth century; I say more of her in
what follows. She is named and the books she wrote listed, in two in-house
booklists, one made not long after her death, the other ca. 1200. Her
memory was revived again, each time with growing veneration, in the
thirteenth, fifteenth and eighteenth centuries, by which time she had
attained the status of a local saint.21 A different kind of evidence is provided
by the well-known pen-drawing of (and presumably by) the professional
scribe Hildebert. I do not think it has been observed how well dressed he is,
suggesting a man of wealth and rank, certainly not any kind of artisan.22
Training: Most scribes represented in surviving manuscripts had been
trained, but we know little about the training unless it was in-house.
Young persons might be trained to write within the context of a school,
usually attached to a religious community, and might go on either to enter
the house or to make a career as a lay professional. Most of what we see
today is the near-finished or finished product, by scribes already well
trained, in complete books. A very precise, detailed and illuminating
study of this is Aliza Cohen-Mushlin’s on the scriptorium of
Frankenthal, an Augustinian house near Worms in the Rhineland,
founded in 1119. A new house such as this one had to build a library
from nothing, so a special effort was called for. From Frankenthal survive
twenty-six twelfth-century manuscripts made locally, the work of more
than sixty scribes active between ca. 1145 and 1200. These were obviously
(in some cases demonstrably) canons, and must have constituted most of
the population of the house over that period. At any one time within these
dates it seems as though between three and fourteen scribes were working
simultaneously, that is, collaborating on the same book or group of books.
The Worms-Frankenthal Bible, made in 1148, was worked on by eleven
men: four of them wrote the text and delineated the initials; another seven
coloured them in, but these colourists were apparently itinerant profes-
sionals, probably from Cologne; the more modest decoration in the other
manuscripts was made locally. Cohen-Mushlin was able to distinguish
between teachers and pupils among the scribes, and between different
specializations; for example, some scribes wrote most of the text, others
did only correction and rubrication. Care was taken to achieve a high level
of uniformity, the senior scribes writing exempla, passages of script at the
beginning of a text or subsection, which the younger scribes were to
imitate. There is also evidence of scribes being selected for further work
 Scribes and Scriptoria 73
and training by being made to write short passages of text as samples.
The same sort of learning on the job can be glimpsed at an earlier date, at
the Benedictine house of Mont-St-Michel, where passages in one of its
books, now Avranches, Bibl. mun. 128, ‘look as if they were written a page
or two at a time for practice by different scribes’.23
Scribal Literacy: There is some evidence that twelfth-century scribes were
generally literate, or at any rate more so than many of their Merovingian
and Carolingian counterparts. This perhaps follows naturally from the
fact that so many of them were monks, nuns and regular canons, generally
taught to read, and also needing to read text that made sense. The evidence
flows, then, from the accuracy of the texts they produced, beginning with
correct word separation, which makes an eye-catching contrast with many
books written in the Carolingian era. Another indication is the prevalence
of correction, often carried out by means of a close reading of the text,
perhaps against the exemplar, by the scribe or another person, who would
enter copy in the margin for insertion into the main text. Finally, there is
the evidence of the errors themselves. Every scribe in any era makes
mistakes, but the range of possible errors committed by literate scribes is
quite different from that made by scribes who are either illiterate or at least
copying mechanically. It is also the case that scribes sometimes corrected
text that was demonstrably wrong or not clearly legible in the exemplar.24
This evidence, that twelfth-century scribes generally understood the text
they were copying, seems to apply to both males and females.25
Working Practices: Before copying could begin on the prepared parch-
ment, an exemplar to copy from had to be provided. How this was done
can only be demonstrated anecdotally, but it appears that there was no set
pattern. A scribe might be sent to the place where the exemplar was held
and the copying done there. A scribe where the exemplar was held might be
engaged to do the work.26 Or an exemplar might be borrowed by the
community wanting to copy it. There are even cases in which an exemplar,
of unknown origin, circulated over a wide geographical area, copies being
made of it at a number of localities. For instance, Hereford Cath. O. III. 2,
a ninth-century west Frankish copy of patristic bibliographical works, was
brought to England in the late eleventh century, becoming the ancestor of
a score of copies until it came to rest at Hereford by ca. 1150.27 This kind of
process can usually only be revealed by painstaking collation of existing
copies, a procedure which has been carried out sufficiently often to show
that it has great potential.28
74 rodney thomson
How many scribes typically wrote in a single book? The range is of
course from one to many, but some generalizations can be made. Roughly
speaking, the higher the quality sought for the book, the smaller the
number of scribes who worked on it. In this respect the Worms Bible
mentioned earlier was rather exceptional, though all of its scribes wrote
well and the impression is of considerable uniformity. But most luxury
liturgical books were written by one scribe or a very few scribes. This is the
case, for instance, with great bibles such as the Carilef, Bury, and Lambeth
Bibles, and the Bible of Stephen Harding.29 The average monastic book, as
far as there is such a thing, was usually the work of no more than two or
three hands.30
How much did scribes participate in other aspects of book-making apart
from writing? They certainly carried out those tasks most intimately related
to their main work, such as folding, pricking and ruling the parchment.
There is evidence that they could participate in at least some of the other
processes that prepared the parchment for writing on, and that they did at
least the simplest coloured initials, whether painted or made with a pen.
Long ago Jonathan Alexander studied the minor (‘arabesque’) initials found
in English books between ca. 1120 and ca. 1175, surmising that these were
typically the work of the scribes who wrote text.31 Additional evidence is
supplied by contemporary representations of persons engaged in both writ-
ing and painting. One of these is the Norman Hugo Pictor, who both wrote
and decorated late eleventh-century manuscripts at Bayeux, Durham and
Exeter.32 His nickname, ‘Pictor’, suggests that he was primarily regarded as
a decorator, but in a miniature made by himself he is shown simultaneously
painting and writing, and his surviving scribal work is well known. Another
example is the already mentioned (presumably German) scribe/artist
Hildebert (ca. 1136), who figures in two pen-worked miniatures, together
with his apprentice Everwin.33 In one miniature he is writing, in the other
painting (and titled ‘H. pictor’). In the one showing him writing Everwin is
painting the ornament for an ‘arabesque’ initial. In the case of
the Augustinian canons at Frankenthal, Cohen-Mushlin has been able to
show that some of the scribes specialized in the pen-and-ink outline of
elaborate initials.34 One of the most famous and often reproduced contem-
porary representations of book-making is the full-page frontispiece to
Bamberg, Staatsbibl. Patr. 5, made ca. 1130 at and for the local Benedictine
community of St Michael.35 Within the rectangular frame are disposed ten
medallions showing the various stages of the making of a manuscript book.
Each medallion encloses the bust of a person engaged in a particular task,
from the preparation of the parchment to the binding of the completed book
 Scribes and Scriptoria 75
(plus teaching from it). While there is no way of knowing how many
individuals were intended to be involved, it is clear that all the stages were
envisaged as being pursued in-house, for all the persons depicted are ton-
sured and appear to be wearing monastic garb.
In an important chapter, Michael Gullick has investigated the speed at
which twelfth-century scribes wrote. This is not the same as the more
interesting question, which he also discusses, of how long it took to
complete the writing of a substantial book.36 Both are difficult to ascertain
except in those rare cases in which a scribal colophon tells us the answer.37
Clearly one of the variables is the amount of time allocated to the task; this
was presumably less per diem for a monk (calculated nonetheless at
a healthy possible maximum of five hours) than for a professional, dedi-
cated scribe. Related to this issue is the length of a scribe’s career, some-
thing rarely considered by those attempting to date manuscripts. Common
sense suggests that a scribe who lived to the age of seventy might have had
a writing career of as much as fifty years. This should influence, more than
it does and in the direction of caution, our attempts to date manuscripts on
the basis of palaeography alone.
Female Scribes: All experts agree that is impossible to tell the difference
between male and female scribes.38 The first question that usually occurs to
anyone interested in this area is: who made books for female communities?
It turns out that, at least in some parts of Europe, women not only made
their own books but also made books for monks, especially in the case of
double houses. The best evidence for this comes from the Germanic realm
and it is abundant: much of it has been gathered by Alison Beach studying
female scribes in southern Bavaria (an area which includes not only
modern Bavaria but also Swabia and Austria).39 Take, for example, the
Carolingian foundation of Wessobrunn, since the mid-tenth century
a community of secular canons, re-established as a Benedictine monastery
at the opening of the twelfth century.40 About 1138 a female convent was
established nearby, and over the next two or three decades, influenced by
the style of reform emanating from Hirsau, the communities merged to
become a double house. At an early stage of this process we learn of an
inclusa, Diemut, who was apparently at first a local solitary, soon joining
the women’s community. The details of her life are very obscure; even the
date of her death has been put within the wide range ca. 1130–ca. 1150. But
of her work we know a good deal. Diemut was a more than competent
scribe, copying both liturgical and library books for the combined com-
munity. The two early in-house booklists mentioned earlier credit her with
76 rodney thomson
copying no fewer than forty-seven volumes, of which fourteen survive
(Figure 4.1). At least two other female scribes collaborated with her and
with each other, as did a local monk named Ludwig. By the late twelfth
century Wessobrunn seems to have possessed a book-collection of 155
volumes; Diemut’s copying accounted for nearly one-third of them.41
At opposite ends of the Germanic world were the nunnery at
Lamspringe in Lower Saxony and the female community within the
double monastery of Admont in Carinthia/Austria. Lamspringe followed
the familiar pattern: founded as a collegiate community for women in the
ninth century, between 1119 and 1130 it became a Benedictine nunnery with
an Augustinian canon (from the nearby community of Hamersleben) as
provost. By mid-century it had its own scriptorium, sufficiently active and
skilled to take in commissions from other houses. Two of its nun-scribes,
Ermengarde and Odelgarde, are known by name, and some of their work
survives.42 One could prolong this account of identified female scribes in
Germanic Europe almost indefinitely, but I shall end with the particularly
interesting example of Admont, a rare case of a community whose medieval
library survives largely intact.43 Founded as a Benedictine house in 1074,
Admont was reformed according to the Hirsau pattern after 1115, adding on
a women’s community between 1116 and 1120. More than 200 of its
twelfth-century manuscripts survive, most apparently made locally.
There is good evidence for the substantial participation of the women’s
community in the making of these books. Five women and ten men have
been identified as copyists; by the late twelfth century, each house had
its own library and librarian, and one female scribe, Adelheit, was
commemorated for her work in the monastery’s necrology. Women copied
books for the men’s library and vice versa. Men and women somehow
managed to collaborate on the same book, despite the strict rules which
segregated the two communities. It is known that female recruitment
at Admont, and doubtless elsewhere in Germany, privileged those of
aristocratic birth who had received a good education in the liberal arts
before entering the cloister. As a result the female scribes at Admont were
literate and the female community remarkably self-confident.44

Scriptoria
As the foregoing discussion has implied, it is difficult to discuss the work of
individual scribes other than in the context of a scriptorium. The word was
in use in our period, though I think not widely.45 For our purposes we
might define it as a locality where more than one scribe was at work, either
 Scribes and Scriptoria 77

Figure 4.1 Handwriting of Diemut, inclusa of Wessobrunn. Munich, Bayerische

Staatsbibliothek, Clm 22009, f. 4v.
78 rodney thomson
contemporaneously or in succession. There is probably sufficient reason to
further tighten the definition, by adding to it the notion of a team of scribes
under the discipline of a master, collaborating in the production of books.
If we were to take this last notion seriously, then it is possible that the mere
fact of a number of scribes working in the same place simultaneously might
not in itself constitute a scriptorium. Take, for instance, the relatively large
and wealthy Benedictine community at Malmesbury in south-west
England. It housed the great historian and scholar William, who during
his adult life, ca. 1115–ca. 1143, both copied books himself and organized
other members of the community to support him.46 Eleven surviving
manuscripts from Malmesbury contain William’s hand and those of
another fifty-four copyists. In their general appearance the hands of these
scribes vary widely from each other, as they vary also in competence. Most
of them were clearly not very experienced and some of them only appear
for a few leaves or even lines before – one presumes – William lost patience
with them. Four scribes wrote long stretches of text in more than one book,
collaborating with each other and with William. One of them at least (my
scribe A) evidently wrote over a substantial period of time, since one can
observe his hand gaining in proficiency. These four individuals might be
thought to have constituted the abbey’s ‘scriptorium’, but their work, like
the rest, is not of the highest quality and is strongly individualized. One has
the impression of book-making at Malmesbury very much managed by
William, an energetic scholar, not a master scribe in charge of the scriptor-
ium. Another example of a ‘non-scriptorium’, as it might be termed, is
Salisbury Cathedral over the last quarter of the eleventh century, a secular
community to be sure, but at this period strongly monasticized through the
imposition of the Rule of Chrodegang of Metz by the Norman Bishop
Osbern (1078–99).47 As Neil Ker and Tessa Webber have shown, Osbern
seems to have drafted as many canons as possible to write new books for the
cathedral, in effect to create a new library.48 No fewer than eighty-odd
books survive from this period, written by twenty-six scribes, probably
canons, over two generations. Like the Malmesbury books, these were
rather humble products, the hands not very calligraphic, highly differen-
tiated from each other, almost entirely Continental or Continental-
looking. A distinctive feature of the Salisbury books is the poorly mixed
red ink used for initials and rubrics, resulting in a washy, pale pink colour
and spreading on the page.
Nonetheless, at the opposite end of the spectrum, ‘scriptorium’ could
sometimes mean a dedicated room or building in which scribes worked
more or less continuously. An example would be the English Benedictine
 Scribes and Scriptoria 79
abbey of St Albans. The house chronicle uses the word to refer to
a physical entity, dating from at least the late eleventh century, in
which two or three paid scribes worked at any one time, active through
most of the twelfth century.49 Most of the surviving St Albans manu-
scripts from the twelfth century are written by one or only a few scribes,
in beautiful script and with handsome decoration, both showing a strong
family likeness. These scribes not only produced liturgical and library
books for the abbey itself, but took in orders from other places, above all
female houses or the abbey’s dependent cells.50 The fact of monastic
scriptoria producing work on commission, for presentation or for sale
was perhaps commoner than we think. I have already mentioned
Lamspringe in Saxony; I mention again Wessobrunn, where books
copied by Diemut were given away or traded: a missal to the bishop of
Trier, a liber officialis to the bishop of Augsburg, and a two-volume bible
to buy an estate from a wealthy widow.51
At other houses a higher percentage of the inmates participated, and our
notion of a ‘scriptorium’ has to be modified accordingly: Frankenthal has
already been mentioned; another is Mont-St-Michel, where between one
and two dozen scribes seem to have been at work simultaneously through
the eleventh century, that is, about one-third of the community. These
examples do not resemble the cases of Malmesbury or Salisbury inasmuch
as the standard of writing is much higher, and there is evidence of training
and scriptorial discipline, widely applied.
One question, perhaps too little considered, is the longevity of scrip-
toria. It is becoming increasingly clear that they did not necessarily operate,
at least continuously or intensively, for a long period. The general rule
seems to be that they lasted long enough to provide the basic stock for the
library of the community. One or two generations of scribes might be
sufficient to achieve this. However, there were many variations, some of
which have been mentioned already. In England, Exeter and Salisbury
Cathedrals started their library-building in the last quarter of the eleventh
century, and had acquired the basic stock by ca. 1120.52 Durham Cathedral
also made at least a strong start during the same period.53 In each case the
impetus seems to have come from a (Norman) bishop. A commoner
pattern, not only in England but across much of Europe, was for the
start being made in the 1120s, the finishing-post being reached around
the middle or third quarter of the century. But of course much depended
on the date of foundation of a particular community, and this especially
concerns the houses of the new orders, often founded well into the twelfth
century. Almost the same applies to houses founded much earlier, but that
80 rodney thomson
were reformed ca. 1100 and thus revitalized in terms of their population
and their commitment to lectio divina. In such cases, library-building, with
the supporting scriptorial activity, might begin later and be prolonged up
towards 1200. Examples of this, Frankenthal and Lamspringe, have already
been given. Another variant is found at the cathedrals of Hereford and
Lincoln, which followed the pattern of English secular foundations in
having only small libraries, containing about 100 volumes of core texts.
At both of these cathedrals small groups of books survive, dating from the
first half of the twelfth century. They look like ‘scriptorium’ products, in
that the same distinctive scribes and decorators appear in more than one
book. But the total number of volumes in each case is so small that one
wonders whether the ‘scriptorium’ was really an ‘atelier’, working on site or
in the town over quite a short span of time.54
House Styles: The notion of well-organized scriptoria, with a master scribe
training others, raises the question of whether this resulted in something
that might be called a ‘house style’, that is, a style of writing which was
homogeneous and which was differentiated from that of other scriptoria.
In fact, I think that the development of a ‘house style’ is probably another
prerequisite for the use of the word ‘scriptorium’. Thus, although at
Malmesbury and Salisbury many books were produced in-house, they
did not develop house styles, and so arguably did not have organized
scriptoria. By contrast, at some places the imposed scriptorial discipline
was so strict and so well learned that it is hard to distinguish between the
work of individual scribes. This is true of communities in quite different
parts of Europe, from the Benedictine abbeys of Gloucester and
Winchcombe in south-west England, to that of Admont in Austria.55
My impression is that German scriptoria seem to have been especially
well disciplined, producing scribes who could write not only very well but
in a very similar way. On the other hand, a good scribe could write in
different styles,56 and of course the script of a particular individual could
change over time, with maturity and with old age.
Regional and National Styles: At the beginning of our period different
scribal styles can be attached to particular scriptoria, to regions and to
countries. In Spain the Visigothic script was on the way out, having
been interdicted for the copying of liturgical books by a council of
1090.57 In southern Italy and Sicily the decorative Beneventan script
continued to be written into the early thirteenth century. Only in
Ireland was Celtic minuscule maintained, against pressure from
 Scribes and Scriptoria 81
Caroline and Gothic, well into the sixteenth century. But ca. 1100 there
was also considerable variation within the family of Caroline
minuscule.58 For example, scribes at Christ Church Canterbury ca.
1100 wrote a distinctive hand which not only differentiates it from
writing at Durham Cathedral around the same time, but even from
some of that written at St Augustine’s Abbey, a mere stroll away.59
Writing in Anglo-Caroline of the late eleventh century is distinguish-
able at a glance from the variety of Caroline written in Normandy at the
same time,60 and both these styles are different from what was com-
monly written in the scriptoria of Germany and the Low Countries.
But these differences tend to diminish as the century wore on. First of
all, English, Norman and northern French script became hard to
distinguish. Germany holds out for longer, but by the mid-thirteenth
century the Gothic style dominated in all these regions as well as
in southern France and even southern Italy. Gothic style, as in
architecture and illumination, was a pan-European phenomenon.
The questions of how and why this happened have scarcely been
investigated. Some of it may have had to do with increasing mobility
of scribes, across frontiers and major waterways. One might think that
its early stages were a consequence of international monastic network-
ing. But by the last quarter of the twelfth century and beyond, the role
of the universities, university towns and professional stationers was
probably predominant. The beginnings of this process are to be dis-
cerned with the evidently commercial manufacture of glossed biblical
books in Paris, during the second half of the twelfth century.61 Not only
do we encounter, for the first time, commercial manufacture on quite
a large scale, the product being marketed across most of Europe, but
associated with it is the development of some of the earliest Gothic
script. The details of both the manufacture and trade as yet elude us.
One of the problems is in differentiating products made at the source
from copies of them that seem to have proliferated, at least through
France and England. Copies which, on other evidence, were demon-
strably made in England are nonetheless indistinguishable, in script and
decoration, from copies definitely made, on other evidence, in Paris.
It is no accident that the principal form of decoration of these books
has been christened the ‘Channel Style’. That makes the point that the
dissemination of these books, in the first instance, was westwards.
German libraries did not show the same keenness to acquire them
until a generation or two later.
82 rodney thomson
Notes
1. Facsimile ed. Delisle 1909.
2. Apart from the facsimile itself, see Ker, English MSS, 16, 34–7, and pls. 14–5.
3. But see the examples of Malmesbury and Salisbury cited in the present
volume, pp. 78–80.
4. I am not aware of any attempt to compile figures for the totality of the
phenomenon. However, with respect to female monasticism, Bruce
Venarde speaks of a fourfold increase in the numbers of houses in England
and France between 1080 and 1172, and much the same increase appears to be
true for Germany: Venarde, 1997, 126. His total of such houses in England
and France ca. 1200 is 477. For Germany, figures of ca. 150 in 1100, 500 by
1250, are given by Bernards, 1.
5. Beach, Women as Scribes; Härtel 2006, ch. 3.
6. I have mentioned this phenomenon in Thomson 2007, 26–7. Nonetheless,
see Stirnemann 1984, 1989.
7. Much of the literature is brought together in Gullick 1995, 55, nn. 17–9.
8. Sears 2006, 75–6 and fig. 1.
9. Gullick 1998.
10. On Eadwine, see Eadwine, esp. 13–4, 180–1; on Manerius, Dodwell 1954,
110–1; Gullick 1998, 15 and n. 70.
11. Thomson, St Albans, 1. 13.
12. Thomson 2006, 25 and nn. 34–5.
13. Liber Ordinis, 78 seq.; Cahn Romanesque Manuscripts, 1. 19–20 and n. 42.
14. De Hamel, Glossed Books.
15. Rouse and Rouse 1997, 34–44. Or, more pithily, following the prologue to the
Pauline Epistles: ‘On your feet, Raulinus; off to the pub’ (my translation).
16. Eadwine, 180, for the complete inscription; see also pl. 32. His name also
appears in a prayer following the collect to Ps. 150. The use of the word
‘scriptor’ is interesting: was it intended to cover his painting as well as writing,
or does it mean that he thought of himself primarily as an excellent scribe? Cf.
the present volume. (p. 72), with respect to Hugo Pictor.
17. Orderic (ed. Chibnall), 2. 87. To this one can add the named and praised
scribes at Liessies and Prüfening (Sears 2006, 79–80, 85–8, 94–5), at
St Michael’s Bamberg, and Anchin (Gullick 2006, 101–3).
18. Matthew Paris, Gesta Abbatum Sancti Albani, 32; translated in Thomson, St
Albans, 1. 13.
19. Alexander, Mont St Michel, 37–8.
20. Hereford Cath. O. V. 10, O. V. 14, O. VI. 10, P. I. 12, P. II. 14, P. III. 7, P. V. 3,
P. V. 4; BL Royal 3 A. xii, 7 F. vi; Oxford, Jesus Coll. 26, 52–3, 62, 67–8, 70;
Watson 1984, nos. 798–803. A particularly elaborate version of the inscription
is in Hereford Cath. P. I. 12, f. i: ‘[L]iber sancte Marie de Cirecestre’ abbatis
primi Serlonis tempore scriptus per manus canonicorum Deodati et Fulconis
postea prioris sub Gilleberto primo tunc cantore.’
21. Beach, Women as Scribes, 43, 60–3.
 Scribes and Scriptoria 83
22. The miniature has often been reproduced, e.g. Egbert 1967, 31; Alexander
1978, 109–10; Stammberger 2003, pl. 7.
23. Alexander 1978, 38 n. 2.
24. William of Malmesbury, for instance, corrected his texts heavily in the
interests of comprehensibility, sometimes unauthoritatively, sometimes
against the exemplar or another copy. See Thomson 2015, 169–85.
25. Beach, Manuscripts and Monastic Culture, 65–6, 70–2.
26. Gullick 1995, 39, 42.
27. Mynors, Cassiodorus, xxxix–xlix; Mynors and Thomson, Hereford, 17–8.
28. Ker, English MSS, 12–5, 54–7; Webber 1996.
29. For the Carilef Bible (only vol. 2 survives, one scribe), see Mynors, Durham,
33–4 and pls. 16–8, Gullick 1990, 63; for the Bury Bible (only vol. 1 survives,
one scribe), Thomson, Bury Bible, 36; for the Lambeth Bible (2 vols., one
scribe as far as vol. 2 f. 196, of 302), Dodwell 1959, MMBL 3. 322–5,
Riedmayer, Lambeth Bibel; the discussion by Shepard, Lambeth, is unfortu-
nately amateurish; for the Bible of Stephen Harding (2 vols., three hands),
Załuska 1989, 63–111, 67–70. Other examples are given by Gullick 1990,
82 n. 74.
30. Decoration was another matter: cf. for instance the notorious case of the
Winchester Bible (Oakeshott 1945).
31. Alexander 1978, esp. 90–6, 100–3; Thomson 2002.
32. Pächt 1950; Alexander 1978, 107–8 (pl. 19); Gullick 1990, 74–5. One could add
the German nun Guda, in a caption to her self-portrait in Frankfurt, Stadt-
und Universitätsbibl. Barth. 42: ‘Guda peccatrix mulier scripsitque pinxit
hunc librum.’ And the scribe of the Carilef Bible: Gullick 1990, 64–5.
33. See n. 22. Both the miniatures are reproduced by Egbert 1967, 30–1, and
Alexander Medieval Illuminators, figs. 18 and 19.
34. Cohen-Mushlin 1983, 74–109.
35. Reproduced, e.g., in Alexander, Medieval Illuminators, fig. 15, and (better)
Stammberger 2003, pl. 5.
36. Gullick 1995.
37. Ibid.
38. True even when male and female scribes wrote within the same (double)
house: Beach, Women as Scribes, 103.
39. Beach, Women as Scribes.
40. What follows is based upon ibid., ch. 2.
41. Thomson 2012, 134–5.
42. Walter-von dem Knesebeck 1995; Cohen-Mushlin 2004, 155–70; Härtel 2006;
Hotchin 2007, 167–71.
43. What follows is based upon Beach, Women as Scribes, ch. 3; also Thomson
2007, 29–32.
44. Beach 2002.
45. It figures in the Oxford Dictionary of Medieval Latin from British Sources,
under ‘scriptorius’, with a wide range of meanings including a writing room.
Unfortunately few examples are given.
84 rodney thomson
46. Thomson 2003, ch. 4.
47. Thomson 2006, 54–8, esp. 58.
48. Ker 1985; Webber 1992, ch. 1.
49. Thomson, St Albans, esp. 1. 52–3 on the evidence for a built scriptorium,
repaired in the late twelfth century.
50. Ibid., 1. 25–7, 37–8, 47, 54, 56–61.
51. Beach, Women as Scribes, 43–4.
52. For Salisbury, see n. 48; for Exeter, Ker, English MSS, 23–4; Gameson 1999,
107–27; Thomson 2006, 48–54.
53. Mynors, Durham, 32–63. More recently, Gullick 1990 and ‘Professional
Scribes’.
54. Thomson 1989, xiv; Mynors and Thomson, Hereford, xvii–xix.
55. These observations are based upon my own first-hand observations of MSS
from these houses. For Gloucester, see also Thomson 1997. For Winchcombe,
see also Ker, English MSS, pls. 24 and 25; Ker 1980.
56. Cf. the extraordinary example, cited by Gullick 1990, 83 n. 75 and pl. 20, of
a late eleventh-century MS from Exeter, in which the scribe wrote two short
passages in a convincing imitation of ninth-century Carolingian minuscule,
presumably as found in his exemplar.
57. For Visigothic, Beneventan and Irish scripts, see Bischoff, Latin Paleography,
83–100, 109–11, with bibliography.
58. For more details than I can provide here, see Chapter 2 of the present volume.
59. Ker, English MSS, 23–30; Webber 1995; Gullick 1998. The ‘Christ Church’
style of script, however, did in due course penetrate the scriptorium of St
Augustine’s.
60. Ker, English MSS, pls. 2 and 3, a dramatic confrontation.
61. De Hamel, Glossed Books.
 part ii
Readers and Their Books
 chapter 5

Scholars and Their Books

Constant J. Mews

The personal collections of individual scholars in twelfth-century Europe

were by their nature transient. With certain notable exceptions, we know
about most of them only through notices of their bequests to established
libraries.1 Nonetheless, we can observe through these notices a significant
growth in the number and range of the books collected by individual
scholars, compared to the situation that prevailed in previous centuries.
I focus here on two parallel tendencies in scholars’ private libraries prior to
the increasing standardisation of scholastic book production in Paris from
around 1200. Within the liberal arts, we see an expansion in the range of
texts and authors being collected by scholars, particularly in the trivium,
and to a lesser degree in the quadrivium. In scripture, theology and canon
law, we see a similar desire to expand the range of texts and authors being
studied, above all through the creation of anthologies. In some cases, we see
both tendencies in the collection of books owned by scholars. Rather than
give an exhaustive account of every known private library in the twelfth
century, I shall simply observe these two trends within what we can learn
about a few personal libraries, acutely conscious of the fragmentary nature
of our evidence. Only occasionally do surviving manuscripts inform us
about the individual scholars to whom they might once have belonged.

The Nature of Miscellanies

A word of caution is needed about medieval inventories and the books to
which they refer. They tend to identify a manuscript only by a single author
or title, without regard for the variety of texts that it might contain. Thus
a medieval book list does not necessarily indicate the range of texts that
a single volume might contain. For example, an inventory of the books
owned by the cathedral chapter of Pistoia, compiled by its dean between 1104
and 1116, lists many volumes that still survive, including one, described
simply as an anthology (C. 101: Scarpso, i.e. ‘scarpsus’ = excerpt), copied in
87
88 constant j. mews
the early twelfth century, in fact containing a ninth-century canon law
collection and various patristic prologues to Scripture.2 On a blank folio
after the Capitula Angilrami, the scribe has added three short poems relating
to gender: the first a verse extract from the eleventh-century dictionary of
Papias about feminine nouns that look masculine; the second a relatively
widely copied pseudo-Virgilian poem (perhaps late antique) about three
Amazonian women who kill their husbands in different ways; while the third
is an otherwise unknown ten-line poem about the five stages by which love
can seduce a young man: sight, conversation, touch, kisses and losing one’s
mind.3 These poems illustrate how a scholar in early twelfth-century Pistoia
distracted himself from copying religious texts by adding literary treasures of
a quite different character.
A similar diversity is evident from a manuscript, copied in the mid-
twelfth century, from northern France (BnF lat. 14793). On f. 1v, imme-
diately preceding theological letters by Walter of Mortagne (d. 1174) and an
annotated copy of Abelard’s Confessio fidei ‘Universis’, is found a poem in
the voice of a young woman to an unforthcoming lover and satirical verses
about worldliness in the schools by Petrus Pictor.4 A single scholar, quite
possibly Walter of Mortagne himself, has assembled texts about love,
theological debate and satire. The manuscript illustrates how a scholar
who collected books relating to scripture and theology could just as well
include within a volume of religious texts verses of a quite different
character. Its varied contents remind us that book lists offer an inadequate
guide to the complexity of texts that a volume might contain.

Expanding Awareness of Classical Literature

Most bequests of books are very small and do not allow us to glimpse the
larger collection that a private scholar might have owned.5 One example of
a larger donation is known from a list of fourteen volumes given to
Beauvais Cathedral by Roscelinus grammaticus in the early twelfth century.
Apart from three specifically Christian texts, namely Augustine’s homilies
on John (a seventh-century uncial copy from Luxeuil), Augustine’s De
doctrina christiana and a troparium, the contents are classical: Priscian,
Macrobius, an Arismetica (Boethius), a Dialectica, a Rethorica de inventione
(Cicero), Boethius’ Consolation of Philosophy, Virgil, Horace, Juvenal,
Ovid’s Metamorphoses and the Thebaid of Statius.6 The list implies that
this grammaticus also taught dialectic, rhetoric and arithmetica. He may
well have been Roscelin, cantor of Beauvais, who, together with a certain
Nevelo of Compiègne, asked Bishop Guy to establish a community of
 Scholars and Their Books 89
canons at St-Vaast, Beauvais, in 1072.7 It seems likely that this is the same
Roscelin of Compiègne who was accused of teaching heresy at Beauvais by
Fulco, a former monk of Bec, installed as bishop of Beauvais in 1088, after
Guy had been ousted from office. The Dialectica owned by Roscelinus
grammaticus may have been similar to the Dialectica of Garlandus, who
taught at Besançon, where Roscelin of Compiègne also held a canonry.8
As John of Salisbury later remarked, such treatises were popular because
they combined the teaching of various authors, Porphyry, Aristotle and
Boethius, into a single book.9 It was much more convenient to have a single
manual than individual copies of glossed authors. The bequest illustrates
the range of texts a teacher might own.
A similar range of classical authors is cited in the Commentum in
Theodolum, composed by Bernard of Utrecht sometime before 1099.10
Bernard exploits the use of pagan authors by Theodolus, a minor early
Christian poet, to justify contemporary study of the auctores. Bernard
draws on Aristotle’s comment in the Prior Analytics about investigating
an author’s intention long before that work became widely known in the
Parisian schools.11 Bernard’s analysis of how any text should be related to
philosophy echoes that of the Glosule on Priscian’s Grammatical Institutes,
also composed in the late eleventh century.12 Conrad of Hirsau expands on
Bernard’s Commentum in his Dialogus super auctores, perhaps from the
1130s. Conrad is cautious about certain of Ovid’s writings, in particular the
Heroides, in which the poet ‘croaks about love’.13 While Ovid’s
Metamorphoses had been known since the Carolingian period, he was not
then as popular as Virgil or Horace. By the late eleventh century, however,
Ovid’s writings seem to have become increasingly available to individual
scholars. In the eleventh century, a bequest of thirty-five volumes made by
a certain Reginfridus to the Bavarian Benedictine house at Benediktbeuern
included both Ovid’s Metamorphoses and De arte amatoria.14 Conrad of
Hirsau’s caustic comment about the popularity of Ovid’s ‘croaking about
love’ hints at monastic suspicion of some of the texts circulating among
students in the early twelfth century. An early witness to this enthusiasm
for the Heroides is provided by the Carmina exchanged by Baudri of
Bourgeuil with women like Constance of Le Ronceray. Another is illu-
strated by the anonymous Deidemia to Achilles, written in the voice of an
anguished princess to her beloved, in imitation of the Heroides.15 While the
Metamorphoses is the only Ovidian text mentioned among the 135 volumes
in the late eleventh-century inventory of a well-endowed female monastery
(possibly that of Chelles), the Heroides was certainly known to educated
young women in the early twelfth century.16 A copy was given by a scholar,
90 constant j. mews
along with Ovid’s De ponte, to Tegernsee, according to an annotation in its
twelfth-century copy of Persius.17 By around 1150, the Heroides had become
a standard textbook, even in schools run by monasteries, as shown by the
popularity of introductions (Accessus) to its text.18

Enlargement of the Trivium

A key figure exemplifying the expansion of the trivium within a personal
library is John of Salisbury, bishop of Chartres (d. 1180). His Metalogicon
and Policraticus summarise key themes of those authors whose study he
esteemed, whether in grammar, dialectic or rhetoric. By the 1150s, John had
become familiar with the Sophistical Refutations and Prior and Posterior
Analytics of Aristotle, texts largely unknown in the 1120s. Many of these
were texts that Thierry, chancellor of Chartres, had included in his
Heptateuchon, donated by him to the cathedral library. There has been
much debate about where John pursued his studies between 1138 and 1141,
whether at Chartres as traditionally thought, or at Paris, as Richard
Southern argues.19 By the late 1130s, the Parisian schools were much larger
than those of Chartres, but students there did not have access to the range
of texts assembled at the cathedral library of Chartres through the activity
of a line of distinguished teachers.
John’s goal was to respond to the explosion in learning confronting
students by the mid-twelfth century. He regretted that hyper-enthusiasm
for dialectic had led to neglect of the traditional auctores, as taught by Bernard
of Chartres and William of Conches. John’s eulogy of a group of humanis-
tically inclined masters, above all of Bernard of Chartres (about whom he had
only heard reports), enabled him to criticise a tendency he observed in the
schools to focus uniquely on dialectic, rather than the broader tradition of
logica, embracing dialectic and rhetoric, with its foundations in the study of
grammatica. Over his career, John would certainly have built up an impressive
library of classical authors. Yet the collection of some twenty-five volumes
that he bequeathed to the cathedral of Chartres may not have constituted his
entire library. It included classical authors: Valerius Maximus, Vegetius,
Eutropius, Seneca and Cicero’s De officiis and De oratore, but many more
patristic authors, including Jerome, Augustine and Origen, and the Celestial
Hierarchy of Pseudo-Denis the Areopagite.20 The copy of the Policraticus that
he bequeathed provided his personal synthesis of much (although not all) of
the classical reading he had acquired. It may well be that many of the more
unusual items in John’s personal library were dispersed among friends, and
that he gave to the cathedral library only those volumes that it wanted.
 Scholars and Their Books 91
John’s literary interests are paralleled by those of Philip of Harcourt,
bishop of Bayeux (1142–64), who bequeathed 140 books to the abbey of Bec
in the final months of his life. Philip was more of a collector than
a scholar.21 His personal library provided thorough coverage of all the
Latin Church Fathers, as well as a significant number of Greek Fathers in
Latin translation, but there was also a solid array of classical texts, above all
of Cicero, Sallust and Seneca, of which almost nothing now survives.
The library at Bec benefited from bequests such as these, accumulated
outside a monastic environment by a bishop strongly interested in the
classical tradition.

The Expansion of Scientific Learning

Scientific texts by non-Christian authors were also increasingly sought by
scholars in the later eleventh and twelfth centuries, building on
a precedent set by Gerbert of Aurillac (d. 1003), who had frequently
asked his friends for copies of scientific texts to which he had no access in
the libraries of northern France.22 In the 1060s, Constantine the African
(an Arabic-speaking Christian who had studied at Fustat or Old Cairo,
where Caliph al-Hakim had established a massive library in the early
eleventh century) took refuge in Salerno. He then came to the attention
of both its Norman rulers and the abbot of Monte Cassino, where he
became a monk. Through the translations of Arabic medical and
scientific writers made by Constantine, as well as of Nemesius made
from the Greek by Alfanus, archbishop of Salerno, a few Latin scholars
could acquire knowledge of Hellenistic scientific and philosophical tradi-
tions. The fact that Anselm sought to obtain a copy of the Articella for
Bec illustrates how new modes of thinking, emphasising rational inquiry,
could transform a monastic library.23 Doctors may have owned many
such texts in their private libraries, only occasionally preserved for
posterity through their bequest to an institution.
Most of the bequests of books made to Chartres Cathedral involved just
one or two items.24 By comparison, Bernard of Chartres, subdeacon and
chancellor, bequeathed twenty-four books to his chapter library.25 Thierry,
archdeacon and chancellor from 1141 to 1150/1, had an even larger collec-
tion. He donated his two-volume Heptateuchon, three volumes of Roman
law and fifty-five other volumes, many of them summarised within his
massive compendium of the seven liberal arts.26 It included not just
Aristotle’s Prior Analytics and Sophistical Refutations, which had only
begun to become widely known by the 1130s, but translations of Euclid’s
92 constant j. mews
Elements and a treatise on numbers by Al-Khwarizmi. Thierry acquired
these texts through a scholarly network that included Adelard of Bath and
Hermann of Carinthia, active as translators in England and Spain,
respectively.27 Another great collector of scientific books was Robert of
Torigny, made prior of Bec in 1149, and abbot of Mont-St-Michel from
1154 until his death in 1186. Many precious philosophical and scientific
texts now preserved at Avranches were undoubtedly collected through his
interests.28
Only a few scholars in the twelfth century were able to procure transla-
tions of scientific and philosophical texts made in Spain by figures like John
of Seville, who effectively initiated the translation movement at Toledo.29
His work paved the way for Robert of Ketton (ca. 1110–60) and Dominicus
Gundissalinus (ca. 1119–90), archdeacon of Segovia and perhaps
a converted Jew.30 When Peter the Venerable travelled to Spain in the
early 1140s, commissioning Robert of Ketton to translate the Quran into
Latin, the abbot of Cluny was implicitly recognising the limitations of
Christian understanding of a broader world, even if he embarked on the
project in order to reaffirm its spiritual superiority.31 His polemic against
Saracen impiety would not impede a further generation of translators,
most famously Gerard of Cremona (1110–87), reportedly a royal physician,
from translating a massive body of medical, scientific and philosophical
texts into Latin. Where or how Gerard acquired his knowledge of Arabic is
not known. Nonetheless, with the help of Gundissalinus, Gerard amassed
a remarkable library of texts that he translated from Arabic, not only
previously unknown texts of Aristotle and Ptolemy, but also the writings
of many Islamic physicians and philosophers, most famously Avicenna.32
Some of these translations were current in Normandy and England before
becoming more widely known within the University of Paris in the
thirteenth century.

The Broadening of Canon Law, Exegesis and Theology

While a small group of scholars built up collections of classical and
scientific works, the disciplines best represented in their libraries were
canon law, biblical exegesis and theology. Over the course of the twelfth
century scholars came to value manuals and anthologies that collected an
increasingly wide range of authorities in these disciplines, arranged in a way
that could provoke rational reflection. The Decretum compiled by Ivo,
dean of St-Quentin at Beauvais until 1090 and subsequently bishop of
Chartres, systematised a much wider range of Christian authors than had
 Scholars and Their Books 93
been gathered in the similar work by Burchard of Worms in the early
eleventh century.33 Ivo urged readers not to be disturbed by the variety of
opinions they might encounter. As he explained in his introduction, the
core principle was Augustine’s teaching, ‘Have charity and do what you
will’. Ivo’s great contemporary, Anselm of Laon (d. 1117), provided
a similar service in systematising access to the Church Fathers by delivering
sententie on major doctrinal questions they raised.34 Whereas traditionally
only major monastic or cathedral libraries might own complete sets of
writings of the Church Fathers, Anselm of Laon made it possible for
scholars to acquire for themselves easy access to their opinions. The lack
of uniformity of manuscripts containing theological sententie from the
school of Anselm of Laon testifies to the freedom with which scholars
compiled collections of his teaching.
In the decade ca. 1130–40, Anselm of Laon and his immediate disciples
produced marginal and interlinear glosses on scripture that made it possi-
ble for individual scholars to own a portable library of patristic and
subsequent exegesis on all the major books of the Bible: Genesis, Job,
Proverbs, Ecclesiastes, the Song of Songs, Matthew, John, the canonical
Epistles, Revelation and quite possibly also on Mark, Luke and the Pauline
Epistles.35 While most surviving copies belonged or were bequeathed to
religious institutions, there can be no doubt that private scholars acquired
their own copies. Robert Amiclas, who taught in Paris in the mid-twelfth
century, left to the Cistercian abbey of Buildwas, near Hereford, some time
after 1176, a collection that makes up an almost complete Glossed Bible,
dating from the 1130s to the 1170s. He also owned a few texts for teaching
grammar (glosses on Theodulus and Juvenal and of Petrus Helias on
Priscian), but no works of systematic theology.36 The completeness of his
collection of glossed books of the Bible illustrates the disparaging comment
of Robert of Melun, bishop of Hereford from 1163 to 1167, about the
reverence such glosses were generating in the schools as more important
than scripture itself.37 The fifth booklet of one of the Buildwas manuscripts
contains, alongside four booklets of glossed texts owned by Amiclas, the
Commentarius Cantabrigiensis, a reportatio of Abelard’s commentary on the
Pauline Epistles.38 His bequest illustrates how it was possible for a teaching
master to benefit from the pedagogical revolution that took place in the
twelfth century.
Peter Abelard’s Sic et non (initially drafted ca. 1120) similarly enabled
students to have access to a much wider range of patristic authorities than
had been initially assembled by Anselm of Laon, but with greater focus on
the need for critical reflection. Differences in perspective, Abelard
94 constant j. mews
suggested, might derive from the rhetoric employed. Thus, while Abelard
repeated Ivo’s version of Augustine’s teaching that one should always ‘have
charity’ (habe caritatem), he also supplied Augustine’s original wording,
dilige et fac quod vis. Abelard attached to the Sic et non the Retractationes
of Augustine, as an example of how the greatest Church Father could
reconsider his opinions at a later date. The Sic et non provided a convenient
portable library for peripatetic students. Abelard’s patristic reading was
unusually wide. He had a passion for authenticity, shown, for example, in
the way that he corrected the attribution of a text (De ecclesiasticis dogma-
tibus) from Augustine to Gennadius.39 The relative popularity of the Sic et
non compared to his Theologia demonstrates how Abelard’s greatest con-
tribution to the discipline of theology was to urge students to evaluate the
arguments of the Church Fathers, without being bound by prior commit-
ment to accepting their point of view. The Sic et non showed what was not
so clear from the summaries of Ivo of Chartres and Anselm of Laon: that
there was great diversity of theological perspective within the patristic
canon, and that it was necessary to read the Fathers with awareness of
the arts of language. Abelard showed that Augustine did not always have
the last word on any disputed question.
Abelard’s technique of debating the precise text of the Fathers was very
different from that promoted by Hugh of St-Victor in his Didascalicon,
written in 1121.40 Hugh combined an introduction to the various disci-
plines of the liberal arts with guidance for sacred reading, the goal of which
was not simply knowledge, but advancement of the soul along the path of
virtue and wisdom. While Abelard emphasised the questioning of received
assertions, Hugh encouraged the application of a meditative focus not just
to Scripture but to all reading. Hugh’s doctrinal synthesis, formulated
most completely in the De Sacramentis, composed during the 1130s, was
much more influential than that of Abelard’s Theologia. Yet their
approaches were not mutually exclusive. Few students could afford their
own copy of Hugh’s massive synthesis. For that reason, the Summa
sententiarum, compiled in the late 1130s by a student of Hugh, was much
more accessible and popular than Hugh’s De Sacramentis. A student’s
manuscript copied in the 1140s and given to St-Victor (Paris, Bibl. de
l’Arsenal 256) contains both Abelard’s Theologia and the Summa senten-
tiarum. At a later date, this manuscript was extended with a copy of John of
Cornwall’s treatise to Pope Alexander III attacking Peter Lombard (ca.
1170).41 The manuscript reveals a fascinating range of theological perspec-
tives. The librarian of St-Victor thought it appropriate to supplement the
Summa sententiarum and Abelard’s Theologia with a treatise warning
 Scholars and Their Books 95
against the theologian Peter Lombard, who himself drew heavily on both
these texts in his own Quattuor libri sententiarum, completed in the
1150s. This manuscript was certainly used by a student of theology in
mid-twelfth-century Paris. Each item in it offers its own perspective on
Christian theology.
One legacy of the teaching of Hugh of St-Victor was increased interest
in the scriptural exegesis of Jewish scholars. Unlike Anselm of Laon, Hugh
gave considerable attention to the exegetical views of Hebraei, generally
disciples of Rashi.42 Hugh’s interest in Hebrew exegesis was a particular
inspiration for Andrew of St-Victor, abbot of the Augustinian canonry at
Wigmore in Shropshire, and a cluster of other English Hebraists. One of
these was Odo, author of the Ysagoge in Theologiam, which drew on both
Abelard’s Sententiae and the Summa sententiarum.43 It was written by an
English scholar familiar with Abelard’s theology, who cited Old Testament
passages in Hebrew script (with Latin transliteration) as part of his desire to
retrieve the authentic text of scripture.44 Another English Hebraist was
Maurice of Kirkham, an Augustinian canon and enthusiast for the study of
Hebrew, writing in the 1170s, who tells us in his Contra Salomitas that
when he was young (ca. 1130?), he had learned Hebrew from certain Jews,
and had copied Hebrew psalters that had belonged to Gerard, archbishop
of York from 1100 to 1108. As royal chancellor, Gerard (previously arch-
deacon of Rouen) may have helped introduce Jews into England.45
William of Malmesbury comments that Gerard was so interested in
astrology and the black arts that the canons of York refused him burial in
their cathedral.46 The comment sheds light on the Hebrew books owned
by Archbishop Gerard, perhaps similar to the surviving example that
belonged to St Augustine’s Abbey, Canterbury, written in England ca.
1150, in which the same scribe wrote the text in Hebrew and glosses in
Latin.47 Maurice exemplifies how knowledge of exegesis in the twelfth
century could expand to acquisition and production of books written in
Hebrew.

The Personal Library of Guido di Castello (Pope Celestine II)

One politically important figure who amassed a remarkable specialist
library was Guido di Castello, a cardinal deacon from 1127, who became
Pope Celestine II (1142–3) after the death of Innocent II.48 Trained as
a lawyer, Guido was a significant figure in the papal court, who served as
papal legate in Cologne in 1131/2 and in France 1139/40.49 He was one of the
many cardinals present with Pope Innocent II at the dedication of an altar
96 constant j. mews
at Morigny in January 1131, an event attended by Peter Abelard, described
by the local chronicler as ‘a monk and abbot, and a religious teacher of the
most distinguished schools’.50 Among the fifty-six volumes he bequeathed
to the church of Città di Castello only two (a Rhetorica and a Seneca) do
not relate to canon law, scripture or theology. Alongside numerous
unidentified scriptural glosses, an excerpt from Ivo of Chartres and the
Candela (a synthesis of patristic and canonical texts compiled by Garland
of Besançon), he owned an important copy of Abelard’s Sic et non and
Theologia cum libro retractationum (the latter referring to Augustine’s
Retractationes, included by Abelard in his Sic et non).51 Because these two
items are found in a manuscript at Monte Cassino (Archivio della Badia
174) copied ca. 1200, it is reasonable to assume that this copy is related to
Guido’s manuscript. It contains Abelard’s heavily annotated working draft
of the future Theologia ‘Scholarium’ as it stood in the early 1130s, a draft also
preserved in Tours, Bibl. mun 85.52 The likelihood is that Guido commis-
sioned a copy of this manuscript after meeting Abelard at Morigny.
Guido’s interest in Abelard was viewed with concern by Bernard of
Clairvaux, who wrote to him immediately after the Council of Sens in
1141 a letter, subsequently introduced within Bernard's correspondence as:
‘To master Guido, who had been a disciple of Peter Abelard, about whom
he presumed greatly, and who was later Pope.’53 William of St-Thierry’s
comment that Abelard’s books had flown across the Alps and could be
found in Rome itself reflects his fear of the support these texts were finding
among senior figures in the curia.54 In 1140/1 Guido’s apparent support for
Abelard was perceived by William of St-Thierry and Bernard of Clairvaux
as dangerous to the cause of ecclesiastical order. Yet the book list shows that
Guido was more interested in canon law and scholastic theology than in
the writings of Bernard and Hugh of St-Victor, neither of whom figure in
his personal library.

The Personal Library of Peter Lombard

Another scholar who built up a revealing personal library was Peter
Lombard, who taught in Paris from the early 1140s until becoming its
bishop (1159–60). At his death, he bequeathed to the cathedral of Notre-
Dame his collection of glossed books of the Old and New Testaments,
a copy of his Sentences and the Decretum.55 Separate testimony about his
library is provided by an enigmatic list of books preserved on the final folio
of Abelard’s Theologia ‘Scholarium’ in Paris, Bibl. de l’Arsenal 256 (already
mentioned) that apparently records an exchange of books between Peter
 Scholars and Their Books 97
Lombard and the canons of St-Victor.56 Its text, not easy to understand,
merits translation:
This is the collection of books which Geoffrey Moricii carried [added
above the line: from the books of master P. bishop]: Ovid’s De arte
[amandi]; Ovid, Heroides; Maximianus; Avianus; Pamphilus [a widely
copied pseudo-Ovidian comedy]; Statius, Achilleidos; Ovid, De Ponto
incomplete; letters of Horace, incomplete; Ovid, Metamorphoses incom-
plete; three books of Lucan; two books of Juvenal; Virgil; glosses of Cato
and Theodolus; Ovid On . . . without title [i.e. the Amores]; bucolics; and
nine quaternions.
[Second hand] Then, when he went to Paris, he took with him the
Apocalypse of John, the canonical letters, the Lamentations of Jeremiah,
and Claudian and two Priscians. The priest R. indeed sent to master
P. through William de Fonte Morini, fifteen quaternions and through cleric
P. twenty [quaternions], and later indeed, through canon G., forty-four
quaternions about sentences and others compiled about the Psalms. He gave
six volumes of the Pentateuch of Moses, and the four Gospels and the book
of Ezekiel to the brothers of the habit. He kept back the Gospel History and
a quaternion of notes and a Bernardinus.57

What are we to make of this list? It was added at the end of a manuscript
containing Abelard’s Theologia and the Summa sententiarum, before John
of Cornwall’s treatise criticising Peter Lombard was appended to it.
Stirnemann uses the fact that the manuscript belonged to St-Victor to
suggest that it may record books coming into the abbey (presumably to be
copied), as well as books that had been copied at St-Victor, either by its
canons or by professional scribes that it employed. The first entry records
classical authors borrowed from the books of someone identified in
a subsequent interlinear note as ‘master P. bishop’ (who can only be
Peter Lombard). Stirnemann suggests that a copy of Lucan from
St-Victor (BnF lat. 15406) could be from this collection.58 The second
entry begins with a list of scriptural texts, without making explicit if they
were being taken to the ‘master Peter, bishop’. Then it lists various texts
sent by the priest R. through various individuals (William de Fonte
Morini, a cleric P. and a canon G[aufridus?]), to ‘master P’. Stirnemann
suggests that the massive collection of forty-four quaternions of sententie
(352 folios) refers to the exemplar of Peter Lombard’s IV libri sententiarum,
supplemented by his glosses on Psalms. An early copy of the Sentences
(Troyes, Mediathèque 900) has a colophon declaring that it was produced
in 1158 by ‘Michael of Ireland’, to whom she attributes other key manu-
scripts also connected to St-Victor.59 While the authority and date of the
98 constant j. mews
Troyes manuscript are disputed, Stirnemann believes that it could have
been made in 1158 from the forty-four quaternions of sententie being sent
back to Peter Lombard. The list also implies that the copyists, presumably
at St-Victor, borrowed a significant number of classical texts from Peter
Lombard’s library, including Ovid’s Heroides and the more recently com-
posed Pamphilus.60 At a broader level, the list gives vivid insight into the
complexity of exchanges of books between scholars and religious commu-
nities in a Parisian environment.

Scholasticism and Monastic Libraries in the Later Twelfth Century

Many scholastic texts, of both classical and Christian authors, were
bequeathed by individual scholars to monastic houses, whose libraries
were often better endowed than those of cathedrals. While monastic
libraries remained repositories of traditional monastic spirituality and
kept a only small proportion of books containing pagan authors or transla-
tions of Islamic scientific writers, they grew through the donations made by
individual scholars of both non-Christian and scholastic Christian texts.61
A simple way of recognising this is by observing the large number of
monastic copies of such intellectual masterpieces as the commentary of
Gilbert of Poitiers on the Opuscula sacra of Boethius. Otto of Freising, who
studied in northern France, quite possibly under Gilbert, would himself be
responsible for introducing many new books, including Aristotelian texts,
into monastic libraries in the German-speaking regions of Europe.62 While
there is no doubt that monastic writers like Bernard of Clairvaux generated
a distinct theological literature in the twelfth century, we should not
assume that monks steered away from reading texts generated in the
urban schools. The fifty-one volumes of Frowin, abbot of Engelberg
(1142–78), offer a collection remarkable for being almost totally classical
in character. The few exceptions are notable: a certain Augustinus super
epistolam Jacobi, sermons on the Eucharist and two summaries of
Abelard’s teaching: the Sic et non (in a volume that also contained glosses
on Macrobius and Priscian) and sententie, taken down from Abelard’s
teaching, also preserved at St Gallen (Stiftsbibl. 69).63 Frowin, who
composed a thoughtful Tractatus de ueritate in response to debates
going on in the schools, brought to Engelberg a range of texts which he
presumably encountered in his studies. These included many works of
Boethius, Cicero on rhetoric and friendship, Horace, Avianus, Statius,
the Latin Homer, Lucan and Ovid. In Germanic regions, such libraries
 Scholars and Their Books 99
continued to play a vital role in promoting a culture that was both
humanist and monastic.
Because monastic libraries in southern Germany escaped the wars of
religion that engulfed so much of France and northern Germany in the
sixteenth and seventeenth centuries, they can paradoxically give us a better
glimpse into twelfth-century scholastic culture than those from France.
One famous example of this is the thirteenth-century manuscript from
Benediktbeuern that contains the Carmina burana. It included many songs
about love and satire that had circulated in northern France a century
earlier. The fact that so many German students chose to enter monastic
houses, bringing their personal libraries with them, undoubtedly helped
maintain an intellectual tradition that would survive in Germany well into
the early fourteenth century. At the same time, most of the books preserved
in monastic libraries related to scripture, theology and canon law.
Individual bequests made to such institutional libraries may sometimes
have included texts of a secular character. The great majority, however,
were books of immediate benefit to the monastic community.

Conclusion
This brief survey can only offer a glimpse of the expanding range of books
individual scholars owned during the twelfth century, relating both to the
liberal arts and theology and exegesis. Over the course of the twelfth
century, donations by individual scholars to institutional libraries varied
in size: from the fourteen volumes donated by Roscelinus grammaticus to
Beauvais in the early twelfth century, to the 140 given by Philip of Bayeux
to the abbey of Bec. While institutional religious libraries were always well
stocked with Christian authors, well-educated scholars like Roscelin and
Thierry were interested in building up their own collections of the cano-
nical auctores, both pagan and Christian, for use in teaching. The greatest
number of texts belonging to individual scholars, however, were in the
domains of canon law, exegesis and theology. Some monks might have
been troubled by the scholastic character of the new books being
bequeathed to monastic libraries. By the mid-twelfth century, however,
it had become increasingly normal for scholars to own not just a good
range of non-Christian authors, but sophisticated works of theology and
exegesis that took for granted an education in the artes. The books they
owned and donated to institutional libraries were important both for
systematising the learning of the past and for broadening mental horizons
beyond the confines of a purely Latin Christian intellectual heritage.
100 constant j. mews
Notes
1. See Genevois, Genest and Chalandon 1987; for Germanic regions, see
MABKDS and Becker, Catalogi.
2. Identified by Savino 1987, 35, 39.
3. Described as grammaticalia in Murano, Savino and Zamponi 1998, 40–1; the
poems are from Papiae Ars grammatica, ed. Cervani 1998, 87–8; Anthologia
latina, ed. Riese 1869, 257–8 no. 392. See Mews 2016.
4. Puella ad amicum munera promittentem, also preserved in Liège, Bibl. uni-
versitaire 77, ff. 72v–73r, was edited by Bulst 1975, 16; see Mews 2014.
5. Munk Olsen 1989, 31–43, esp. 39 on the collections of individual
scholars. He notes donations to Egmond of thirteen volumes by master
Baudouin (1057–1105), eleven by a priest Simon (1130–61), fourteen to
St-André-de-Rosans by Robert de Galone, twenty-six by master
Alexander to Jumièges, where he became abbot from 1198 to 1213.
6. Cited by Delisle 1886, 160.
7. I argue that he is Roscelin of Compiègne, who came into conflict with Fulco,
a monastic bishop of Beauvais, supported by St Anselm, in Mews 1996,
esp. 17.
8. While De Rijk identified its author as the eleventh-century Garlandus
Compotista, he is more likely to be a younger Garland, active in Besançon
between 1118 and 1136, attested as accompanying Thierry of Chartres in 1148 to
Frankfurt; see Robert, and B. de Vregille in Dictionnaire d’histoire et de
géographie ecclésiastique 20. 887. Roscelin’s name is mentioned by
Garlandus, Dialectica, 107.
9. John of Salisbury, Metalogicon 3. 4.
10. Commentum in Theodolum, in Accessus ad Auctores, 55–69.
11. Ibid., 66–7, referring to Aristotle, Analytica Priora I, 24a, ed. Minio-Paluello,
5 and 433.
12. Gibson 1979.
13. Conrad, Dialogus super auctores, 114–5; on this passage, see Tilliette 1998.
14. MBKDS, 4. 2 Bistum Freising, 750 no. 107.
15. Baldricus Burgulianus, Carmina; Stohlmann 1973.
16. Ovid’s heroines are mentioned by the young woman in Ep. 45 of the
Epistolae duorum amantium, ed. Könsgen 1974, 24. On a library perhaps
similar to that of Argenteuil (BnF lat. 943, ff. 154v–155r), see Turcan-
Verkerk 2007.
17. MABKDS, 4. 2 Bistum Freising, 751 no. 108 (BSB Clm 19490).
18. Accessus ad auctores, ed. Huygens 1970 (from BSB Clm 19475), 29–38.
19. Mews 2014.
20. Cartulaire de Notre-Dame de Chartres; also ed. by Webb 1941.
21. Becker, Catalogi, 199–200, no. 86.
22. See Riché 1988.
23. On connections between science and new theological thinking in Normandy,
see Gasper and Wallis 2004.
 Scholars and Their Books 101
24. Giacone 1974, 42, identifies nineteen donations of books, but most are very
small; Cartulaire de Notre-Dame de Chartres 3. 225; also Obituaires de la
Province de Sens 2. 71.
25. Ibid. (2 June).
26. Cartulaire de Notre-Dame de Chartres 3. 108 (mistakenly reporting forty-five
rather than fifty-five vols.); Obituaires de la Province de Sens 2. 206.
27. The preface is edited by Jeauneau; see also Lejbowicz 2003.
28. Nortier 1971, 39–42.
29. On the broader contribution of Petrus Alphonsi, Adelard of Bath and their
contemporaries, see Burnett 1997 and Chapter 15 of the present volume.
30. For an overview of this group of translators, see Burnett 2011.
31. Burman 2007, 14–7; Burnett 2011.
32. Bertolacci 2011.
33. Ivo of Chartres, Decretum, Prol.: PL 161. 47B.
34. On their influence, see Giraud 2010, 331–405.
35. Stirnemann 1994, 262.
36. Thomson 1995 and Smith 2013.
37. Robert of Melun, Sententiae, Prefatio, 15.
38. Thomson 1995, 242–3.
39. See the correction in Abelard, Theologia christiana 1. 28 (Petri Abaelardi Opera
Theologica 2. 83).
40. Dominique Poirel re-dates the Didascalicon to 1121 in Poirel 2011, esp. 307–9.
41. See the introduction to Abelard’s Theologia ‘Scholarium’ in Petri Abaelardi
Opera Theologica 3. 245–6, and Stirnemann 1998. While the poor text of
Abelard’s Theologia in the Arsenal manuscript makes it unlikely to be the
actual manuscript owned by Peter Lombard, as suggested by Stirnemann
(Petri Abaelardi Opera Theologica 3. 264–6), it could have been copied by
a student drawing on copies belonging to Peter Lombard; on its list of books
borrowed from the library of magister P. episcopus, see n. 56.
42. Moore 1998, 76–89.
43. Ysagoge; see also Luscombe 1968.
44. An English student added a record of otherwise unattested glosses of Abelard
to BL Cotton Faustina A. X, copied in 1148 in the region of Worcester and
Gloucester, alongside copying various moralistic texts; see Burnett and
Luscombe 2005 and Álvarez López 2012.
45. Excerpts from the Contra Salomitas, preserved in complete form (along with
an exchange of verses with Roger, archbishop of York) only in a fifteenth-
century copy, Bodl. Libr. Hatton 92, are edited by James.
46. William of Malmesbury, Gesta Pontificum Anglorum 118. 2–3 (392–4).
47. Lieftinck 1955.
48. Wilmart 1923; Nebbiai-Dalla Guarda 1992, 107–9.
49. Malacek 1981, 64 and 72.
50. La Chronique de Morigny, 53–4.
51. Garland, Candela. His teaching on the Eucharist was questioned by Hugh
Metel: Mews 2001, esp. n. 50.
102 constant j. mews
52. See my introduction to Petri Abaelardi Opera Theologica 3. 210–7 and 268.
53. Bernard, Ep. 192, in Bernardi Opera Omnia, 8. 43–4.
54. William of St-Thierry, Ep. 326, PL 182. 531BC.
55. Cited by Brady in his Prolegomena to Peter Lombard, Sententiae, I/1. 19*.
56. Stirnemann 1998; see n. 41.
57. The text given here is that edited by Stirnemann 1998 (302), which improves
on that offered in CCCM 13, 243–7: ‘Hec est summa librorum quod
Gaufridus Moricii portauit [de libris m[agistri] P. episcopi]: O[uidium] de
arte, O[uidium] Heroidum, . . ., librum Maximiani, librum Auiani, librum
Pamphili, Stacium Achilleidos, O[uidium] de Ponto imperfectum, epistolas
Oracii imperfectas, O[uidium] Metamorphoseos imperfectum, tres libros
Lucani, duos libros Iuuenalis, Virgilium, glosas Catonis et Teodoli, et
O[uidio] de . . . sine titulo, bucolica et novem quaternulos. (b) Deinde,
quando iuit Parisius, tulit secum Apocalypsim Iohannis, epistolas canonicas,
lamentations Ieremie, et Claudianum et duos Priscianos. R. uero presbyter
misit postea magistro P. per Willelmum de Fonte Morini, xv quaternulos et
per P. clericum xx, postmodum uero per G. canonicum xliii quaternulos de
sententiis et alios super psalmos compilatos, et vi volumina Pentateuchum
uero Moysi, et iiii euangelistas et librum Ezechielis, et dedit fratribus de
habitu. Hystoriam euangelicam retinuit et quaternulum de notulis et
Berhardinum.’
58. Stirnemann 1998, 306.
59. Ibid., 309, suggesting that the same scribe produced an early copy of Peter
Lombard’s glosses on St Paul (BnF lat. 7246), another copy of the Sentences
(Oxford, St John’s Coll. 49), given to Hilary, bishop of Chichester (d. 1169),
and an account of the miracles of St Victor (Paris, Bibl. de l’Arsenal 939).
60. Dronke 1979 questions the usual view that Pamphilus originated in the Loire
valley, suggesting that it came from south-eastern Germany in the late
eleventh century.
61. Mews 2007.
62. Rahewin, Gesta Friderici 4. 14 (250).
63. Becker, Catalogi, 233, no. 103; MBKDS 1. 32–3.
 chapter 6

The Libraries of Religious Houses

Teresa Webber

The period between the late eleventh century and end of the twelfth
witnessed significant expansion in the manuscript holdings of religious
communities. Collections similar in scope had been formed during the
Carolingian period, but their number, geographical density and spread by
the end of the twelfth century reflect the extraordinary increase in religious
foundations and benefactions to existing communities across Western
Europe that took place from ca. 1000.1 The broad contours of develop-
ments in the contents of the collections of male communities during this
period are already well known: significant expansion in the holdings of the
Church Fathers and, in many communities, of the pagan authors of
ancient Rome, as well as the introduction of texts, commentaries and
other compendia and tools associated with the schools.2 Detailed study
of these developments at a local and regional level, however, remains
incomplete: a consequence not only of variation in the scale of loss and
dispersal of the books themselves but also of the limited extent to which the
surviving evidence has been collated and edited.3 It is therefore not yet
possible to present a fine-grained synthesis that traces variation in the
chronology, pace and scale of these developments across Latin Europe,
and differences of emphasis in the texts acquired by individual houses.
Further work is also required to establish the extent to which particular
patterns of acquisition and practice can be identified that were common to
either a particular region or monastic affiliation, or that point to other,
perhaps more personal networks.4 For the most part, the holdings of
religious houses across Latin Europe during this period comprised
a common core of the same or closely similar texts. Significant variation
from house to house is most likely to have been either one of scale,
conditioned by differing economic circumstances, or of scope, perhaps
largely determined by the interests of individual members of the commu-
nity or external benefactors, whose personal collections (discussed in
Chapter 5) often made their way into the communal holdings. This chapter
103
104 teresa webber
goes beyond the individual volumes and categories of text (which are
examined in other chapters) to focus upon the overarching concept of
a library in this period, and to investigate whether the expansion in the
scale and scope of the holdings of religious houses was accompanied by any
significant change in the ways in which they were perceived and used. This
is a question which has rarely been posed, but which is fundamental to any
inquiry that uses manuscripts and their contents as evidence for the
spiritual and intellectual activities of religious communities.5 It will be
addressed in this chapter, first, through an examination of the terminology
used to refer to the communal holdings, in conjunction with the material
evidence for their physical organisation, and second, through an analysis of
the evidence provided by monastic customaries and other written sources
concerning those to whom the custody of the books was assigned and their
range of responsibilities. These material and written sources reveal the
presence of shared practices and perceptions within the different monastic
traditions and across Latin Europe. They also show that, despite develop-
ments in the storage of the books (in response to their increasing numbers),
there was underlying continuity throughout this period in the understand-
ing of what constituted a library: a resource that served the liturgical and
devotional as well as the intellectual requirements of the community.
Furthermore, of all these requirements, it was the provision of books for
the communal observances in the choir, chapterhouse and refectory that
was given most emphasis in the duties of the person to whom custody of
the library was assigned.

The Library as Idea and Physical Entity in the Twelfth Century

Twelfth-century references to libraries as physical structures or discrete
spaces are rare. William of Malmesbury provides an unusually evocative
example in the preface to his Abbreviatio Amalarii, in which he recalls the
occasion that prompted him to undertake the task of abbreviating
Amalarius of Metz’s lengthy, allegorical exposition of the liturgy, at the
request of a junior monk named Robert:
For lately, when we were sitting in our library [‘bibliotheca’], and each of us
turned the pages of our books for the purpose of study, you were struggling
with Amalarius’ De ecclesiasticis officiis. Since you recognised the subject-
matter from the very first sight of the title, you embraced the opportunity to
breathe life into your training in the rudiments of your new profession, but
the complexity of its interpretations and the difficulty of its vocabulary
affected your spirit’s zeal, and you asked me to make a digest of it.6
 The Libraries of Religious Houses 105
The vignette is unusual for its description of a library as a physical space in
which books were read, rather than as a term for the holdings themselves.
The ‘bibliotheca’ to which William refers, however, is most unlikely to have
been a room set aside specifically for books and reading, but was almost
certainly simply an area within the cloister galleries in close proximity to the
cupboards housing the community’s principal book collections. Library
rooms or buildings specially designated for both the storage and consultation
of books had been a creation of the ancient world,7 but were not reintroduced
to Western Europe until the thirteenth century, and were widely adopted by
monastic communities only in the fifteenth.8 Nevertheless, the great libraries
of antiquity, such as the public and papal libraries of Rome, were known to
William and his contemporaries from such texts as Isidore’s Libri etymolo-
giarum (Book VI. 3, ‘De bibliothecis’) and the histories of Suetonius and
Eusebius. This knowledge may perhaps explain the evocation of a library in
apparently discrete spatial terms, even if this was not the physical reality.
The heading to a twelfth-century catalogue of the books of the Norman abbey
of St-Évroult, for example, states: ‘These are the books which are contained in
the library [‘in bibliotheca’] of St Évroult’.9 The synonym, ‘librarium’, was
also used in this way, as in a late twelfth-century record from the Benedictine
community of Michelsberg, near Bamberg: ‘Hi sunt libri quos Rutgerus in
librario invenit sub Wulframo abbate’ (‘These are the books that Ruotger
found in the library during the abbacy of Wulfram’ (1172–1201)).10 The more
physically dispersed reality, however, is made explicit in a late twelfth-century
set of customs from the English Benedictine abbey of Abingdon, in which the
‘bibliotheca’ as a whole is assigned to the custody of the cantor, but more
specific reference is also made to the separate cupboards (‘almaria’) in which
the books were stored.11 This physical reality is also reflected in the continued
widespread use of the term ‘armarium’ (or variant forms) also to refer
collectively to a community’s books,12 as well as its derivative, ‘armarius’, to
designate the office of the person given custody of the books.13 The customs
compiled by Abbot William (d. 1091) for the German abbey of Hirsau, for
example, and his principal source, the Cluniac customs recorded for him by
Ulrich of Zell, explain that the title ‘armarius’ derives from the fact that the
library (‘bibliotheca’), which was customarily in the hands of this officer, was
also known as the ‘armarium’.14 Books, however, were understood to be so
central to the spiritual economy of the community that, as with the other
essential resources – the buildings, relics, vestments, altar vessels and other
furnishings – a special term was not always required. It was sufficient to
describe them simply as ‘the books’, as in the following headings: ‘Isti sunt
libri ecclesiae sancti Nycolai’ (a mid-twelfth-century list of the holdings of
106 teresa webber
the Augustinian canons of St Nicholas, Passau),15 and ‘Hii sunt libri qui
continentur in Radingensi Ecclesia’ (the formal record of the books of the
English Benedictine abbey of Reading included in its late twelfth-century
cartulary).16
The physical locations for books reflect their centrality to the communal
religious life. By the twelfth century, throughout much of Latin Europe, they
were usually kept in the cloister, at or near the junction between the outer wall
of the nave of the church and the south (or north) transept (depending on
local topography).17 The four-sided cloister as the usual physical arrangement
at the heart of the monastic enclosure and the principal site of communal
monastic observance was first articulated during the Carolingian monastic
reforms of the early ninth century, but did not immediately become widely
adopted.18 During the eleventh and twelfth centuries, the spread of monastic
observances inspired by both the practices and written customs of houses
associated with various currents of monastic reform that had begun in the
tenth century (especially those of Cluny) and by those of the newly emerging
orders and congregations of monks and regular canons – the Hirsau con-
gregation, the Cistercians, Carthusians, Victorines, Premonstratensians and
Arrouaisians – contributed to its widespread adoption throughout Latin
Europe. It was positioned with one range adjacent the nave to the south or
the north, the east range incorporating the south (or north) transept, with the
chapterhouse beyond, and the refectory on the far side, facing the church.
This arrangement contrasts with the more dispersed structures of the earlier
Middle Ages, and which continued to characterise, for example, Irish mon-
astic enclosures until the arrival of the Augustinian canons and Cistercian
monks in the twelfth century.19 The cloister, together with the buildings that
opened onto it, thus provided an enclosed space within which the daily
communal observance took place, and was perceived in various ways to
represent the spiritual ideals of the monastic life.20
Later rebuilding and, in many cases, subsequent destruction have oblit-
erated the twelfth-century evidence at the great majority of monastic sites,
while much of what survives has not been (or cannot be) very closely dated.
Thus far, extensive archaeological and architectural study has been con-
fined largely to Cistercian houses in France.21 Nevertheless it is clear from
this evidence, and from studies of individual houses of other monastic
traditions, that arrangements for the storage of books differed from house
to house and were subject to modification over time.22 The first phase of
building at new foundations may have comprised, wholly or partially,
wooden structures, with or without a full cloister.23 The books were
generally stored in one or more recesses (with grooves and rebates to
 The Libraries of Religious Houses 107
accommodate shelves and doors) set into either the west-facing wall of the
transept or adjacent sacristy/vestry, as, for example, at the Cistercian
abbeys of Boquen (Côtes-d’Armor) and Escaladieu (Hautes-Pyrénées),24
or the outer wall of the nave, as at La Garde-Dieu (Tarn-et-Garonne)
where a series of six twin round-arched recesses with grooves for shelves
survived until 1984.25 The dimensions vary, but none of these recesses is
large: for example, at the Cistercian house of Silvanès (Aveyron) they are
3 m high by 2.85 m wide and only 0.75 m deep,26 and at Escaladieu
(Figure 6.1), 1.57 m high, 0.96 m wide and 1.30 m deep.27 The earliest
principal repository for books at the Augustinian house of St-Victor in
Paris seems to have consisted of one or more recesses in a stone wall: its
customary, the Liber ordinis (dating from before 1139) prescribes that the
book-cupboard (‘armarium’) be lined with wood in order to prevent
damage from damp.28 The lack of any contemporary lists of the contents
of a surviving wall recess makes it difficult to establish the number of
volumes that could be accommodated. Estimates must take into account
the fact that, during the Middle Ages, books were stored flat rather than
upright, and must allow for the varying dimensions of different kinds of
volume and different forms of binding: simple wrappers or wooden boards,
the latter perhaps with projecting metal furnishings or over-covers.29

Figure 6.1 Recesses for books in the cloister at Escaladieu. Photograph by the
author.
108 teresa webber
It is not known whether these wall recesses were supplemented by
free-standing wooden chests or cupboards in the cloisters or close by in the
church or elsewhere.30 No such furniture dating earlier than the thirteenth
century is known to have survived. The late eleventh-century Chronica maiora
of the Benedictine abbey of St-Wandrille (Seine-Maritime) contains
a pictorial representation of a chest with its lid propped open, in
which books are visible, but its location is not depicted, and the image may
reflect an older pictorial tradition rather than contemporary practice.31
Supplementary arrangements are likely to have differed from community to
community in response to local needs. A set of glossed books of the Bible
acquired by Abbot Simon (1167–83) for the abbey of St Albans, for example,
was accommodated in a special painted cupboard Roger the hermit in the
abbey church near the shrine of Roger the Hermit, providing both secure
storage and a visually prominent means of ensuring the commemoration of the
donor.32 It was also common practice for certain kinds of liturgical book to be
kept elsewhere. The most precious books – those with treasure bindings and
other volumes used by those performing a sacerdotal role (as celebrant, deacon
or subdeacon) in the liturgy of the Mass – were usually kept with the other
liturgical ‘ornamenta’ in the sacristy/treasury.33 Volumes used daily in other
locations, and not required for any other purpose, might also be kept in situ.
The late twelfth-century catalogue of the books of Reading Abbey records the
presence of breviaries in the ‘capella claustri’, the ‘capella abbatis’, the guest-
house and the infirmary, as well as a plenary missal, sacramentary, gospel book,
epistolary and two graduals in the ‘capella abbatis Ioseph’, and a plenary missal,
sacramentary, epistolary and breviary in the ‘capella abbatis de Hida’.34
During the second half of the twelfth century, the holdings of some
communities had grown so much as to necessitate new arrangements to
accommodate significantly larger numbers of books than the wall recesses.
These too were located at or near the junction of the church and the transept.
At the Cistercian house of Fontfroide (Aude), for example, a small room,
3.20 m wide and 2.75 m deep, was formed in the space under the stair
connecting the church to the dormitory at the far east end of the south
cloister gallery.35 A more widely adopted solution, and one that allowed for
the creation of a larger room, was to customise part of the space in the east
cloister gallery between the end of the transept and the chapterhouse. Where
a sacristy or vestry already existed here, the part of the chamber fronting onto
the cloister was adapted as a book room (Figure 6.2). At La Garde-Dieu, for
example, a room 5.80 m wide and 3.70 m deep replaced or supplemented the
recesses in the west-facing wall of the transept.36 Cistercian houses were not
the only ones to deploy this area in such a way. At the cathedral priory of
 The Libraries of Religious Houses 109

Figure 6.2 Ground plan of Roche Abbey. Adapted from R. Gilyard-Beer, Abbeys:
An Introduction to the Religious Houses of England and Wales (London, 1958).

Christ Church, Canterbury, a passageway between the end of the transept

and the chapterhouse was closed off at the east end to form a book room
sometime between 1160 and the 1220s.37 Local circumstances sometimes
required more unusual solutions. At Winchester Cathedral Priory it may
have been a lack of ground-floor space beyond the end of the south transept
110 teresa webber
that led to a room for the storage of books being built onto it at first-floor
level. Eight arched recesses (1.4 m in height) were constructed within the
recesses of the two earlier windows, replacing or supplementing a series of
recesses (now filled in) at ground-floor level in the wall between the entrance
to the chapterhouse and the church.38 Despite their greater size, the new
book rooms functioned in essentially the same way as the earlier wall recesses:
as secure places of storage, under the custody of a designated officer.39 Their
introduction need not have been accompanied by a significantly different
conception of their holdings.
The organisation of books within the cloister recesses and book
rooms is extremely difficult to establish. Elements of author or subject
classification in a number of twelfth-century booklists cannot be assumed
to reflect spatial organisation and may only represent long-established
hierarchies of authority or textual categorisation.40 Nevertheless, two late
twelfth- or early thirteenth-century booklists from Rievaulx Abbey may
provide evidence that such concepts and categories could have had
a bearing (at least in part) upon the groupings of volumes upon the shelves.
In both lists the entries are arranged in alphabetically labelled sections,
which perhaps denote individual cupboards within the book room – an
arrangement made explicit in a late medieval catalogue from the
Yorkshire Cistercian abbey of Meaux.41 The sections labelled B and
C comprise volumes containing Augustine; section D, Bernard,
Anselm and Ailred; section I, Bede; section N, homiliaries and passionals,
and section P, glossed books of the Bible. The contents of other sections,
however, are (or became) more diverse in their authorship and subject-matter.
If these lists do represent the contents of Rievaulx’s book room (which by
this date formed the cloister-facing half of a structure between the
transept and chapterhouse), then they are valuable evidence of the numbers
of volumes that such rooms could accommodate: in this instance, more than
220.42
The evidence of customaries and booklists suggests that patterns of
use rather than textual categories were the more dominant principle
underlying the physical organisation of the books. The late twelfth-century
customs of Abingdon Abbey, for example, refer to ‘almaria puerorum,
iuvenum’, which presumably housed the various kinds of book
used in the education and training of the boys, youths and novices.43
More plentiful evidence, however, survives of the practice of
reserving as a special, non-borrowable collection those books that
were required on a daily basis for the purposes of rehearsing (and then
delivering) the public reading within and outside the liturgy, and for
 The Libraries of Religious Houses 111
the singers to rehearse the chants. These are known in several twelfth-century
customaries as the ‘libri communes’, and the introduction of special
arrangements restricting their use may have been a twelfth-century
development.44 They included bespoke liturgical books (such as
graduals and antiphoners, which contained the chants for the Mass
and the Office), as well as other books whose contents were in
calendrical sequence, such as homiliaries and passionals, but whose
purpose was not exclusively liturgical.45 But among the books read
aloud in the annual cycles of reading in the Night Office and in the
refectory were also the Bible, patristic biblical commentaries and
treatises, and, in the evening collation, the writings and Lives of the
desert fathers.46 These were the same texts that formed the core of the
personal reading of monks and canons; where a community possessed
only a single copy, that volume must have seen more than one kind of
use. In the Liber ordinis of St-Victor (the earliest surviving text to
describe the libri communes in detail), such books are deemed ‘maiores’,
since they were especially useful and necessary for the instruction and
edification of the canons.47 Although their placement among the libri
communes meant that they could not be removed from the cloister, there
was no prohibition on their use there for personal study and devotion.
From their treatment in the Liber ordinis and other customaries, it is
clear that the libri communes were regarded as the most important
collection of books within the wider holdings of the community.
The range of books they encompassed strongly suggests that the modern
conception of a library – as constituting books specifically for personal
study, distinct from all those used within the liturgy and other
communal observances – did not apply in the twelfth century.
The books used in teaching of grammar and the other liberal arts
might comprise a physically discrete collection, but the books used for
biblical study and consideration of theological and doctrinal questions
were not distinguished from those used for oral delivery in the choir,
refectory and chapterhouse; the same volume might serve more than
one purpose. For example, it is not uncommon for copies of certain
patristic texts such as Gregory’s homilies on the gospels or his commen-
tary on Ezekiel to contain marginal annotation denoting passages to be
read aloud in the Night Office of Matins.48 Further evidence for the
absence at this date of a single bipartite division between liturgical and
non-liturgical books survives in the written descriptions of the duties of
and resources allocated to those assigned overarching custody of the
library.
112 teresa webber
The Custody of Books
The period from the late eleventh century onwards is significant for the
widespread adoption of the practice, first traceable in the eleventh-century
Cluniac customaries, of combining the duties of the armarius (the office to
which custody of books and responsibility for the liturgical and
non-liturgical readings had been assigned in monastic customaries of the
late tenth and earlier eleventh centuries) with that of the cantor (the person
with overall responsibility for the sung elements of the liturgy).49 This
combination of roles may have been prompted both by the increasing
practice of recording in writing the melodic component as well as the texts
of the chants and by a heightened concern to assign the correct sets of
proper chants and readings to the appropriate days, weeks, months and
seasons over the course of each liturgical year. This was a highly complex
process of alignment that required adjustment each year in order to
accommodate the variable date of Easter and the different correlations of
calendrical dates with the days of the week.50 It may not always have been
possible to find someone sufficiently expert to fulfil both roles adequately.
Some customaries allowed for a temporary separation of the two roles,
should the cantor not be suited to the office of armarius,51 while the
twelfth-century customs of the canons of Prémontré departed from con-
temporary norms in keeping the two offices distinct.52
Late eleventh- and twelfth-century customaries use either one or both of
the existing titles, ‘cantor’ (or ‘precantor’/‘precentor’) and ‘armarius’, to
describe the combined role,53 and describe a broadly similar range of
responsibilities. The choice of wording frequently reflects textual borrow-
ing between the different religious traditions.54 As far as the library is
concerned, the cantor-armarius was given custody of all the books of the
community. This is either stated explicitly, as in the Cluniac-influenced
customs Lanfranc compiled for his community at Christ Church,
Canterbury, and in the similarly worded Victorine Liber ordinis,55 or by
implication from the responsibilities assigned.56 It is only the closely
related customaries of Ulrich of Zell and William of Hirsau that refer to
the ‘bibliotheca’ in their description of the role of cantor-armarius.57
Nevertheless, the close association between his custody of the books, the
holdings themselves and the library as a physical space can be inferred from
the continued widespread use of the related terms ‘armarius’ and ‘armar-
ium’, as noted earlier. It is also reflected in a thirteenth-century narrative
from Christ Church, Canterbury, which refers to the recently created book
room between the transept and the chapterhouse as the ‘armariolum . . . in
 The Libraries of Religious Houses 113
quo libri praecentoris includuntur’ (‘the armarium in which the precen-
tor’s books are kept’).58
Custody of the books involved not only their distribution on a daily and
(for those communities that followed the practice of the Lenten distribution
as set out in the Rule of St Benedict) annual basis59 but also, when necessary,
their correction.60 Indeed, according to the customs of St-Victor, only the
armarius (or those authorised by him) was permitted to make any alteration
to the books – whether deletions, additions or corrections.61 Documentary
evidence for the allocation of specific resources to the cantor-armarius also
frequently specifies both the correction and repair of the books.62 In the
customaries, however, special emphasis is placed upon the preparation of the
books of chant and those used for public reading. These were to be carefully
corrected and, where necessary, punctuated, so that those to whom they were
distributed each day for rehearsal in the cloister would not stumble when
singing or reading aloud.63 Several customaries couple the cantor-armarius’
duty to correct the books with the armarius’ long-standing responsibility to
show the readers where to end their readings, and himself (or a deputy) to
hear them rehearse.64
Twelfth-century customaries thus make clear that books used within or
in preparation for the performance of the liturgy and other communal
observances were a core component of ‘the library’ for which the cantor-
armarius had responsibility. Nevertheless, one particular category of litur-
gical book came to be excluded from the remit of the cantor-armarius
during the twelfth century. These were the books used by the officiants
(the celebrant, deacon and subdeacon) in the celebration of the Mass, and
which came increasingly to be specified, along with the other ornaments
and furnishings of the church itself, as the responsibility of the sacrist.
The late eleventh-century customs Lanfranc compiled for Canterbury
assign to the sacrist custody of ‘all the ornaments and utensils and furnish-
ings of the church’, but without specifying what they comprised.65 Some
twelfth-century customs, however, do so, and list the ‘textus’ – the term
commonly used to refer to the gospel book carried in procession – alongside
the candelabra and other ecclesiastical paraphernalia.66 Some customaries
go further. The Victorine Liber ordinis includes among the ‘ornamenta’ in
the care of the sacrist not only the gospel books but all of the books used in
the service of the altar, and which were kept in the treasury (or sacristy): the
missals, epistolaries and gospel books.67 In the late twelfth-century customs
from Abingdon Abbey, the books officiants used in the Mass are excluded
from the allowance made to the cantor to inspect books during the cano-
nical hours and Mass.68 An early thirteenth-century customary of the
114 teresa webber
English Benedictine abbey of Eynsham repeats this allowance and its
restriction word for word, and then makes its implications explicit by
reserving to the care of the sacrist all of these books (missals, gospel books
and epistolaries), as well as lectionaries and books for the guest house, and
benedictionals.69 Further evidence that this category of book was conceived
as a part of the sacristy and distinct from the library survives in a catalogue of
books drawn up in 1165 at the German Benedictine abbey of Prüfening.
It provides a commemorative record of those things in the sacristy (‘sacrar-
ium’) and library (‘armaria’) that had been acquired through the gift of
divine grace and the munificence of the abbey’s founder and abbots.70
The books of the sacristy (the missals, benedictionals and other altar
books) are listed first, together with the other ecclesiastical objects and
furnishings, and are described as representing the clothing (‘togas’) of
spiritual joy and sacred devotion – evoking the visual element of their role
within the liturgy. The importance of the books of the armaria, however,
lay wholly in their contents, which are described as constituting the spiritual
arms of the community.71

Claustrum sine armario, quasi castrum sine armamentario

The Prüfening catalogue’s use of the punning metaphor of arms (‘arma’)
for the books in its armarium echoes the saying included in a letter of
a Norman Augustinian canon, Geoffroy de Breteuil (d. 1194): ‘A cloister
without a library is like a castle without an armoury’. Although evidence
has yet to be found for the wider circulation of this dictum,72 its sentiment
can be regarded as axiomatic. Its assertion that a library was essential to the
proper functioning of a religious community is all the more self-evident
given that twelfth-century conceptions of a library continued to encompass
all (or almost all) of the books required to fulfil the precepts of the Regula
Sancti Benedicti and the Regula Sancti Augustini (one or other of which
underpinned the religious observance of all communities of monks and
regular canons by this date). Nevertheless, although the fundamental
conception of a library had not changed, various impulses contributed to
stimulate more intense activity in recording the communal ownership of
the books which comprised it.73 The various means employed – inscrip-
tions of ownership and donation in the books themselves; inventories of
the holdings; narrative records of donation and acquisition embedded
within local, national or even universal histories – all have their precedents
in the earlier Middle Ages (and especially during the Carolingian period),74
but the cumulative effect of these efforts was to sharpen the distinction
 The Libraries of Religious Houses 115
between institutional ownership and temporary, personal possession (a
distinction fully articulated by the friars in the thirteenth century) and
thus, in principle (if not wholly in practice), endow their library with
a greater level of stability.
One important impulse was the increasingly common and sometimes
acrimonious practice of separating the property and revenues of the abbot
(and sometimes other leading office holders) from those of the community
as a whole.75 Abbots and other members of the community continued to be
recorded in narrative histories and in inventories as donors of books and
other resources to their communities throughout the twelfth century and
beyond, but such records may have played a dual role: a means not only of
commemorating their benefactions but also of providing testimony of the
institution’s ownership. The issue of communal ownership was further
complicated by the increasing numbers of adult recruits who brought
personal possessions (including books) with them. Permission may have
been granted for such men to continue to use their books, but ownership,
in principle, now lay with the institution. Thomas of Marlborough (d.
1236), for example, in his history of his abbey of Evesham, recording the
prolonged dispute over property and revenues between the community
and Roger Norreis, abbot from 1189 to 1213, also recorded his own bene-
factions to the community as prior and subsequently abbot.76 He distin-
guished between the books he brought with him on entering the
community and those he commissioned after becoming prior,77 presum-
ably drawing upon the revenues assigned by custom at Evesham to the
office of prior, specifically for the purchase of parchment and the main-
tenance of scribes.78 Among the former were ‘books of both branches of the
law, canon and civil, which he had used before becoming a monk, when he
lectured in the schools at Oxford and Exeter’.79 He doubtless continued to
make use of these when he presented the abbey’s case to the curia at Rome,
defending its rights of exemption against Mauger, bishop of Worcester
from 1202 to 1212.
A second significant impulse, which became increasingly pressing during
the later twelfth century and beyond, was the recognition of the potential for
books borrowed and taken beyond the walls of the cloister not to be
returned. It may be no coincidence that the earliest and most detailed records
designating the libri communes as a closed, non-borrowable collection are
found in the customs of communities of regular canons, whose practice of
the religious life was more likely to entail activity beyond the abbey walls.
By the second half of the century, growing numbers of monks as well as
canons were required to engage in secular and ecclesiastical affairs (often in
116 teresa webber
defence of the property and privileges of their own communities) or were
travelling to study at the schools of higher learning, thus increasing the
potential for books to stray. The increasingly widespread practice of entering
inscriptions of ownership that incorporated elaborate anthemas may, at least
in part, reflect such concern. By the beginning of the thirteenth century, it
had become a requirement for the canons of St-Victor to record a standard ex
libris inscription in the books they copied since entering the abbey.80
The proximity of St-Victor to the emerging university of Paris, and
a population of masters and students who required access to books, may
well have made the issue of borrowing especially acute for the canons. Some
communities reacted by restricting still further the practice of loans, to the
extent that, in 1212, the Council at Paris forbade such prohibitions (including
the imposition of anathemas), arguing that the loan of books to the poor was
a very special work of mercy.81 Stephen Langton, a Paris theologian and later
archbishop of Canterbury (1207–28), went even further in his commentary
on the fifth commandment, ‘Thou shall not kill’, asserting that ‘not to lend
books is a type of homicide’.82

Conclusion
Any comparison between the contents of the libraries of religious houses at
the beginning and end of the twelfth century will reveal a striking increase
in the number of volumes and the range of their contents. That growth
came about in part through the deliberate enhancement of the existing
resources and in part through the unplanned absorption of the books
acquired by adult recruits, some of whom had studied at the Continental
schools. Nevertheless, a study of the libraries of religious houses that
examines them primarily through the lens of the intellectual developments
that have been characterised as ‘the Twelfth-Century Renaissance’ will be
incomplete and anachronistic. The twelfth century witnessed the culmina-
tion of a process that had begun with the reforms of the Carolingian
period, was reinvigorated during the tenth century and gathered pace
during the course of the eleventh and twelfth centuries: namely the estab-
lishment of a communal library as a more permanent entity, reinforced in
different communities to varying degrees by written expressions of own-
ership. It has been suggested that this in itself constituted a new conception
of the library.83 If so, it was one steeped in long-standing ideals and
precedents.
 The Libraries of Religious Houses 117
Notes
1. Constable 1996, 65, 88–91.
2. Bischoff, Latin Palaeography, 213–8. The book collections of female commu-
nities, or of the nuns and canonesses in double houses, have received less
attention, and the evidence is more exiguous: Webber 2006, 117–8; Beach,
Women as Scribes, 68–84, 105–9; Golding 1995, 177–87.
3. For surveys and overviews at the national level, see Nebbiai-Dalla Guarda
1994; Munk Olsen; Bondéelle 1989; Becquet 1989; Webber 2006; Thomson
2012. For two preliminary regional surveys that illustrate the different chal-
lenges faced and how they might be addressed, see Mews 2002; González 2015.
Much of the evidence for Spain survives only as fragments; this is even more
dramatically the case for the Nordic countries: see, for example, Brunius 2005;
Karlsen 2013. The collation of the evidence of surviving books and medieval
lists of books in institutional ownership is most complete for Britain: Ker 1964
(see also www.mlgb3.bodleian.ox.ac.uk) and the ongoing volumes of
CBMLC. For the booklists from Germany and Switzerland: Krämer 1989–
90 and MBKDS; from Austria, Mittelalterliche Bibliothekskataloge; and from
Belgium, Derolez. For an online portal to digitised materials for the study of
Austrian medieval manuscripts and libraries, see also http://manuscripta.at.
A project to provide descriptive handlists of Italian booklists is under way:
Fiesoli (ed.) 2009–, RICABIM. For the current state of research in France, and
the IHRT’s project, see ‘BiblIFraM’: www.libraria.fr/sites/default/files/Bibli
fram%20programme%20scientifique%20pr%20Libraria.pdf.
4. See, for example, Falmagne 2000.
5. But see also Nebbiai-Dalla Guarda 2013. On common elements and emphases
in different streams of monastic reform and their shared inheritance of late
antique and early medieval ideals, see Constable 1996, esp. 169–208.
6. ‘Nuper enim cum in bibliotheca nostra sederemus et quisque pro studio libros
evolveret, impegisti in Amalarium De ecclesiasticis officiis. Cuius cum mate-
riam ex prima statim tituli fronte cognosceres, amplexus es occasionem qua
rudimenta nove professionis animares, sed quia confestim animi tui alacrita-
tem turbavit testimoniorum perplexitas et sermonum asperitas, rogasti ut
eum abreviarem.’ Pfaff 1981, 128.
7. Casson 2001; Lapidge 2006, 5–22.
8. Gameson 2006; Prache 1989; Lehmann 1957; Clark 1901.
9. BnF lat. 10062, f. 80v, ‘Hec sunt volumina qui in bibliotheca Sancti Ebrulfi
continentur’: Nebbiai-Dalla Guarda 1994, 207 and 255, pl. 4.
10. MBKDS 3/3, 366.
11. ‘Bibliotheca erit sub cantoris custodia’: ‘De obedientiariis’, 373; ‘Cantor
almaria puerorum, iuvenum, et alia in quibus libri conventus reponentur,
innovabit, fracta praeparabit, pannos librorum bibliothecae reperiet, fracturas
librorum reficiet.’ Ibid., 371.
12. For this use of the term between the ninth and twelfth centuries (and
beyond), see Genest 1989, 141–9.
118 teresa webber
13. The late tenth-century customs of Fleury describe the custodian of the books
as ‘armarius’; he was also the schoolmaster: ‘Armarius, qui et scole preceptor
vel librarius’: Consuetudines Floriacenses, 16. For eleventh- and twelfth-century
use of the term, see Vernet 1989, 165–6.
14. ‘Praecentor (qui: om. Ulrich) et armarius, armarii nomen obtinuit, eo quod in
eius manu solet esse bibliotheca, quae et in (in: add. Ulrich) alio nomine
armarium appellatur.’ William of Hirsau, Constitutiones 2. 113; Ulrich of Zell,
Consuetudines Cluniacenses, 748.
15. MBKDS, 4/1, 53, no. 11.
16. CBMLC 4, 421 (B71).
17. For French Cistercian houses with the cloister located to the south of the
church, and those (somewhat less numerous) with the cloister to the north,
see Aubert 1947, 1. 112 n. 1.
18. Horn 1973.
19. Ó Néill 2006, 84–6.
20. Meyvaert 1973; Pressouyre 1973; de Jong 2000; Klein 2004.
21. Aubert 1947, 2. 39–47; Kinder 2002; Robinson 2006. Detailed studies of the
remains at individual Cistercian houses include Swartling 1969; Fergusson
and Harrison 1999; Coppack 2009.
22. The various phases of adaptation are usually no longer visible or yet to be
uncovered. For structural modifications at the Cistercian abbey of Bordesley
revealed by the horizontal stratigraphy of the remains, see Hirst, Walsh and
Wright 1983, 116–22.
23. Coppack 2009, 22–5.
24. Aubert 1947, 2. 40–1, fig. 338; Kinder 2002, pl. 5/iv, B and D.
25. A photographic record is all that remains: Bondéelle 1989, 67.
26. Aubert 1947, 2. 40.
27. Masson 1972, 10.
28. ‘Ipsum autem armarium intrinsecus ligno uestitum esse debet, ne humor
parietum membranas rubigine aliqua siue humectatione aliqua inficiat’: Liber
Ordinis, 78.
29. Note, for example, the forty-odd ‘parvi libelli’ (booklets probably with
just some form of parchment wrapper) among the contents of the ‘mag-
num armarium’ recorded in a late twelfth-century booklist perhaps
from the English Augustinian priory of Bridlington: CBMLC 6, 18–22
(A4.79–118).
30. Vezin 1989; Genest 1989.
31. Le Havre, Bibl. mun. 332, f. 41v: reproduced in Vezin, 364.
32. Matthew Paris, Gesta Abbatum, 59b (184): ‘Whoever wishes to see those books
will find them in the painted cupboard in the church opposite the tomb of
holy Roger the Hermit, where the abbot himself ordered them to be placed.
By contemplating them he will come to understand how great a lover of the
Scriptures the abbot was’ (Thomson, St Albans, 1. 52).
33. See, for example, the inclusion of the gospel books in treasure bindings among
the vestments, vessels and other furnishings rather than with the books ‘in
 The Libraries of Religious Houses 119
armario’ in a mid-twelfth-century description of the possessions of Ely Abbey:
Liber Eliensis, III. 50.
34. CBMLC 4, 193–4, 198–201 (B71.146). For Joseph, abbot of Reading (el. 1173,
resigned 1186, d. 1191), and Thomas, abbot of Hyde (el. 1175, resigned and
retired to Reading, 1180), see Coates 1999, 20–1.
35. Aubert 1947, 1. 41.
36. Ibid., 42–3.
37. Ramsay 1995, 350–1.
38. Oakeshott 1954, 7–8, pls. IIIa–b, IVa. For the adaptation at Valcroissant
(Drôme) of the last bay of the transept as a sacristy and book storeroom, see
Aubert 1947, 1. 174; 2. 44.
39. Nebbiai-Dalla Guarda 1997, 31–44.
40. Nebbiai-Dalla Guarda 1989.
41. CBMLC 3, Z19 and Z20 (Rievaulx); Z14 (Meaux).
42. Fergusson and Harrison 1999, 87.
43. ‘De obedientiariis’, 371.
44. Nebbiai-Dalla Guarda 1996, 258–60.
45. The customary in the Cistercian Ecclesiastica Officia specifies the following:
‘pro communibus libris, scilicet antiphonariis, hymnariis, gradalibus, lectio-
nario, collectaneo, kalendario et illis qui in refectorio et ad collationem
leguntur’. Choisselet and Vernet 1989, 324 (ch. 115); repeated verbatim in
the customs of the canons of Arrouaise: Constitutiones Ordinis
Arroasiensis, 165.
46. Webber 2013, 226–38.
47. ‘Debet etiam armarius inter hos libros, qui ad cotidianum officium ecclesiae
necessarii sunt, etiam de aliis aliquot quos ad instructionem uel ad aedifica-
tionem fratrum magis commodos et necessarios esse perspexerit, in commune
proponere, quales sunt bibliotecae et maiores expositiones et passionarii et
uitae patrum et omeliarii’: Liber Ordinis, 82.
48. For example, Cambridge, St John’s Coll. B. 13 (35), a late eleventh-century
copy of Gregory on Ezekiel from Bury St Edmunds, and Oxford, University
Coll. 191, an English twelfth-century copy of Gregory’s Homiliae XL in
Euangelia of unknown Cistercian provenance, are both annotated with
roman numerals demarcating passages to be read as lections in Matins.
49. Fassler 1985, 39–51.
50. Webber 2017.
51. As, for example, in the customs drawn up in the late eleventh century by
Archbishop Lanfranc for Christ Church, Canterbury: ‘De uniuersis monas-
terii libris curam gerat, et eos in custodia sua habeat, si eius studii et scientie
sit, ut eorum custodia ei commendari debeat’: (‘He [the cantor] takes care of
all the books of the house, and has them in his keeping, if his interests and
learning are such as to fit him for keeping them.’): Lanfranc, Monastic
Constitutions, 122–3.
52. Van Waefelghem 1913, 28–9.
120 teresa webber
53. The Cluniac-influenced customs of William of Hirsau use both terms:
‘Praecentor qui et armarius’ (William of Hirsau, Constitutiones, 2. 113), while
the Victorine Liber Ordinis retains ‘armarius’ (Liber Ordinis, 78).
The Cistercian customary in the late twelfth-century Ecclesiastica Officia uses
‘cantor’ (Choisselet and Vernet 1989, 322 (ch. 115)).
54. For correspondences in the content and wording of the chapter on the office
of cantor-armarius in the customs of the Cistercians, the canons of
Premontré, of Arrouaise and of Oigny, see Lefèvre 1972, 803–7, Coûtumier
de l’abbaye d’Oigny, xlix–lv.
55. ‘De uniuersis monasterii libris curam gerat, et eos in custodia sua habeat’:
Lanfranc, Monastic Constitutions, 122; ‘Armarius omnes ecclesiae libros in
custodia sua habet’: Liber Ordinis, 78.
56. As in the Cistercian Ecclesiastica Officia, which specifies only particular duties
rather than a general responsibility for the custody of the books: Choisselet
and Vernet 1989, 322–4 (ch. 115).
57. ‘in eius manu solet esse bibliotheca’: Ulrich of Zell, 748; William of Hirsau,
Constitutiones, 2. 113.
58. Gervase of Canterbury, 2. 121; Ramsay 1995, 350 n. 48.
59. The practice is also recorded in the customs of the canons of Arrouaise:
Constitutiones Ordinis Arroasiensis, 165.
60. See, for example, the similarly worded customs of the Arrouasians and
Premonstratensians: ‘Ad eum etiam [eum etiam: armarium Premonstr.] perti-
net libros custodire et emendare’. Constituiones Ordinis Arroasiensis, 165; Van
Waefelghem 1913, 28. On the concern for textual accuracy (especially in the
Bible and other texts delivered in the liturgy) as an aspect of twelfth-century
reform ideals, see Constable 1996, 154–5.
61. ‘Nemo alius praeter armarium, siue in his [i.e. the libri communes], siue in aliis
quibuslibet libris uel demere, uel addere, uel mutare quicquam praesumat,
nisi ei specialiter concessum fuerit uel iniunctum’: Liber Ordinis, 82.
62. For evidence from twelfth-century England, see Gullick, ‘Professional
Scribes’, 2–4, and Webber 2017.
63. ‘quos (libros) praecipue armarius diligenter emendare debet et punctare, ne
fratres in cotidiano officio ecclesiae, siue in cantando, siue in legendo, aliquod
impedimentum inueniant’: Liber Ordinis, 81–2.
64. As, for example, in the customs of the canons of Oigny: ‘Ad eumdem pertinet
libros custodire et emendare, et fratribus cum necesse fuerit distribuere;
lectiones ipse, vel alius ab eo monitus, terminare et auscultare’: Coûtumier
de l’abbaye d’Oigny, 49.
65. ‘Ad secretarii officium pertinent, omnia ornamenta monasterii, et omnia
instrumenta et suppellectilem, que ad ipsum monasterium pertinent, custo-
dire; horas prouidere’: Lanfranc, Monastic Constitutions, 122–3.
66. Webber 2017.
67. ‘Ad officium sacristae pertinent omnia, quae in thesauro sunt, custodire:
reliquias et omnia ornamenta altaris et sanctuarii ac tocius ecclesiae, siue in
auro, siue in argento, siue in ostro et palliis et tapetibus . . . calices et textus et
 The Libraries of Religious Houses 121
cruces et thuribula et candelabra et cetera uasa, quae uel ad ministerium,
uel ad ornatum altaris et sanctuarii tociusque ecclesiae pertinent, libros
quoque missales, epistolares et euangelia’: Liber Ordinis, 86–7.
68. ‘Cantori licet sine reprehensione horis canonicis et ad missas in libros inspi-
cere exceptis libris ad officium missae assignatis’: ‘De obedientiariis’, 370.
69. ‘Sacrista tamen de missalibus, evangeliariis, epistolariis curam gerat et lectio-
nariis et libris hospitii et benedictionali’: Customary of Evesham, 164–5.
70. MBKDS, IV/1, 421–7, no. 41: ‘hec sacrario nostro et armarie pertinere recog-
nita sunt, et scriptus est liber hic monumenti in conspectu Domini’.
71. ‘Sed primum quidem spiritalis leticie et sacre devotionis togas cum ceteris
utensilibus evolvamus, deinde libros armarie nostre, in quibus videlicet
arma nostra, non carnalia sed spiritalia, Deo locata sunt, conputemus’.
Ibid., 421–2.
72. It owes its modern popularity to the inclusion of the letter in the Thesaurus
novus anecdotorum, compiled and published by the French Benedictines,
E. Martène and U. Durand ([Paris, 1717], 1. 510–2) and later repr. PL 205.
844–5. Silvestre 1964; Kottje 1982, 125–6 n. 2.
73. Nebbiai-Dalla Guarda 2013, 83–92.
74. McKitterick 1989, 157–205.
75. See, for example, Webber 2017. On relations between abbots and their
communities, within the context of an increasing institutionalisation of
religious communities, see also Constable 1996, 176–83.
76. Thomas of Marlborough, History, 490–9, 502–11.
77. Ibid., 490–3. See also CBMLC 4, 134–8 (B29).
78. Thomas of Marlborough, History, 392–3, 540–1.
79. ‘Iste . . . attulit secum libros utriusque iuris, canonici scilicet et ciuilis, per
quos rexit scolas ante monachatum apud Oxoniam et Exoniam.’ Ibid., 490–1.
80. ‘Districte precipitur omnibus ut in libris quos scribi fecerunt, ex quo
venerunt ad conversionem, titulum communem apponant, hunc scilicet:
Iste est liber Sancti Victoris, etc., et eos de cetero nullatenus alienare presumant’
(BnF lat. 14673, f. 277v): Delisle 2. 227 n. 4.
81. ‘Interdicimus inter alia viris religiosis ne emittant juramentum de non com-
modando libros suos indigentibus, cum commodare inter precipua miseri-
cordiae opera computetur. Sed adhibita consideratione diligenti, alii in
domo ad opus fratrum retineantur, alii secundum providentiam abbatis,
cum indemnitate domus, indigentibus commodentur. Et a modo nullus
liber sub anathemata teneatur et omnia predicta anathemata absolvimus.’
Concilium Parisiense A.D. 1212, Part 1, c. vii, in J. D. Mansi, Sacrorum
conciliorum nova et amplissima collectio (Venice, 1778), 22. 821. Nebbiai-
Dalla Guarda 1996, 254–7.
82. ‘Ergo genere homicidii est quaternos non accommodare’: Oxford, Trinity
Coll. 65, f. 258rb: Smith, 268.
83. Nebbiai-Dalla Guarda 1994, 205: ‘elle va acquérir désormais un rôle propre,
devenant ainsi pleinement représentative des besoins spirituels et culturels de
la communauté’.
 chapter 7

Modes of Reading
Jenny Weston

There has been a tendency in scholarship to juxtapose monastic and

scholastic cultures in the twelfth century. This distinction can largely be
attributed to Jean Leclercq’s seminal 1957 study, L’amour des lettres et le
désir de Dieu, which contrasted medieval monastic and scholastic culture
based on a number of factors, including methods of reading.1 Some
scholars continue to gesture towards this dichotomy by focusing on the
apparent differences between the lectio practised in the monasteries and the
scholastic lectio that flourished in the urban schools.2 Tomas Sodeika warns
that ‘it would be a mistake to identify “scholastic” lectio with the monastic
one’, while Jacqueline Hamesse writes that ‘the new scholastic model’ was
‘totally different’ from the kind of lectio practised in the monasteries.3 Such
distinctions largely stem from the understanding that monastic and scho-
lastic readers were motivated by distinct goals related to their respective
cultural milieu – monks read for devotion, and urban scholars read for
intellect.4 The dichotomy becomes further entrenched when comparisons
are drawn between scholastic lectio and the specific practice of lectio divina –
an integral and often private form of devotional reading practised in the
monastic setting. Whereas scholastic lectio relied on a selective and non-
sequential approach to the text, lectio divina encouraged a comprehensive
and sequential reading, punctuated by personal prayer and
contemplation.5 These notable distinctions perpetuate an assumption
that monastic and scholastic reading were wholly different enterprises in
the twelfth century.
This chapter challenges aspects of this perceived division by broadening
the comparative scope of monastic lectio (beyond the specific activity of
private lectio divina) to include corporate reading practices. A close inves-
tigation of the types of reading that took place during the celebration of the
liturgy and the Divine Office, for example, reveals the regular practice of
reading only select passages of text and often in non-sequential order –
a method that directly aligns with the scholastic model. Further parallels
122
 Modes of Reading 123
can be drawn between the physical design of books used during the liturgy
and those used in the urban classroom. This study is part of a growing
movement among scholars to find ways to reconcile the perceived cultural
divide between the monastic and scholastic milieux.6 It presents a new
premise that reading practice during the twelfth century often transcended
cultural and intellectual traditions. To articulate this point further, I shall
highlight the work of three individuals, John of Fécamp, Anselm of
Canterbury, and Hugh of St-Victor, all of whom presented new reading
strategies that cannot be contained within either the traditional monastic
or scholastic paradigms. Instead, they blended tradition and innovation,
and offered strategies that were applicable to readers from various cultural
contexts. Through the study of reading, this chapter draws attention to the
liminal space between the monastic and scholastic milieux – a space
ultimately shared by monks, scholars, and students alike.

Scholastic Lectio
To demonstrate the commonalities between scholastic reading and forms
of corporate reading in the monastery, I begin with a concise overview of
scholastic lectio and its emergence in the urban schools of the early twelfth
century. During this period, the Latin West experienced a rapid expansion
of intellectual and educational enterprises.7 It has been argued that new
methods in teaching and learning motivated the development of a new
type of reading, now commonly referred to as scholastic lectio.8 One such
teaching method was the popular exercise of quaestio and disputatio, where
two opponents would debate and discuss the solution to an argumentative
problem or idea.9 To prepare, students endeavoured to gather as much
information on a topic as possible, preferably from a range of different
sources. As Leclercq explains, students in the ‘urban schools’ were ‘seeking
important, concise, and interesting extracts for doctrinal studies’.10
Students and scholars were not expected to read a book from cover to
cover, but instead they were trained to mine the text for useful and relevant
passages and to skip over any superfluous material. J. P. Gumbert notes
that readers in this scholastic context preferred to read ‘short passages in
several texts, rather than of one entire text’, and as a result, the ‘habit of
“looking things up” for “reference”, or “consultative literacy”, was born’.11
To make the task easier, some readers chose to compile multiple texts into
single, easily accessible volumes known as florilegia. As Hamesse explains,
these compendia ‘gave what was essential in a work or a topic’ and ‘they
often presented the texts in short, easily memorized sentences’.12 For some
124 jenny weston
modern observers, this method of selective reading presented a drastic
change from the slow and comprehensive kind of lectio that seems to
have dominated monastic book-reading.13

Monastic Lectio
Certainly, clear distinctions can be drawn between scholastic lectio and the
private mode of monastic reading commonly referred to as lectio divina.
While the term lectio divina or ‘divine reading’ can be applied to devotional
reading in general, it is most often used to describe a specific type of
personal reading exercised by monks.14 In many communities, including
those that followed the Rule of St Benedict, each member was given a book
during Lent to read over the course of the year, and time was allotted each
morning, during the various rest periods of the day, and on Sundays to
engage in private lectio.15 Reading at these times has been characterised as
a solitary practice.16 Finding a quiet moment, the reader cleared his mind
of distraction and sought communion with God through the act of holy
reading. To enhance and extract the value of the experience, readers
followed a series of interrelated steps.17 The first step involved the undis-
tracted reading of the text (lectio); the reader would then turn the words
over in his mind, committing them to memory (meditatio); he was also
encouraged to pause from his reading and engage in spontaneous prayer
(oratio); and finally he could embrace a period of contemplation (contem-
platio), reflecting upon the experience and the wisdom he may have
received. This type of reading demanded a patient and focused approach,
and, ideally, the practice was not restricted by the limitations of time but
was perceived as a constant lifelong endeavour that led to the gradual
reception of spiritual wisdom;18 it was meant to involve the comprehensive
reading of whole texts over time. As stipulated by the Rule, each book
issued during the period of Lent should be read ‘per ordinem ex integro’
(straight through, in its entirety).19 Given the comprehensive and deeply
devotional nature of lectio divina, it is easy to see why monastic and
scholastic lectio have been framed as dichotomous practices. However, if
we expand the scope of monastic reading to include corporate modes, we
find less divergence between the two models.

Corporate Monastic Reading

In addition to lectio divina (as described earlier), most monastic communities
also engaged in other forms of reading that were not solitary, comprehensive,
 Modes of Reading 125
or textually sequential. These other reading activities took place each day
when the community gathered together for the Mass, during the eight prayer
cycles of the Divine Office, mealtimes, the daily Chapter reading, and the
evening gathering of Collation. At these times, reading was practised com-
munally with a single lector reading to a congregation of listeners.20 Despite
the fact that most of the community would not have had a book in hand, the
act of listening was still considered a kind of divine reading; listeners were
encouraged to focus on the words and contemplate their inner meaning.21
Unlike private forms of lectio divina, however, the individual listener could
not manage his own experience directly. He was unable to reread passages at
will for the purpose of memorisation (meditatio) or pause from the text to
engage in spontaneous personal prayer (oratio). And, most important for the
context of this study, corporate reading did not typically involve the com-
prehensive reading of entire books. As the following evidence shows, the
readings performed during most communal settings were tied to the events
of the liturgical calendar – a timetable that guided all rituals, prayers, and
activities throughout the year. In order to maintain synchronisation with the
calendar, the reading procedure was necessarily selective and, as such,
directly corresponds to aspects of scholastic lectio.
Since as early as the fifth century, readings performed during the Mass, for
example, demonstrate a varied programme that shifted depending on the
calendar.22 The celebration of the Mass was made up of a series of prayers,
readings, and responses, some of which were fixed (Ordinary) and others that
were subject to change depending on the liturgical calendar (Proper). Part
of the ‘Proper’ of the Mass included the epistle and the gospel readings.23
In a lectionary known as the Liber comicus, copied in the eleventh century
and used by the Benedictine community at the abbey of Santo Domingo de
Silos, we find a list of texts designated for the epistle reading of the Mass
service.24 During the first week of Easter, when Mass was celebrated each day,
the readings were primarily drawn from the Apocalypse and the Acts of the
Apostles.25 The readings for the week, however, did not comprise the entire
text, nor did they follow the original sequence of the text: on the third
ferial day of Easter (Tuesday), for example, the reading is listed as Apocalypse,
Book III, chapters 7–13; the following day (Wednesday), the reading begins
with Apocalypse Book II, chapters 12–7.26 Similarly, the gospel reading
performed during the Mass focused on selections or excerpts from the
text.27 Both the epistle and gospel readings were not meant to be compre-
hensive, nor were they meant to be textually sequential – the lector was
directed to skip certain chapters and focus on material related to the weekly
events, effectively moving backward and forward in the sequence of the text.
126 jenny weston
Selective reading did not just take place during the Mass, but it was also
used to fulfil the scriptural readings during the Divine Office (and in some
cases the readings performed at mealtimes in the refectory).28 The Office
was divided over the course of eight prayer services that took place
throughout the day and into the night: ‘Vespers in the late afternoon
before dusk, Compline after dark and before retiring to bed, Matins in
the middle of the night between midnight and dawn, Lauds around day-
break, Prime in the early morning followed by Terce later in the morning,
Sext at midday, and None in the mid-afternoon.’29 The primary focus
during each Office was the recitation of the Psalms (all 150 Psalms were
supposed to be recited over the course of each week), although the com-
munity would also perform prayers and scriptural readings.30 While most
of these meetings were relatively brief, the Night Office known as Matins
(beginning usually around 2:00 A.M.) was substantially longer than the
rest and involved short scriptural readings (‘lessons’ or lectiones) grouped
together into a set (known as a ‘nocturn’).31 During regular weekdays
(known as ‘ferial days’), Matins typically consisted of one to three lectiones.
On Sundays and major feast days, the Matins reading programme was
expanded to include twelve lectiones (three nocturns, with four lectiones in
each).32 Like the epistle and gospel readings of the Mass, the choice of what
to read for each nocturn depended on the cycle of the liturgical calendar,
which changed on a regular basis.33 For example, the first nocturn readings
were typically meant to be taken from the Bible, with the goal of reading
the entire Bible over the course of the year. This programme, however, was
consistently interrupted by special readings on feast days and other special
occasions that demanded a new set of texts.34 These temporary transitions
to alternative textual material required the reader to close one book and
open another; they also required the reader to access material from multi-
ple books. In order to link the content of scriptural readings during the
liturgy with the Christian calendar, the lector relied on a selective pro-
gramme of reading – choosing passages that matched the theme of the
season and the day of the week, which was then recited for the benefit of
the listeners.
Broadening the scope of monastic lectio to include both private and
corporate reading enabled the monks to approach the task of reading in
more ways than one. The fact that certain types of monastic lectio included
selective reading also reduces the perceived division between the monastic
and scholastic cultural milieux in the twelfth century. Although readers
from each context may have been motivated by different objectives, at
times they both relied on the same practical procedure – just as urban
 Modes of Reading 127
scholars searched for specific passages of text related to the quaestio and
disputatio, monastic readers focused on specific chapters and passages
related to the liturgical calendar. What all of these readers shared was
a reliance on selective reading. This observation contributes to our ongoing
efforts to identify common threads that bind together two seemingly
distinct cultural and intellectual contexts.

The Manuscript Book

Further parallels between the two contexts can also be drawn if we take
into account the surviving manuscript record. It has been argued that the
scholastic lectio practised in the urban schools also prompted the devel-
opment of a new book format during the twelfth century.35 The physical
book needed to support an increasingly complex system of teaching,
learning, and reading; it not only needed to present the text, but it also
needed to function as a searchable repository of information that could be
quickly accessed by masters and students engaging in scholastic activities
like quaestio and disputatio. Hamesse comments that the twelfth-century
‘scholastic reader’ wished to ‘find what he was looking for with ease,
without having to leaf through pages’.36 Richard and Mary Rouse have
also demonstrated that by the late twelfth century it was common to find
books equipped with a variety of tools for ‘finding’, such as a ‘clearly
displayed text, with its chapter lists, running headlines, and marginal
apparatus’.37
In the context of selective reading, however, monastic readers also relied
on many of the very same book formats and navigational tools that were
commonly used by readers in the scholastic milieu.38 By the eleventh and
twelfth centuries, it was common for liturgical material to be presented in
one of two formats: either a compendium of collected excerpts arranged in
liturgical order (such as lectionaries, epistolaries, missals, and breviaries)39
or ‘whole texts’, arranged in biblical order, equipped with navigational
reading aids, such as chapter tables, chapter numbers, running titles,
paragraphs, or lection marks that helped the reader locate desired
material.40 The development of liturgical books that contained excerpted
readings assembled according to the calendar indicates that selective read-
ing was commonplace in the liturgical readings, and some communities
wished to limit the time required to locate designated material without
having to flip through multiple volumes. There seems to be a direct parallel
here to the use of florilegia in the urban schools, where students and
128 jenny weston
scholars collected excerpts of important works together to avoid having to
refer to multiple texts when preparing an argument.
Not all monastic communities chose to use collections of excerpts during
the celebration of the liturgy, however, but instead preferred to use whole
texts. A representative study of surviving manuscripts from the Benedictine
abbey of Fécamp demonstrates that many of the books designated for use
during the celebration of the Mass or Office feature navigational aids
embedded into the design of the book.41 A Giant Bible produced at
Fécamp ca. 1060–80, for example, contains a large collection of readings
suitable for the liturgy. The text is equipped with chapter tables (Figure 7.1)
at the opening of each new book (listing the relevant incipit headings and
associated chapter numbers), as well as running titles, marginal lection marks,
and navigational paragraph marks placed throughout the text (Figure 7.2).42
The imposing size of the book – 495 mm × 350 mm – further indicates
that it was designed to be stationary (such as on a lectern) and not likely
intended for the private activity of lectio divina.43 Similar examples of
books equipped with navigational reading aids (and with texts suitable
for the readings of the Mass and Office) can be found in collections from
other Norman Benedictine monasteries including Jumièges, Mont-St-Michel,
and St-Évroult.44
These navigational features are most often associated with the type of
reading pursued in the scholastic context. However, it is important to
acknowledge that they also regularly appear in monastic books that were
used during the liturgy and Office readings, at which times selective reading
was practised. Monastic and scholastic readers, therefore, did not just have
selective reading in common, but they also relied on many of the very same
tools and principles of book design to accommodate this specific approach to
the page.

Individual Approaches
As demonstrated thus far, some elements of reading practice during the
twelfth century seem to have transcended cultural and intellectual tradi-
tions. This premise is further exemplified in some of the individual
approaches to reading practice that extended across traditional boundaries
of the monastic and scholastic milieux. Expressed in various treatises,
letters, and prologues, these individual approaches further complicate the
general picture of reading in the twelfth century and show it to have been
a dynamic activity that shifted along with developing theories of spiritual,
personal, and doctrinal inquiry.
 Modes of Reading 129

Figure 7.1 Chapter table. Rouen, Bibliothèque municipale, 1, f. 17r.

130 jenny weston

Figure 7.2 Chapter numbers and lection marks. Rouen, Bibliothèque municipale, 1,
f. 19r.
 Modes of Reading 131
John of Fécamp
Writing in the eleventh century, John of Fécamp (ca. 990–1078) presented
a slightly modified understanding of lectio divina that would influence later
authors and readers.45 In his work, titled the Confessio theologica (written
ca. 1018), John offered an interpretation of reading largely modelled
upon Augustine’s Soliloquia and Confessions, which included a deeper
sense of personal interiority and self-examination achieved through read-
ing and prayer.46 More specifically, John adjusted his conception of
meditatio in the scheme of lectio divina. Traditionally, meditatio involved
the repeated reading of text for the purpose of memorisation and rumina-
tion. For John, however, this stage was conceived as an opportunity and
means to engage in deep, self-reflective prayer.47 While earlier descriptions
of lectio divina also accommodated spontaneous prayer (oratio), for John
meditatio itself would become a form of prayer achieved through ‘a move-
ment of active, personal implication in the reading’.48 As Brian Stock
explains, ‘devotional reading becomes a type of reflective thinking and
writing in which self and text are closely integrated’.49
What was also new about John’s approach was the type of reader and type
of reading procedure he imagined for this kind of prayer-focused devotional
reading. In a prefatory letter added to one of John’s compilations, the Libellus
de scripturis et verbis patrum (largely drawn from the Confessio theologica),
John first addressed an unnamed nun, and then rededicated the work to
Agnes of Poitou (ca. 1025–77), the widow of Holy Roman Emperor Henry
III.50 The dedication of the work to an aristocratic lay woman suggests that
John envisioned a wider audience for his writings that extended beyond
religious professionals and male readers. It is in this prefatory letter that John
also explained how the book should be read. First, he asked for the text to be
read ‘reverently’ and meditated upon with ‘due fearfulness’.51 He then added
that it should be read frequently, ‘especially when you feel that your mind
has been touched by the desire for Heaven’.52 Although John’s vision of how
the text should be read was rooted in the principles of lectio divina, the
method he described also implied a more leisurely and personal approach to
the material than previously encountered, as well as an activity that may have
appealed to non-monastic readers.53

Anselm of Canterbury
Following the direction initiated by John of Fécamp, Anselm of
Canterbury (ca. 1033–1109) also presented an approach that placed greater
132 jenny weston
emphasis on the close relationship between reading, prayer, and
interiority.54 Anselm expressed these connections most clearly in his earlier
devotional writings, in particular his popular collection of prayers and
meditations (Orationes sive meditationes).55 Like John’s writings, Anselm’s
Prayers and Meditations turned the focus of meditatio away from the
memorisation of words and towards the rumination of ‘ideas and
feelings’.56 By invoking the reader to apply critical examination of the
feelings presented in the text, Anselm also opened a path to personal
introspection. This ‘movement towards interiority’ and subjective
response as a result of reading is considered a major development in
processes of spiritual and personal inquiry in the eleventh and twelfth
centuries.57
Also like John, Anselm sent his collection of Prayers and Meditations to
a variety of individuals, including aristocratic lay women and fellow
monks: in 1072, he sent an early selection to Princess Adelaide, daughter
of William the Conqueror; in 1073–4, he presented three prayers to his
friend Gundulf, monk at the abbey of Bec; and in 1104, he sent the entire
collection to Countess Matilda of Tuscany.58 In a preface to the collection,
Anselm explained the purpose of the texts (‘to stir up the mind of the
reader to the love or fear of God, or to self-examination’), and he explained
how they should be read: not in ‘turmoil, but quietly’, not ‘skimmed or
hurried through, but taken little at a time, with deep and thoughtful
meditation’.59 Anselm deviated from earlier descriptions of lectio divina,
however, in his description of how much to read and where to begin:
The reader should not trouble about reading the whole of any of them, but
only as much as, by God’s help, he finds useful in stirring up his spirit to
pray, or as much as he likes. Nor is it necessary for him always to begin any
one always at the beginning, but wherever he pleases.60
Like the instructions provided by John, Anselm’s reader is invited to
browse the text ‘at leisure’ and ‘at will’, reading only as much as needed
to inspire prayer.61 The potentially selective aspect of this approach also
resembles the type of reading performed in the urban schools, where
students typically read only as much as was needed to support the devel-
opment of specific arguments or ideas. This more abstracted method also
reflects a departure from the comprehensive private lectio described by the
Rule of St Benedict. Of course, it is not likely that Anselm intended this
method to be applied to all forms of devotional reading; after all, he
describes this approach in relation to the reading of his own Prayers and
Meditations and not necessarily to the reading of Scripture. That said,
 Modes of Reading 133
Anselm’s method reflects an emerging sense of freedom among authors and
readers to adjust traditions in order to accommodate new sensibilities, and,
in some cases, new types of readers.

Hugh of St-Victor
Another individual who initiated a major shift in reading practice was
Hugh of St-Victor (ca. 1096–1141), an Augustinian canon and master at
the school of St-Victor in Paris. In his Didascalicon, Hugh presented
a comprehensive survey of knowledge that also included lengthy instruc-
tions of what, why, and how to read.62 Like John and Anselm before
him, Hugh adjusted the focus of meditatio in his scheme of devotional
reading. While he expressed an interest in penitential prayer and self-
examination, Hugh added a new analytical layer to the practice of
meditatio that had not been exercised before.63 As he explained in the
Didascalicon, Book III, ‘Meditation is sustained thought along planned
lines; it prudently investigates that cause and the source, the manner and
the utility of each thing.’64 Instead of using the stage of meditatio as
a means to memorise the text, Hugh perceived it as an opportunity to
analyse the text, using intellectual and rational lines of investigation.65
In some ways, this approach might best be described as a hybrid mode of
monastic and scholastic lectio, which was motivated by an emotional
desire for ‘penitential purification’, but which also included the scientific
evaluation of text.66

Conclusion
Nearly thirty years after Leclercq penned his foundational study,
L’amour des lettres et le désir de Dieu, his view on the relationship between
the monastic and scholastic milieux underwent a profound shift – where
he had once seen opposition, he now also saw commonality. Writing on
the topic of theology, Leclercq proposed a new challenge: ‘how to fit
“monastic within scholastic” and “scholastic within monastic”’.67 This
study responds to his call by focusing on reading practices in the eleventh
and twelfth centuries. It challenges assumptions that monastic and
scholastic lectio were wholly distinct modes of reading, and instead
draws attention to the shared practice of selective reading and related
elements of book design – that is, how aspects of monastic lectio existed
within scholastic lectio.
134 jenny weston
The similarities in reading practice not only help to diminish the
perceived division between these two cultural settings, but they also
prompt further questions about the emergence of selective reading in the
urban context. In particular, if we understand that selective reading was
a practice that had been deeply entrenched in the daily rituals of the
monastic tradition for centuries,68 then is it possible that the monastic
environment served as a direct influence for the development of selective
reading in the urban schools? The established view is that changes to
programmes of teaching and learning sparked the use of selective reading
in the scholastic context. But could the basic tenets of this approach have
been (at least partially) based on the liturgical and Office readings regularly
performed in the monastic setting? Indeed, many scholars and students
who lived and worked in the urban context would have been familiar with
the readings performed during the celebration of the Mass and may have
witnessed the practice of selective reading. Others who received training in
the monastic school or who spent time as community members may have
had even more direct experience with this mode of reading. Peter Abelard
(1079–1142), for example, was a popular master in Paris in the early twelfth
century and a proponent of the quaestio method; he also lived for a time as
a monk at the abbey of St-Denis.69 While this hypothesis remains spec-
ulative until a more conclusive study can be realised, it would help to
explain the rapid rate at which selective reading emerged and came into
regular use in the urban schools of the twelfth century.
Although the relationship between the monastery and the development
of scholastic lectio demands further exploration, this chapter firmly estab-
lishes that readers in the twelfth century often blended convention with
newer developments, as well as monastic ideals with scholastic practice.
The development of individual approaches to reading that suited personal
interpretations of pedagogy and practice is reflective of a growing general
interest in notions of ‘selfhood’ and the ‘individual’ that arose in this
period.70 The subtle fusion of monastic and scholastic ideals, as well as
the invitation of new audiences to engage in devotional reading, also shows
that reading practices were less bound by cultural context than previously
imagined. While later generations of readers would continue to diversify
and update their practices, a rich spectrum of reading already existed in the
eleventh and twelfth centuries.71 Through the merger of tradition and
innovation, as well as monastic and scholastic methods, readers could
choose from a myriad of practices designed to suit their spiritual, intellec-
tual, corporate, or personal needs.
 Modes of Reading 135
Notes
I would like to extend special thanks to Erik Kwakkel, who provided critical
guidance as I developed many of the ideas presented in this chapter. I am also
grateful for advice received from Rodney Thomson and Teresa Webber.
1. Leclercq 1947. While this work was not originally intended for an academic
audience, it continues to serve as a foundational text for those who study
medieval reading.
2. See Studzinski 2009, 15; Newman 1996, 37; Rahner 1975, 1540; and n. 3.
3. Sodeika 2005, 24; Hamesse 1999, 104.
4. Leclercq 1947, esp. 72 and 84; Studzinski 2009, esp. 15 and 17; and Ferruolo
1985, 50.
5. For a general overview of scholastic lectio, see Hamesse 1999; Rouse and Rouse
1982; Parkes 1976; Illich 1993, Vineyard; Southern 1979; Hankins 1990, 20–1.
For discussions of lectio divina in the monastery, see Leclercq 1947, 2012;
Irvine 2010; Edsal 2000; Sandor 1989; Studzinski 2009; Robertson 1996;
Casey 1995; Illich 1993, Vineyard; Pennington 1998; Vandenbroucke 1966;
and Weston 2015.
6. Leclercq himself would later reduce the opposition he initially applied to the
monastic and scholastic milieux. See Leclercq 1986, esp. 194; also Mews 2000,
2007; and Lefler 2014.
7. See Southern 1982, 1995; Ferruolo 1985; Verger 1999.
8. Hamesse 1999; Parkes 1976; Gumbert 2009; Leclercq 1947; Rouse and Rouse
1982; Illich 1993, Vineyard.
9. For a general overview of the scholastic programme of learning, see
‘The Scholastic Method’ in Seel 2012, 1170. For quaestio, see Marenbon
2004, 23; Elders 2003; Lawn 1993; Grabmann 1909.
10. Leclercq 1947, 182.
11. Gumbert 2009, 232; Gumbert’s quotations refer to Parkes 1976, 35. Gumbert
points out that this form of reading was not exclusive to this period: ‘It should
be clear . . . that these are not absolutes; neither way was exclusive in its period
(nor did the scholars who described these ways of reading imagine they were).’
12. Hamesse 1999, 107.
13. As Richard Rouse and Mary Rouse argue, this was not a context in which
students pursued ‘reflective reading’, but instead they read their books with
the intention of ‘seeking out specific information’ (Rouse and Rouse
1982, 206).
14. See Rees 1978, 265; Studzinski 2009.
15. St Benedict, Rule, ch. 48. Depending on the time of year, it has been
calculated that Benedict allocated more than three hours a day to this practice;
Studzinski 2009, 123. Other religious orders and groups that did not follow
the Rule of St Benedict, including the Carthusians and Augustinian canons,
also practised forms of lectio divina.
16. Lectio divina was often a solitary activity, but on occasion monks would sit
together to read. Even then the practice was still relatively private, as monks
136 jenny weston
were encouraged to read quietly to themselves so as not to disturb others. St
Benedict, Rule, ch. 48.
17. The Carthusian monk and prior of Grande Chartreuse monastery Guigo II
(d. ca. 1188) likens each of these steps to the rungs on a ladder that extend into
the heavenly realm: as the reader completes each step of the lectio divina
process, he moves higher on the ladder and closer to perfection. He also
explains the synthesis between these four steps using a common metaphor of
alimentation. See Guigo II, The Ladder of Monks, 82–3.
18. According to Cassian, the monk’s desire to read the Word of God is never
satiated. See Cassian, Conferences, 10. 2 (514).
19. St Benedict, Rule, ch. 48. See also Studzinski 2009, 125; Kardong 1996, 391–2;
Casey 1995, 5–11, esp. at 9; Holzherr 1994, 232–4; William of St-Thierry,
The Golden Epistle, XXXI. 120 (51); Edsall 2000, 50.
20. For further details about corporate reading and the oral delivery of texts, see
Harper 1991; Boynton 2011; Webber 2013, 2014, 2015.
21. Guigo II writes that ‘listening is a kind of reading, and that is why we are
accustomed to say that we have read not only those books which we have read
to ourselves or aloud to others but also which our teachers have read to us’
(Edsall 2000, 62). In St Benedict, Rule, ch. 4, titled ‘Quae sunt instrumenta
bonorum operum’ [What are the instruments of good works], Benedict writes
that monks ought ‘Lectiones sanctas libenter audire’ [To listen gladly to holy
reading].
22. Lampe et al. 1969, 225, make clear that by the fifth century this practice of
continuous reading during the Mass was gradually interrupted to incor-
porate various feasts and celebrations, such as the Epiphany, Easter,
Ascension, Pentecost, and the ‘feasts of the martyrs’. For further discus-
sion on the selective reading of the Bible during Eucharistic services, see
Vogel 1986, 299–301.
23. The epistle reading was usually taken from the Pauline Epistles, but it could
also come from the Acts of the Apostles or passages from the Old Testament.
The gospel reading was always taken from one of the four gospels. Leclercq
1922, 246.
24. BnF nouv. acq. lat. 2171. The list of readings is reprinted in Leclercq 1922.
A complete edition of the lectionary can be found in Liber comicus.
25. Leclercq 1922, 266–7.
26. For further discussion on the selective reading of the Bible during Eucharistic
services, see Vogel 1986, 299–301; Lampe et al. 1969, 225.
27. See, for example, the list of pericopes added to the back of BnF lat. 258 from
the abbey of Fécamp (f. 156rv). For comparable lists of gospel readings, see
Leclercq 1922, ‘Évangiles’. For a full description of pericopes and gospel books
with capitulare evangeliorum, see Palazzo 1998, 91–7. He notes that these lists
are often preceded by a heading, such as ‘capitulare evangeliorum’, ‘breviarium
lectionum evangelii’, or ‘ordo evangeliorum per annum’ (93 n. 270); cf. Vogel
1986, 316–20. For a description of how pericopes were generally used in gospel
books, see Sheerin 1996, 172.
 Modes of Reading 137
28. Sometimes mealtime readings followed the Office programme (the time in
the refectory used to complete any unfinished readings from the Office), and
at other times they were chosen based on the events of the year. See the ordinal
of the abbey of Fécamp for a detailed list of refectory readings (listed as
‘Lectiones ad prandium’), including saints’ days, feast days, and other special
occasions. The Ordinal of the Abbey of the Holy Trinity Fécamp, 2. 674–85.
These readings have also been edited by Grémont 1971, 3–41.
29. Boynton 2011, 11. For a breakdown of the Divine Office with modern hours,
see Collamore 2000, 5. For a general discussion of the Divine Office in the
Middle Ages, see Crichton 1992; Harper 1991.
30. Originally, the reading list of the Divine Office comprised solely the Psalms
and prayers. By the fifth century, scriptural readings were added to the service,
followed later by the addition of hagiographical texts, patristic works, and
sermons. See Lampe et al. 1969, 233.
31. Matins was also known as ‘Vigils’ or the ‘Office of Readings’; Collamore
2000.
32. Boynton 2011, 23.
33. Boynton notes that by the end of the seventh century, the scriptural lessons
for the first nocturn ‘had been organized into an annual cycle’ that followed
the liturgical calendar (2011, 23–4).
34. Webber 2013, 21–2. For the variety of readings performed during the Office,
see The Ordinal of the Abbey of the Holy Trinity Fécamp, 1. 6–7; 2. 391.
35. See, for example, Parkes 1976, 1993, 2008; Rouse and Rouse 1979, 1982.
36. Hamesse 1999, 103. She adds that readers needed to have ‘convenient ways to
find the passages they wanted to use and know what arguments were indis-
pensable in a given domain’ (107).
37. Rouse and Rouse 1982, 209. For further examples, see the following: Parkes
1976, esp. at 135; Parkes 2008, 66–7; Rouse and Rouse 1979, 29; Rouse and
Rouse 1982, 207. Most scholars have acknowledged that many of these finding
aids were not new in the twelfth century, but can also be found in much older
manuscript traditions. Nonetheless, there is a prevailing assumption that
earlier manifestations of these features were not typical, nor were they neces-
sarily required to support modes of reading that preceded the scholastic
model. See Parkes 1976, 122; Rouse and Rouse 1982, 29.
38. For a more detailed argument, see Weston 2015.
39. For a more detailed discussion of liturgical books, see Chapter 10 of the
present volume and Webber 2014.
40. For a brief comment on lectio marks, see Webber 2015, 48–9.
41. Weston 2015.
42. Rouen, Bibl. mun. 1.
43. Private lectio happened at various locations throughout the monastery which
required readers to move their books regularly. For some examples of reading
and location, see St Benedict, Rule, ch. 48.
44. For further details, see Weston 2015.
138 jenny weston
45. Leclercq and Bonnes describe John of Fécamp as a ‘précurseur’ to develop-
ments that take place in later generations. See their comments in Leclercq and
Bonnes 1946, 89.
46. ‘Augustine allowed the rational spirit to community with the “interior of the
soul” and in this manner made self-knowledge the goal of reading’ (Exalto,
210); Stock 2001.
47. Mcginn 1994, 136.
48. Robertson 1996, 140.
49. Stock, 63.
50. Leclercq and Bonnes 1946, esp. 211–2; For Agnes of Poitou, see McNamer
2010, esp. 77–80; and McLaughlin 2010, 117–22.
51. Leclercq and Bonnes 1946, 211–7.
52. Ibid.
53. As Robertson suggests (1996, 144), ‘Lectio divina is now to be rediscovered by
a solitary individual in an unregimented situation, removed from the mon-
astic routine.’
54. For the influence of John of Fécamp on Anselm’s works, see Bhattacharji
2014, esp. 153–66; Southern 1990; and Evans 1974, 105–15.
55. Anselm, Prayers and Meditations, in S. Anselmi . . . Opera Omnia, 2. 2–91. For
Anselm’s influence, see Davies and Leftow 2004.
56. Robertson 1996, 149.
57. Studzinski 2009, 144–5; Robertson 1996, 133; Leclercq, Vandenbroucke, and
Bouyer 1968, 163–4; Salmon 1965, 122–3; Stock 2001, 122–3.
58. Robertson 1996, 145.
59. Anselm, Prayers and Meditations, 89.
60. Ibid.
61. Robertson 1996, 146.
62. Hugh of St-Victor, Didascalicon. For a comprehensive commentary on
Hugh’s text, see Illich, Vineyard.
63. Taylor 2002, 129; Studzinski 2009, 165.
64. Didascalicon, 3. 10.
65. Robertson 1996, 220. Hugh’s method of textual exposition was similar to that
of Gregory’s threefold process of historia, allegoria, and tropologia. For a brief
overview, see Robertson 1996, 217; for a more detailed study of Hugh’s
theology and practice, see Illich, Vineyard.
66. Robertson 1996, 220.
67. Leclercq 1986, 194. See also Mews 2007.
68. See n. 23.
69. Peter Abelard’s major contribution to this field was his work Sic et non; Köpf
2000, esp. 157–60; Marenbon, esp. 61–4; Grodecki 1975, 279–86.
70. For studies on the ‘rise of the individual’ in various contexts of twelfth-century
intellectual life and literature, see Bynum 1982, esp. 82–90; Ullman 1966; Dronke
1970; and Southern 1953, esp. 219–57.
71. For later developments in reading, see Illich 1993, esp. 32.
 chapter 8

Practices of Appropriation: Writing in the Margin

Mariken Teeuwen

The history of strategies of reading and writing has come into view only
over the past few decades.1 The field was given a big push by the initiative
of large public libraries to make digital facsimiles of medieval
manuscripts from their collections available online. For the first time,
modern scholars can see not only a few but large quantities of manu-
scripts. They can compare manuscripts kept in one library with those
kept in another on the screens of their computers. Digital images of
manuscripts are now available to any scholar interested in them and not
only to those for whom they were traditionally part of their research: the
philologists, who focused on the content of books, on textual variants and
stemmatological questions, and the manuscript scholars, who focused on
their appearance, studying and analysing codicological and palaeogra-
phical aspects of the book. Now they have become part of the research
material of a much greater variety of scholars, interested in historical,
social, cultural or intellectual questions of how books were made, read,
studied, used, touched, carried around, traded, treasured and discarded.
An important feature of manuscripts that has come into view only as
a corollary of their new online existence is the annotations in the margins
and interlinear spaces. Whereas both philologists and cataloguers, by
tradition, were generally inclined to ignore these, these features were
largely hidden until very recently, but digitisation and the interest in the
history of reading has put them back in the spotlight. In this chapter, my
focus drifts to the edges of pages, rather than the middle. I shall address
questions such as: What did twelfth-century makers and readers of books
do to store, sort, select and summarise their reading?2 How did they
engage with their books, in order to optimise their use of them? Was
annotating books common or special? Are certain practices of annotating
specific to certain textual genres, or shared by all genres alike? How do
they compare over the chronological length of the century, or the
geographical area of the Latin West? How did practices shift, potentially,
139
140 mariken teeuwen
with the intellectual demands and ideals of their time, and how do they
reflect these?
In the grand narrative of intellectual history, the twelfth century is
a century of change and innovation, a cultural renaissance which saw the
birth of scholasticism, the first universities, the introduction of a new
stream of texts from the Greek philosophical tradition, enriched with
Arabic interpretations. In terms of manuscript production, it is generally
argued that the face of the book itself changed so as to accommodate a new
culture of reading: the mere growth of material to be read and studied by
a scholar caused a growth in strategies of summarising, structuring and
organising texts, and the development of tools such as the index in order to
facilitate the consultation of texts.3 Next to the monastic, contemplative
practice, a scholastic model of reading developed, aimed at strategies of
reasoning, the selection and comparison of authorities and dialectical
engagement.
Parts of this grand narrative seem flawed and in need of revision. Many
of the tools twelfth-century readers and writers used to engage with the
texts assembled in their books, to aid them in their reading, study and
appropriation of the texts, were not new.4 Their appearance may have
changed, but certainly the majority of them were already in use in earlier
times. Recent scholarship has shown, moreover, that the goals of the
Carolingian reform stimulated a culture of careful, correct reading and
diligent text transmission, which involved many of the same intellectual
strategies that were so important in the dialectical, scholastic age.5
It seems timely, therefore, to try and compare annotating practices
from different times and different areas, and to take stock. A full analysis
of such practices is not yet possible, since scholarship has only recently
begun to explore the margins of medieval manuscripts from this point of
view. But a tentative comparison between some material collected from
two sets of data, one of manuscripts from the period 800–1000, and one
of manuscripts from the Long Twelfth Century, will bring a number of
interesting observations to light.6

Manuscripts of the Classics, Manuscripts for the Classroom

‘Medieval copies of classical works do not naturally constitute a distinct
category. The classics were produced, like all other books, according to the
nature of the text and the destination of the manuscript.’7 With this
statement, Birger Munk Olsen opens his chapter on the production of
manuscripts containing classical texts in the eleventh and twelfth centuries.
 Practices of Appropriation 141
He is certainly right, of course: a book with classical texts was produced by
the same scribes and used by the same readers as many other literary genres.
Yet two characteristics of the classical genre make them more interesting
than average in the particular context of writing in the margin and in
between the lines: first, they often come with a set of glosses, scholia or
even a full commentary. Secondly, they were generally produced for usage
in education. These two features make them a fruitful category of manu-
scripts for the exploration of techniques of appropriation, and a safe choice
if one wants to observe specific practices of writing in the margin.
Munk Olsen counted around 2,500 manuscripts or fragments with
classical texts produced in the eleventh and twelfth centuries, and lists
twenty-nine texts which survive in more than twenty-five copies. Of these,
about 30 per cent are dated to the eleventh century, 70 per cent to the
twelfth.8 Popular were Cicero’s moral and rhetorical works, along with the
pseudo-Ciceronian text Rhetorica ad Herennium, Lucan’s Bellum civile,
Statius’ Thebaid, the Satires of Juvenal and Persius and Terence’s
Comedies. The Golden Poets, Horace, Virgil and Ovid, were also popular.
In this first section I present six twelfth-century manuscripts containing
classical texts or which were designed for use in the classroom: a Virgil
manuscript, a book combining the prosimetric works of Boethius
(The Consolation of Philosophy) and Martianus Capella (the first two
books of his encyclopaedia of the seven liberal arts: The Marriage of
Mercury and Philology), a copy of Calcidius’ commentary on Plato’s
Timaeus, a booklet with Ciceronian works on rhetoric and two copies of
Priscian.

Leiden, Universiteitsbibl. BPL 92A: Virgil’s Bucolica, Georgica

and Aeneas
BPL 92A is a beautiful book made of very high-quality, perfectly smooth
parchment and enriched with illustrated initials in black, blue, green and
red inks.9 It is dated around 1150 and was probably produced in England.10
The layout reckons with the addition of commentary: a rather narrow
single column is used for the main text, and a column in the outer margin
which is almost as wide is pricked and ruled for the purpose of commen-
tary. Each line of the main text potentially houses two lines of commentary
text. The lines of the main text are widely spaced, so that between the lines
annotators had ample room to write their glosses and comments as well.
The pages measure ca. 240 mm × 150 mm (a rather narrow book),11 the
text-space ca. 200 mm × 62 mm, in the Aeneas part of the manuscript
142 mariken teeuwen
200 mm × 73 mm, which makes the percentage of marginal space vary
between 60 per cent and 66 per cent of the page.
The commentary added here represents a settled commentary tradition:
it has elements of Servius, Donatus and several anonymous established
commentaries on the texts at hand, together with a mix of vitae of Virgil
and accessus.12 The commentary text is split up into blocks, which surround
the main text and which are placed so as to line up, as much as possible,
with the lemmata they refer to. Where the placement could cause
confusion, minute tie marks are inserted. The set of signs or squiggles
used here are combinations of dots, circles and lines.
In general, each annotation is marked at the beginning with a paragraph
sign or hook (Γ), the top of which is elegantly curved upwards. Remarkably,
most of the interlinear glosses also begin with this sign. On some pages, the
marginal annotations are boxed or demarcated with lines, elegantly broken
with small ‘ribs’, so as to set them apart as a unit for the reader.
Apart from the stylishly entered commentary, the manuscript carries
very little evidence of use. There are no later layers of commentary added to
it, or ad hoc reader’s notes. The whole manuscript, with text and com-
mentary, must have been produced in one go, and afterwards it was kept in
pristine condition.

Leiden, Universiteitsbibl. BPL 144: Boethius’ Consolatio

and Martianus’ De nuptiis
In BPL 144, two prosimetric texts, notably Boethius’ Consolation of
Philosophy and Martianus’ The Marriage of Mercury and Philology (Books
I–II), are combined. This book was possibly produced in the Netherlands
in the last quarter of the twelfth century.13 Just as in the previous book, the
scribe reckoned with commentary that needed to be added around the
main text: he created a rather narrow main text column in the centre of the
page, and provided for two extra columns left and right which could house
annotations. The pages measure circa 225 mm × 165 mm, with a writing
space for the main text of 150 mm × 75 mm (a marginal space of
70 per cent). In the Martianus Capella part of the manuscript, this
marginal space is at times made even larger, so as to create space for even
more commentary. The main text lines are pricked and ruled; for the
marginal text columns, only the vertical lines are pricked and ruled. Each
line of main text has the same height, approximately, as two lines of
commentary.
 Practices of Appropriation 143
In the margins and interlinear space a first, contemporary set of annota-
tions has been added followed by a second one, roughly a century younger.
The basic layout of the first, contemporary layer of commentary is similar
to that of the Virgil manuscript discussed earlier: the annotations are
arranged around the main text as blocks of text, if possible placed so as
to match the line of the lemmata in the main text. As a rule of thumb, the
marginal annotations start with a paragraph sign (Γ), the interlinear glosses
do not. No tie marks are given. In the annotations, the lemma is often
underlined. In the Martianus Capella part of the manuscript, the volume
of commentary is at times so large that the marginal space is almost
completely filled. Only small strips of space at the top and bottom of the
pages are left blank.
As to the use of text and commentary, the differences between the
original layer of commentary and the later one are striking. Whereas the
scribe of the first copies a settled tradition, the second – to be dated in the
last quarter of the thirteenth century14 – consists of ad hoc additions,
personal notes and lively interactions with the text at hand.
The annotator adds nota signs, pointing out which part of the text caught
his interest, often enlivening them with pointing fingers and faces, or other
small drawings. On f. 10r he adds a nice drawing of Boethius’ wheel of
fortune (Figure 8.1). Many of his nota signs, hands and faces are right at the
edges of the pages, which may be only because of the binder’s knife that cut
the edges down in modern times, but one could also imagine that they were
purposefully put there to allow the annotator to browse through his book
and be reminded of passages he liked or which left an impression. This
personal voice is absent from the Martianus Capella part of the manuscript,
but present again in a part which must have been added to the first two at
a later stage: a copy of Macrobius’ Commentary on the Dream of Scipio,
produced in France around 1100. Only a few folia of this text survive, but
they caught the interest of our late thirteenth-century reader, who wrote
lengthy comments and illustrated points in the text with small cosmologi-
cal diagrams.

Leiden, Universiteitsbibl. BPL 64: Calcidius, Commentary

on Plato’s Timaeus
From Macrobius, it is a small step to Calcidius, a second great authority
in matters of cosmology. BPL 64 is a manuscript composed of multiple
codicological units. The first part (Plato’s Phaedo) was produced in the last
quarter of the thirteenth century, the second (Calcidius’ Commentary on
144 mariken teeuwen

Figure 8.1 Two hands entering commentary in the margins. Leiden,

Universiteitsbibliotheek, BPL 144, f. 10r.

Plato’s Timaeus) dates to the twelfth. This twelfth-century part is followed

by two more parts: a supplement of a missing part of Calcidius’ commen-
tary dated circa 1300, and a copy of works of Aristotle, a.o. De caelo et
mundo, copied in the second half of the thirteenth century. The entire
manuscript may have been produced in the southern Netherlands.15
The interesting part for the purpose of this chapter is the second part: ff.
37–124, with Calcidius’ commentary.
In this part, we see, again, a beautifully executed copy of the text, fitted
with extra-wide margins and ample interlinear space for the purpose of
commentary. The pages measure 290 mm × 200 mm, the single main text
column 225 mm × 130 mm, which leaves about half the page free for
commentary. The outer margin is the space reserved for commentary, but
the narrow inner margin, the upper and lower margins are at times also used
to add notes, explanations and excursions. It is clear that multiple hands were
at work in this manuscript, both for the copying of the text and for the layers
 Practices of Appropriation 145
of annotation. Some of them are contemporary, some later (perhaps added
only in the thirteenth century, when the other parts of the manuscript may
have been added), but they are difficult to distinguish with certainty.16
In the margin, we can observe phenomena which are by now familiar:
blocks of text, arranged next to the main text so as to line up with the
lemma they respond to; as a rule of thumb marked at the beginning with
a paragraph sign (Γ). Occasionally, tie marks are used, such as the letters
B or θ. Remarkable in this case are the many small diagrams which are
added to the set of diagrams already available in the text: they illustrate
abstract concepts such as numerical ratios, or visualise the relations
between terms and their characteristics. We can see added nota signs,
calling attention to certain passages in the text. We can also observe how
in the annotations multiple authorities are referred to: Augustine,
Tertullianus and Remigius are all mentioned in relation to the text at
hand. A new phenomenon in this particular manuscript is the use of ‘co’
(for commentarius) at the beginning of some of the annotations, which fits
with the impression that this book was tailored for the classroom.
It emphasises the fact that a teacher could be treating both text and
commentary to guide the reading process of his students.17

Leiden, Universiteitsbibl. BPL 189

BPL 189 is a very different example. It is a booklet measuring just 175 mm ×
105 mm, a composite of several parts which were written in the eleventh
and twelfth centuries, containing a peculiar, seemingly incoherent or
haphazard collection of texts, ranging from computus (Johannes
Constantiensis, Epistola de luna paschali) to canon law (Fulbert of
Chartres, Liber penitentialis), and from classical comedy (Terence) to
rhetoric (Cicero’s De inventione and Thierry of Chartres’ commentary on
Cicero).18 Especially this last part of the manuscript (ff. 42–47), which was
made in France in the second half of the twelfth century, is interesting for
the purpose at hand. Its layout is the opposite to the books described
earlier: it has small, fully filled pages, with just a few millimetres of margin:
a writing space of 167 mm × 97 mm on a page of 175 mm × 105 mm –
a marginal space of just 12 per cent. Yet a reader still felt invited to add little
notes and signs in the margin: we can see a few key words, sometimes
written at a 90 degree angle so as to fit on the page, some hooks, crosses,
quotation marks and a few faces, both smiling and non-smiling, which
illustrate the content of the text in a playful way: the frowning face stands
next to a passage about ‘empty’ eloquence without sapientia (wisdom); the
146 mariken teeuwen
smiling face stands next to a passage about how the faculty of eloquence
makes mankind superior to animals (Figure 8.2).19 While it is difficult to
put a date to the faces and signs, we can be certain that the key words and
interlinear glosses are written in a contemporary hand. Since the ink of
these markings matches that of the signs and faces, it is safe to assume that
these too were contemporary with the main text. They seem to have been
created to personalise this copy of Cicero’s rhetorical treatise to the taste of
the twelfth-century owner of these pages.

Leiden, Universiteitsbibl. BPL 91 and 92: Two Twelfth-Century

Priscian Manuscripts
Finally, two manuscripts of Priscian’s Grammar, BPL 91 and 92,20 could be
expected to reveal practices of teaching and learning in the twelfth-century
classroom. In fact, however, their annotations are not unequivocal about
their function. BPL 91 is an Italian manuscript of high quality produced in
the second half of the twelfth century, with numerous contemporary and
later annotations, but only in the first few pages. After these, the annota-
tion quickly dwindles to almost nothing. The pages measure 260 mm ×
165 mm, and the writing space measures 205 mm × 110 mm, leaving about
48 per cent of the page free for commentary. Thus we can see how the pages
of this manuscript were laid out to contain commentary, but in this case
they were prepared with less care than we saw in the other manuscripts:
a column is left free on the outside of the text, but it is not pricked with
extra vertical lines, or ruled. The designated place for commentary seems to
have been the outer margin, but the top, inner and lower margins are
occasionally used as well. The interlinear glossing is dense, but only at the
beginning. The beginnings of annotations are, as a rule of thumb, marked
with the usual paragraph sign, although some of these were added by later
hands. A later hand has also inserted paragraph signs in the text at the
beginning of each sentence. In this manuscript, furthermore, an inventive
set of tie marks is made from dots, circles, lines and squiggles, in imagi-
native combinations. The large amount of correction is remarkable, again
in different hands and belonging to different chronological layers of work-
ing on this text.
BPL 92 is a composite manuscript, consisting of four parts. Each of these
units contains a part of Priscian’s Grammar, and they must have been
brought together with the purpose of creating one more or less complete
copy. The first, second and fourth parts were all made in France in the
twelfth century;21 the third was made in the thirteenth century, probably in
 Practices of Appropriation 147

Figure 8.2 Faces on the edges. Leiden, Universiteitsbibliotheek,

BPL 189, ff. 44r and 45r.
148 mariken teeuwen
the Netherlands, in order to supplement the other parts. Here too, space is
available for commentary, especially in the outer margins, but it was not
pricked or ruled for the purpose. In this case, the writing lines are not
widely spaced either to provide for glossing. The marginal space available
for annotations in the twelfth-century parts of the manuscript varies
between 50 per cent and 62 per cent.
Many annotations are added in the margins, but most of these are not
from the twelfth century. Thirteenth- and fourteenth-century hands left
their traces in this manuscript, which suggest that it was only used for
intensive study after the different parts had been assembled in their new
composition. In the fourteenth century, the text was enriched with nota
signs, pointing hands and faces to mark passages of special interest,
indexing glosses to guide the reader through the text, and comments
which engage with the book's contents. In the twelfth century, the
marginal activity consisted mainly of corrections and the completion of
a few lacunae.
The two Priscian manuscripts can thus be characterised as reflections
of a teaching tradition which was, perhaps, no longer as fully alive as one
can observe in some Carolingian manuscripts: here a full commentary is
often added, including minute signs to mark grammar, syntax, scansion,
Greek vocabulary and other phenomena.22 BPL 91 starts with a full copy
of the settled commentary tradition from its exemplar, but quickly stops.
In BPL 92, the text of Priscian is copied, but there is relatively little room
for commentary and the marginal space was not prepared to contain
any.23

Books of History
When we turn from the manuscripts of the classics and classroom texts to
books of history, a different picture of annotating practices arises. Whereas
in the first category commentary was often available and needed to be fitted
on the page together with the main text, in the historical genre this was,
generally speaking, not the case. If we look at the margins of these
books, they are rather empty at first glance, but if we look closer, several
techniques which helped the readers to work with the texts they read come
to the fore. They showcase a different practice of working with text than
the manuscripts discussed earlier, yet we can still see how this was done
with the use of the same marginal phenomena.
Leiden, Universiteitsbibl. VLF 39 is a large book (315 mm × 230 mm)
produced in the first quarter of the eleventh century, probably in the
 Practices of Appropriation 149
monastery of Mont-St-Michel, but corrected and annotated in the twelfth
century. The book contains Gregory of Tours’ History of the Franks and
Ado of Vienne’s Chronicle.24 The text is laid out in two columns, leaving
about 55 per cent of the page blank. In all this white space, however, just
a few signs are inserted: a couple of notae, r’s for require (‘check’, or ‘look
up’) and an occasional q for questio (‘question’). Some Roman numbers are
added to give the text structure, and careful corrections are inserted over
erasures or in the margin, with tie marks. On folia 5r, 6r and 11r, further-
more, the twelfth-century annotator added ‘falsum est’ in the margin, also
carefully placed in the text with a tie mark, a remarkable sign of a critical
reading of the text, which matches the corrections and the few require- and
questio-signs found elsewhere in the manuscript.
Leiden, Universiteitsbibl. BPL 30, only slightly smaller (308 mm ×
206 mm), was produced in the middle of the twelfth century, probably
in the Benedictine monastery of St-Peter in Corbie.25 It has a part of
Eusebius’ Chronicle, and Sigebert of Gembloux’ continuation of it.
The layout varies according to the number of columns the text needed,
but generally slightly less than half the page is marginal space.
Annotations are added in several layers, the earliest contemporary, the
latest post medieval. Remarkable and probably part of the contemporary
layer of annotations are the long chi-rho combinations which are found in
the margins at several occasions: these signs, called chresima, mark pas-
sages of special interest or good use (after the Greek word chresimos,
‘useful’ or ‘usable’).26 Furthermore, crosses mark either the beginning of
passages or suspicious passages which needed to be checked. Nota signs
and key words are occasionally added to help the reader find his way in
the text, and red Roman numerals indicate the years. Because the annota-
tions consist in large part of signs and not text, it is difficult to assess the
chronological layers to which they belong. The use of these signs, how-
ever, points rather to an earlier than a later date: they were part and parcel
of annotation practices of Carolingian times, and seem to have become
less frequent after the tenth century. They thus seem a remnant of an
older exemplar, copied together with the text in which they were used to
guide the reader. Whether their use and function was still understood in
the middle of the twelfth century is hard to guess. To some of the chresima
extra nota signs have been added in a later hand, perhaps to clarify their
meaning, but this is not always the case. More examples are needed of the
use of these signs in the later period before we can fully understand their
meaning and function in this manuscript.
150 mariken teeuwen
Two Medical Manuscripts
I end my quick exploration of annotating practices with another textual
genre: two medical manuscripts, an eleventh-century one from Monte
Cassino and a twelfth-century one, probably from Spain. They are now
kept in The Hague, Koningklijke Bibl. 73 J 6 and 73 J 7. The first, 73 J 6, is
the famous Liber pantegni, the eleventh-century medical handbook of
Constantine the African, dedicated by him to Abbot Desiderius of Monte
Cassino.27 The copy now resting in The Hague may have been written
under supervision of the author himself. It contains the first ten theoretical
books of Constantine’s medical manual; the practical part is missing.
The book is long and narrow, 235 mm × 122 mm, but still laid out in two
columns each 50 mm wide. Only 36 per cent of the page is left blank.
The text itself is well articulated by red section titles. Very few annotations
are added in the margin. I counted just two pointing hands, a single
R (require?) and some lacunae made good in the bottom margin. In the
beginning (ff. 2r–4v) and at the end of the book (ff. 85v–89r), additional
medical texts and recipes have been copied in the twelfth and thirteenth
centuries. In this part, a few nota signs have been added in both contem-
porary and later hands.
The Hague, Koningklijke Bibl. 73 J 7 gives a completely different
impression. Here we have a much-used copy of Gariopontus of Salerno’s
Passionarius,28 still in its original twelfth-century binding of linen covered
with tawed skin. It is a fragile little book, measuring 200 mm × 140 mm,
with a writing area of 175 mm × 113 mm, only 30 per cent marginal space.
In the first quire of this book many marginal annotations have been added,
which summarise material from the main text, function as a marginal
index, and explain and expand the main text. As F. E. Glaze has argued
convincingly, Gariopontus’ Passionarius seems to have been used mostly as
a schoolbook; the annotations in this particular copy confirm this. They
are part of a settled commentary tradition, which was copied from one
manuscript to another. There are no signs of a personal, ad hoc engage-
ment with the text.

Practices Compared
In the earlier sections, I briefly described marginal phenomena observed in
six books containing works of classical or school-authors, two books of
history and two books with medical texts from the eleventh and twelfth
centuries. I chose on purpose examples from different textual genres, so as
 Practices of Appropriation 151
to be able to see different sets of annotating practices, with varying func-
tions. My small selection is neither well balanced nor representative: it is
a random selection of examples, certainly distorted and incomplete.
Marginal practices are only recently starting to come into view as valuable
sources for intellectual history and much of the terrain is still uncharted;
bold conclusions are, therefore, yet out of place. Yet my observations in the
small set of examples do reflect the wider patterns that I saw when going
through the collection of photos assembled for Erik Kwakkel’s research on
the transformation of the book in the twelfth century.29 A tentative com-
parison with earlier books and their practices of annotating (as inventoried
in my own data set) can thus now be presented. And because some of the
examples contained layers of annotations from the thirteenth century,
some preliminary suggestions can also be made on how practices of
annotations may have changed in the period beyond the twelfth century.
All of these observations, however, will need to be backed up by proper
systematic investigations in future research.
First, we can observe that just as in the ages before, books were in the
twelfth century more often enriched with annotations than not. This
phenomenon is not unique to the twelfth century: it is certainly valid for
the period before this age, and also for the period after.30 A book with
empty margins is an exception. Marginal spaces are almost always put to
some purpose, be it for the addition of a complete commentary or for the
notation of just a few signs, corrections or lacunae. Just as in the previous
ages and in the succeeding ages, the writing of a book was not finished with
the writing of the last word of the main text on the final page: a process of
correction, and in many cases also annotation, is part of its making. It is an
essential characteristic of the handwritten book, which survives well even
into the period of printing culture.31
Most of the activity we find in the margins of medieval books is witness
to this continuing process of correction, explanation and organising.
The exposition of a text in the shape of glosses and commentary (as we
have seen with the manuscripts of the classics, the Priscians and
Gariopontus’ medical handbook) often suggests that the text was used in
a classroom, be it within a monastic, cathedral or other kind of school.32
The enrichment of a text with added texts points at a continued reading
and supplementing of the text, both by the author himself and by more
distant readers. In only a minority of cases are we dealing with the ‘voice of
the reader’, ad hoc annotations which reflect a personal interest in or
engagement with the text. These witnesses, however, offer valuable insights
into the culture of reading and writing; they give us a unique look over the
152 mariken teeuwen
reader’s shoulder. They can take the shape of added variants, comments,
parallel texts or a-textual markings, such as signs (or drawings) flagging
interest, approval or suspicion. All of these shapes that we encountered in
my small set of examples are part of the traditional practices of annotating
text, except for the pointing fingers or hands and the faces. It is hard to date
these a-textual elements, but on the basis of my survey it seems safe to
conclude that the adding of pointing hands and faces was an innovation of
the late twelfth- or thirteenth-century reader. They serve the same function
as the earlier nota signs, which, however, remain in use in the twelfth
century and later. Some other symbols, which were used to mark interest
and approval in earlier times, such as the chresimon or asterisk, are, so it
seems, gradually eclipsed by the new practice.
It has been argued that the book changed face in the twelfth century so as
to accommodate a new ‘book fluency’, or the ability to read a text quickly
and accurately. As constituents of this new book format, reading aids such as
running titles, paragraphs, quotation marks, marginal notes, cross references
and diagrams have been mentioned.33 Each of these phenomena, however,
was already in use in the Carolingian world, and some may go back even
further.34 Running titles, for example, are a regular feature of the oldest
books that survive to us from the fourth and fifth centuries; they are, in fact,
interpreted as a consequence of the development from scroll to codex.35
We can also see how hierarchically distinguished scripts were used to visually
mark titles, subtitles, incipits and explicits from very early on.36 Quotation
marks are, again, a frequent phenomenon in Carolingian and older books:
they are used to flag passages which are quotes, sometimes with and some-
times without explicit cross-references to the authors who are quoted.37
The creation of thematically coherent compilations, for example of exege-
tical texts from different Church Fathers, is one of the more prominent
intellectual activities of Carolingian theologians. The thirteenth century is
often called the ‘age of compilation’,38 but the ninth century could carry the
same label. The annotating practices that Carolingian scholars used to
perform the activity shaped the way in which texts were read, analysed,
used and digested.39 They are the cradle for the dialectical practices of
twelfth-century logicians and theologians, who used comparison and textual
analysis to build their arguments.
Marginal notes are also not new in the twelfth century. These were
present in abundance in earlier ages, both in manuscripts of texts which
came with a commentary tradition and in an ad hoc form. In the
Carolingian copies of Virgil, Terence, Persius, Lucan, Boethius and
Martianus Capella – all texts which came with settled commentary
 Practices of Appropriation 153
traditions – text and commentary were brought together on the page to
form a coherent whole, where the voice of the author and the voice of the
expositor were both present in such a way that they could still be easily
separated by the reader.40 The layering of such commentary traditions is
a well-known phenomenon: a first layer is usually entered into a marginal
space which may be laid out for the purpose, with columns and writing
lines, but more layers may be entered in this same marginal space in
contemporary or later hands. The level of variance and flexibility is,
therefore, much higher in these marginal texts than in the set texts that
feature in the middle of the pages of our medieval books.41 The layout of
such commentary texts varies, from blocks of texts placed close to the
lemmata they refer to, to margin-filling continuous texts, in which the
lemmata are distinguished by underlining, the use of capitals or coloured
ink. This format is, if not invented, then at least used frequently in the
ninth century, and continues to be used in the tenth and eleventh
centuries. In ninth- and especially in tenth-century manuscripts, the
use of tie marks is common, employing different styles of signs: letters
from the Greek or Latin alphabets, reading signs such as dots or asterisks,
Tironian notes, musical notation symbols and newly invented
graphemes.42 In the twelfth-century manuscripts we saw a similar var-
iance, be it that the beginning of an annotation was generally marked
with a hook or paragraph sign. This particular feature I did not encounter
in earlier manuscripts.
Diagrams are definitely not new either in the twelfth century: they
appear, for example, in ninth-century copies of quadrivial or logical
texts, explaining abstract matters with visual means, probably to enable
an easier understanding or a better imprint on the memory. The fact that
diagrams travel from one text to another certainly suggests that this new
kind of visual literacy worked for at least some readers, and that they felt
invited to quote their diagrams just as they felt invited to quote
definitions.43
So, I would argue that in our comparison of the practices of annotating
books in the centuries before, roughly, 1100 and thereafter the changes are
not grand but subtle. First of all, I would point to the disappearing of
certain practices. In the Carolingian period Tironian notes are a common
phenomenon in the marginal and interlinear space of manuscripts. They
are used in the copying of commentary, where space was scarce and the
system of shorthand notation came in handy. They were also used to
annotate text with unobtrusive personal marks, to indicate which part of
a text was to be copied, remembered or studied, or which context was to be
154 mariken teeuwen
used for its interpretation.44 I am not aware of any annotations in Tironian
notes from the twelfth century: perhaps the practice of using them did not
completely disappear, but at least the frequency of using them in the
margin dwindled. A second practice which seems to have disappeared is
the use of signs to mark suspicion, disagreement or even rejection of the
main text at hand. In the earlier period a number of signs were used to flag
unease or warn the readers, including the obelus (a horizontal stroke, with
or without dots) and the theta. These signs were described by Isidore and
Cassiodorus, further developed and prominently used in the Carolingian
period, when they were an intrinsic part of the scribal toolkit.45 Their use
did not altogether disappear in the twelfth century, but they are certainly
less frequent. In the examples assembled here, an explicit ‘falsum est’ was
used instead of such a sign. On the other hand, the personal engagement
with the text by means of marking it up with pointing fingers and faces
seems to have intensified. This practice may have started in the twelfth
century and was, according to my quick survey of material, in full swing in
the thirteenth century. It may have been the result of the fact that
a growing number of books, over the course of these ages, were owned
by individuals rather than institutions and were, hence, more likely to be
annotated with personal notes. The faces, hands and drawings that we
found in later layers of annotations in some of the examples presented here
may also point in this direction. More comparative research is certainly
needed here.
Perhaps the new kind of reading of the twelfth century is not so much
attested by marginal techniques as such. The differences in approach to
text may be clearer from the contents of the books, with the introduction of
the logica nova, new kinds of dialectical collections such as summa and
distinctiones. In the margin, twelfth-century readers and writers mostly
continued using the many marginal writing practices their predecessors
already used before them, in order to correct, give structure and add
commentary. They preferred to abandon a number of them rather than
to add new ones. My sample was too small to be conclusive, but it would be
worthwhile to further explore the hypothesis that the nature of marginal
scholarship changed from active to more passive in the twelfth century, and
that its active nature became more prominent only in the thirteenth
century. The active part of twelfth-century scholarship thus may not
have found its expression so much in the development of new techniques
for the handling and appropriation of transmitted texts, but rather in the
creation of new texts.
 Practices of Appropriation 155
Notes
1. Cavallo and Chartier, Reading, 1–5.
2. For these ‘four S’s of text management’, see Blair 2010, 3.
3. Rouse and Rouse 1982; Hamesse 1999, 103–11; Kwakkel 2012, 79–80.
4. Tura 2005; Teeuwen 2011.
5. Contreni 2014.
6. The data sets are collected by the NWO-VIDI projects ‘Marginal
Scholarship: The Practice of Learning in the Early Middle Ages (ca. 800–ca.
1000)’, led by Mariken Teeuwen, and ‘Turning over a New Leaf: Manuscript
Innovation in the Twelfth Century Renaissance’, led by Erik Kwakkel.
7. Munk Olsen 1996, 1.
8. Ibid., 5, 17.
9. Munk Olsen 1982–9, 2. 726. BPL 92A is one of six twelfth-century Virgils in
the Leiden collection; the others are BPL 5, BPL 35, BPL 43, BPL 1048 and
VLQ 42.
10. Gumbert 2009, 48 (01187–8), with added comments from Erik Kwakkel.
11. Kwakkel defined a book of such measurements as a ‘holster book’ and argued
that the narrow format is a good fit for classroom use. Kwakkel 2012, 41–4.
12. Munk Olsen 1982–9, 2. 726.
13. Gumbert 2009, 70–2 (01300–01).
14. I thank Erik Kwakkel for dating the second marginal hand for me.
15. Gumbert 2009, 36 (01126–9).
16. Again, I thank Erik Kwakkel for his dating. He dated the most active later
hand to the first half of the thirteenth century.
17. Reynolds 1996, 105.
18. The size, modest appearance and peculiar selection of texts suggest that this
booklet may have been a personal vademecum of some scholar (from
Chartres?), who may have carried it around for his own studies or teaching
purposes.
19. I thank Irene O’Daly for her observations about the interpretation of the
faces.
20. Gumbert 2009, 48 (01181, 01183–6), with added comments from Erik
Kwakkel.
21. Kwakkel dates the first part to ca. 1100, the second and the fourth to 1100–50.
22. See, for one example among many, St Gallen, Stiftsbibl. 904 (www.e-codices
.unifr.ch/en/list/one/csg/0904); the marginal and interlinear annotations in
this manuscript are fully explored by Hofman, online edition at www
.stgallpriscian.ie/ (last consulted March 2016).
23. When in the twelfth century the grammatical handbooks of Donatus started
to disappear, Priscian was still copied in relatively large numbers, only to be
eclipsed by the new grammatical handbooks of Alexander de Villedieu and
Évrard de Béthune in the thirteenth century. The presence of rather un-
annotated copies of Priscian, however, may be an early indication of its
eventual loss of popularity: Holtz 2009.
156 mariken teeuwen
24. De Meyier 1973, 1. 84–6.
25. Gumbert 2009, 26 (01072).
26. Steinová 2016, 211–4, 408–9.
27. Burnett and Jacquart, Constantine, 322; Kwakkel and Newton, in press.
28. The history of transmission and use of this text is described by Glaze 2008.
29. I am well aware that my selection is not consistent or well balanced: it is
a random selection, chosen on the basis of the scanning photos assembled by
Erik Kwakkel and my own quick search focused on manuscripts which would
be easy to access for me, to wit the collections of Leiden and The Hague.
Kwakkel’s collection of photos (close to 4,700 in number) started with a full
analysis of the pictures assembled in the Manuscrits datés series and expanded
from there to include an even wider selection of manuscripts. My own data set
contains a full analysis of the marginal activity in about 350 Carolingian
manuscripts.
30. Jardine and Grafton 1990; Saenger 1999, 131–48.
31. Rouse and Rouse 1991; Blair 2010.
32. For a good analysis of annotations which could point to the intended use in
a school setting for a manuscript, see Reynolds 1996.
33. See, for example, Illich, Vineyard, 93–114. Notably, although Rouse and
Rouse 1982 are often mentioned when ideas about a direct relation between
a changing face of the book and changing literacy are discussed, they them-
selves were careful when it comes to ascribing manuscript innovations to the
twelfth century. In the work of Richard and Mary Rouse, it is generally
emphasised that phenomena used in twelfth-century books had a history
and that they were not newly invented.
34. Caillet 2009; McKitterick 2012, 23–31.
35. Lowe 1925, 1928.
36. Ganz 1995, 798–9.
37. McKitterick 2012, 25–7; Steinová 2016, 200–6; Teeuwen 2016.
38. Rouse and Rouse, Authentic Witnesses, 221–55.
39. Contreni 2014, 120.
40. Teeuwen 2015, 34–41.
41. Zetzel 2005.
42. In ninth-century manuscripts, tie marks are found in many shapes and sizes,
but in the tenth century an even greater variety of signs is used to make the
layout of text and commentary precise.
43. Eastwood 2011.
44. Teeuwen 2015, 41–3.
45. Steinová 2016, 121–51; Van Renswoude and Steinová 2017.
 part iii
Types of Books
 chapter 9

Hebrew Books
Judith Olszowy-Schlanger

The intellectual innovation of the Christian world characteristic of the

so-called Twelfth Century Renaissance (ca. 1075–1225), including the
establishment of universities, scholastic methods of learning, new aes-
thetics in visual arts and architecture, changes of techniques and modes of
book production and the appearance of the Gothic script type, had
a lasting impact on Jewish culture, books and script. For the numerically
small and relatively newly established Jewish communities in Franco-
German lands, this time was marked by growing hostility from the
Church and by accusations of ritual murder, massacres and
persecutions.1 Despite the fact that massacres accompanying the First
Crusade of 1096 accounted for the deaths of almost half of the Jewish
population in such places as Cologne, Worms or Rouen, for European
Jews the twelfth century was a time of intense intellectual activity, with
the development of new communities, their internal institutions and the
unprecedented flourishing of learning.
Jewish scholars in France, Germany and England (the region collectively
referred to in Jewish historiography as Ashkenaz, ‘Germany’) pursued the
intense exegetical and legal work of previous generations, and introduced
new dialectical and textual approaches to the study of the foundational
religious texts: the Bible and the Babylonian Talmud. The French school
of Rabbi Solomon ben Isaac (Rashi) of Troyes (1040–1105) reinvigorated
the teachings of the masters of the Rhineland schools in Worms and Mainz
(themselves inspired by the Oriental tradition of the rabbinic academies
[yeshivot] in Iraq and Israel), by developing a literalist and contextual
method of biblical interpretation (peshat). In addition, the main activity
of the twelfth-century scholars in northern France was the philological and
legalistic analysis of the Talmud. The so-called Tosafists (‘glossators’),
Rashi’s descendants and disciples, including his two famous grandsons,
Samuel ben Meir (known as Rashbam) and Jacob ben Meir (known as
Rabbenu Tam), and his most faithful pupils, Simhah of Vitry and
˙
159
160 judith olszowy-schlanger
Shemayah, engaged in writing new-style commentaries to the previous
generation’s exegesis of the Talmud, as well as monographs on jurispru-
dence, liturgy, grammar and ritual reading of the Bible. The more tradi-
tional German Judaism also witnessed important intellectual changes and
the production of original texts in the field of jurisprudence (halakhah) and
liturgy, including commentaries on the liturgical poems (piyyutim).
The German ascetic movement of Hasidei Ashkenaz, which achieved its
apogee with Judah the Pious (ca. 1150–1217), produced theological and
mystical books, such as the Sefer Hasidim.2
North-western European Jewish communities were not isolated
islands of settlement. They maintained economic and intellectual rela-
tions with Jewish communities in Italy, whose influence was essential to
northern European rabbinic learning and for the development of Hebrew
script,3 and also on the Iberian Peninsula, in southern France, and even in
the Near East. This unprecedented dynamism was also the outcome of
close interactions between the Jews and their Christian neighbours. This
is evidenced in the development of new epistemological methods in
Jewish learning and changes in Jewish book-making techniques and
script.4
Going well beyond similar clothing and casual marketplace discussions
in the shared vernacular, contacts between Jews and Christians also
involved intellectual pursuits. As far as Jewish book-making is concerned,
the old tradition assimilated new elements, including Christian influences
on book-materials, new conceptions of page layout and Gothic elements in
the Hebrew script.5 The development of Gothic features in the Hebrew
script in particular echoes the progressive passage from the Carolingian to
the Gothic type in Latin scribal tradition.6
The intense text-based and text-oriented study in both the French and
the German rabbinic schools was naturally accompanied by the manufac-
ture and use of books. Of course, the Jews had become the proverbial
‘People of the Book’ long before the twelfth century. Their religious and
intellectual life had centred around books and around the Book, the Bible,
since antiquity. In the Middle Ages, their liturgy and daily life were literally
governed by reading and writing. The medieval period had inherited the
school system, in which young boys from the age of seven learned to read
and write in Hebrew. Overall, the male part of Jewish society was highly
literate by medieval standards; indeed, its very existence depended on
literacy skills and bureaucracy. While no scriptoria or institutional libraries
comparable to those in Christian monasteries and cathedrals are attested,
 Hebrew Books 161
Jewish scholars in France and Germany possessed well-furnished private
libraries which contained books produced locally as well as abroad.7
However, while the developments of the Latin script can be studied
through hundreds of dated and localized books and documentary writings,
extant Hebrew manuscripts from this period are unfortunately very few
and far between. This is due to the tragic vicissitudes of the Jewish people,
as well as their practice of discarding worn-out writings in a genizah rather
than keeping them for posterity in institutionalized libraries and archives.
Indeed, no dated Hebrew manuscripts from north-western Europe earlier
than the twelfth century are known to us, and from the twelfth century
only eight explicitly dated books survive (from between 1177 and 1216):8
Florence, Bibl. Naz. II.I.7 (1177, Babylonian Talmud);9 BL Arundel Or. 51
(1188/9, Mahberet of Menahem ben Saruq);10 London, Sotheby & Co.,
Valmadonna ˙ Trust Libr. 1 (1189,
˙ Bible);11 Bologna, Bibl. Universitaria 2208
12
and 2209 (1193, Bible); New York, Jewish Theological Seminary 8092
(1204, Mahzor Vitry);13 BAV Vat. ebr. 468 (1215, Bible);14 BAV Vat. ebr. 482
˙ 15 and BL Arundel Or. 2 (1216, Bible).16 With two exceptions,
(1216, Bible)
BAV Vat. ebr. 468 (Figure 9.1) and 482, both written by the same scribe,
Hayyim ben Isaac, in La Rochelle, their place of production is not
˙
indicated.
The corpus can be broadened by some books that can be approximately
dated, either by the presence of extra-textual inscriptions or on palaeogra-
phical grounds, and whose north-western European origin can be ascer-
tained. For example, a siddur (daily prayer book), Oxford, Corpus Christi
Coll. 133, can be dated to the twelfth century because its flyleaves contain
accounts of a Spanish Jew who settled in England and supported himself as
a moneylender. The list of his creditors contains well-known individuals
(such as William de Chemillé, who left England in 1196 to become bishop
of Angers and died in 1200 or 1202).17 A Pentateuch, Jerusalem, Nat. Libr.
of Israel Heb. 4°5827,18 is datable because its handwriting is similar or
identical to that of Valmadonna 1. The earliest manuscript containing the
Mahzor Vitry (a liturgical and legal compendium), Paris, Victor Klagsbald
˙
Collection (olim Sassoon 535), is dated on the basis of calendars it contains
to ca. 1145–64.19 Another liturgical text, contained in a binding fragment
detached from Cambridge, Pembroke Coll. 59, must date from before
1200, because it was reused in a twelfth-century binding from Bury St
Edmunds, possibly after the expulsion of the Jews from Bury in 1190.
Several manuscripts can be dated ca. 1200 on palaeographical grounds,
such as Oxford, Corpus Christi Coll. 6 (Rashi), 165 (Rashi), Bodl. Libr.
Opp. 627, Or. 6 (psalter), Opp. 717 (bible, Hagiographa), and Laud. Or.
162 judith olszowy-schlanger

Figure 9.1 Bible, masora magna and parva, La Rochelle, 1215. Vatican City,
Bibliotheca Apostolica Vaticana, Vat. ebr. 468, f. 25r.
 Hebrew Books 163
168; Paris, Alliance Israélite Universelle 147 (Babylonian Talmud), BnF
hébr. 326 (anthology of customs of the French school); Parma, Bibl.
Palatina 2574 (Mahzor Vitry); BAV Vat. ebr. 109 (Babylonian Talmud),
ebr. 113 (Babylonian ˙ Talmud) and others, including small fragments pre-
served in the bindings and several manuscripts in Hebrew copied by
Christian scribes.20
However small, the corpus of extant dated and datable manuscripts from
the twelfth century shows a well-established tradition of book-making, with
a diversity of technical and scribal devices and an array of literary genres,
including the Bible, the Babylonian Talmud, commentaries, liturgy and
grammar. The book materials, and notably the parchment, resemble those
used in contemporary Christian manuscripts. Indeed, Jewish scribes in
Europe used to acquire parchment from local non-Jewish parchment
makers. The composition of the quires can of course vary slightly according
to the book’s quality or the position of a quire in the volume, but the
majority of the quires are quaternios, unlike quires of the Hebrew manu-
scripts from the East or Italy, which usually contain five bifolios. As for the
page and text layouts, they depended to a large extent on the type of text
copied.
The extant twelfth-century corpus includes several bibles. This funda-
mental Jewish text was the basis of the synagogue ritual as well as study.
The weekly portions (parashiyot, plural of parashah) of the Pentateuch
(Torah) accompanied by additional readings from the Prophets (haphtarot)
were read in a one-year cycle. Five books known as the Five Scrolls (hamesh
megillot) were read as a part of the ritual for specific festivals: Song of˙ Songs
for Passover, Ruth for Shavuot, Lamentations for the ninth of Av,
Ecclesiastes for Sukkot and Esther for Purim. The Pentateuch and the
book of Esther were read from books in the traditional ancient scroll
format. Unfortunately, no scrolls from twelfth-century northern Europe
are known to us, but a number of rabbinic responsa and commentaries
from the period and earlier deal with the highly codified rules of the scrolls’
production and liturgical use. From the ruling of Rabbenu Tam (Tosafot
on Babylonian Talmud, Megillah 19a), we gather, for example, that the
liturgical scrolls in his time were written on parchment, on the flesh side.
This is in contradiction to the rule of the Talmud itself (Babylonian
Talmud, Megillah 19a ditto) that the scrolls should be written on hides
treated with salt and flour and then tanned with gall nuts. The cream-white
Ashkenazi scrolls on parchment of the later centuries (similar to the
parchment used in Christian books) differ from the light-brown scrolls
written on the hair-side of leather-like hides (gevil) which continued to be
164 judith olszowy-schlanger
produced according to the traditional recipe in some Oriental
communities.21
The Torah scrolls were the property of the community as a whole and
were kept at the synagogue. This can be gathered from a responsum
concerning a legal conflict in eleventh-century Troyes, where the commu-
nity feared that a certain ostracized family might retaliate by stealing the
communal scrolls.22 Such scrolls were costly items, and some small com-
munities apparently could not afford them, as suggested by discussions on
the possibility of reading the weekly portion from a Pentateuch codex
instead.23 However, although the evidence is late, a list of Jewish books
confiscated by Alfonse of Poitiers in 1268 contains many more references to
scrolls than to Bible codices.24
Unlike scrolls, several Bible codices of the twelfth century have been
preserved. They display different book types and formats, and they contain
either the entire Hebrew Bible, its three major sections – Pentateuch
(Torah), Prophets (Nevi’im) and Hagiographa (Ketuvim) – or parts of the
Bible such as separately written psalters25 and Pentateuchs with the Five
Scrolls (see later in this chapter). Two volumes of the Bologna Bible
(Bologna, Bibl. Univ. 2209 and 2208), property of the convent of San
Salvatore until the nineteenth century, contain the Hagiographa and the
Prophets. The colophon of the scribe, Isaac ben Jacob, tells us that the book
was commissioned by Menahem ben Yehosadaq and that the project covered
˙ Today the˙ Pentateuch is lost,26 but it was
the three sections of the Bible.
described as the first volume of this tripartite bible by Bernard de
Montfaucon on his trip to Italy in 1702.27 It is uncertain whether this large
bible (405 mm × 330 mm, after trimming) was originally bound in one
volume or three, as it was found in 1702. Traces of similar holes and stains at
the end of the Hagiographa and the beginning of the Prophets indicate that
the two extant volumes at least were bound together before 1702. However,
the wrong order of the binding – Ketuvim preceding Nevi’im – shows that
such binding was done at a later stage by new owners who ignored the
correct order of the Hebrew Bible codex. The consonantal text of this bible is
provided with full vocalization, cantillation signs (te‘amim) and the critical
apparatus of both the masora parva and the masora magna, as well as the
Aramaic translation of the Pentateuch (Targum Onkelos).28
The two bibles from La Rochelle, BAV Vat. ebr. 468 and 482, copied by
Hayyim ben Isaac in 1215 and 1216, are both pandects, despite their small
˙
dimensions (218 mm × 165 and 225 mm × 170 mm). They contain vowels,
te‘amim and the masora parva and magna, as well as the Aramaic transla-
tion of the Pentateuch (in ebr. 482).29 The colophon of the richly
 Hebrew Books 165
illuminated Vat. ebr. 468 (f. 481r) calls the volume the ‘Twenty Four’
(‫)עשרים וארבע‬, referring to the twenty-four books of the complete Hebrew
Bible, rather than to its three separate units. Copying of the Bible both in
one volume and in three is well attested among Oriental Hebrew codices as
early as the tenth century.
A feature of European Hebrew manuscript production which is not a part
of the Oriental heritage is codices containing only those biblical books and
sections which are relevant for the liturgy. These books are not arranged
according to the canonical order prescribed by the Babylonian Talmud30
(with variants in medieval manuscripts), but follow the yearly cycle of
liturgical reading. These volumes, well attested in Franco-German lands in
the thirteenth and fourteenth centuries (much less frequently in Spain),
contain the Pentateuch, the Prophetical readings corresponding to the
pericopes of the Pentateuch and the Five Scrolls. The earliest example of
such a ‘liturgical Pentateuch’31 is Valmadonna 1, written in 1189, probably in
England. This calligraphic manuscript of large dimensions (382 mm ×
312 mm), unfortunately incomplete, contains vowels, te‘amim, as well as
both masora parva and masora magna and the Aramaic version. Jerusalem,
Nat. Libr. of Israel, Heb. 4°5827, the handwriting of which closely resembles
Valmadonna 1, is also a large vocalized Pentateuch (360 mm × 280 mm) with
masora parva and magna and the Aramaic Targum. BL Arundel Or. 2 also
contains the Pentateuch, Haphtarot and Five Scrolls with vowels and
te‘amim, but lacks the masoretic critical apparatus.
While it is by no means certain that these ‘liturgical Pentateuchs’ were
indeed used for liturgical reading, even as private copies read silently
during the public reading of the scroll by the cantor in the synagogue,32
they were certainly used for the study of the texts which are a part of the
liturgy. It is possible that the most calligraphic codices served as models for
the copying of liturgical scrolls.33 It is relevant that MS Valmadonna 1 and
Nat. Libr. of Israel Heb. 4°5827 lay out some specific texts, such as the Song
of the Sea in Exodus 15, in a pattern prescribed for the liturgical scrolls, and
use the forms of the letters such as curled pe (pe lefufah) or short additional
vertical strokes known as ‘crowns’ (taggin) on the top of the letters shin,
‘ayin, tet, nun, zayin, gimel and Sadeh, and sometimes also heth and qoph,
which˙ are also a feature of the scrolls
˙ ˙
rather than codices (Figure 9.2).
A special arrangement of the text of the poetic parts of the Pentateuch is
also found in Oriental and Spanish manuscripts. The layout of the narra-
tive parts of Franco-German manuscripts is also reminiscent of the
Oriental tradition. The main text is written in two columns (the afore-
mentioned Bologna, Bibl. Univ. 2008, 2009, Vatican ebr. 468 and 482) or
Figure 9.2 Pentateuch with interlinear Targum, masora magna and parva, England or Normandy, late twelfth century. Jerusalem,
National Library of Israel, Heb. 4°5827, ff. 25v–26r.
 Hebrew Books 167
three columns (the aforementioned Valmadonna 1, Nat. Libr. of Israel
Heb. 4°5827) per page. The short notes and symbols of the masora parva
are placed in the outer lateral margins of each column, while the long
quotations of the masora magna are written in long lines in the upper and
lower margins. On many pages, the masora magna in the upper margin
contains fewer lines of text than the lower margin: for example, one versus
two in Valmadonna 1 and two versus four in Nat. Libr. of Israel Heb. 4°
5827. Instead of always running straight, the lines of the masora magna
sometimes form decorative micrographic patterns, of geometrical design in
earlier manuscripts, including Oriental, Italian and Sephardi ones, and
increasingly represent humans, animals and fantastic beasts in the
thirteenth-century Franco-German tradition, despite the opposition of the
Pietists.34
In addition to the main Hebrew text, provided with vowels according to
the Tiberian system and in some cases the cantillation signs and the
masora, some of the bibles under consideration also contain the late
antique Aramaic translation (the Targum), usually the Targum Onkelos
for the Pentateuch (e.g. MS Valmadonna 1, BAV Vat. ebr. 482). No longer
a Jewish vernacular, Aramaic gained in the Middle Ages the status of
a learned tongue alongside Hebrew. Various sources from medieval
France and Germany show that the Targum was part of the educational
curriculum, that it was extensively used as heuristic means by Bible
commentators, and that it played a role in Jewish liturgy. The ancient
tradition of public liturgical reading of the Aramaic translation of the
weekly Hebrew parashah, advocated by the Babylonian Talmud,35 was
maintained in medieval western Europe only on some festivals (the
seventh day of the Passover and on Shavuot).36 But the Targum also
remained part of the educational syllabus and knowledge of it was seen
as a mark of erudition and personal devotion.37 Its role in the learning
curriculum may explain why European Jewish communities continued to
produce bilingual Hebrew-Aramaic Bibles even though Aramaic was not
an essential part of the public liturgy.
As for its layout, the Targum may alternate verse by verse with
Hebrew in the main columns of the text, as in Valmadonna 1. There
are no graphic differences between Hebrew and Aramaic: the characters
are written in the same style of script, of the same dimensions, and
they are both vocalized with Tiberian vowels. This alternating layout is
attested in some Oriental Bibles (e.g. Oslo, Martin Schøyen Coll. 206,
eleventh century38), sometimes with Arabic (often the Tafsīr of Sa‘adya
Gaon) replacing Aramaic,39 and is found in most copies of the Targum
168 judith olszowy-schlanger
in Ashkenazi manuscripts until the fifteenth century.40 In the thir-
teenth century, however, some manuscripts of the Targum display
a different layout. It is no longer intercalated with the main text but
removed to its own separate column written in parallel with the
Hebrew text. The earliest Ashkenazi example of the layout in columns
is BAV Vat. ebr. 482. Here the Targum is neatly written in columns
placed in the outer margins of each page, by the main scribe. These
columns are narrower than those of the Hebrew text, and the char-
acters are smaller. In order to correspond to the Hebrew, the lines of
the Targum are arranged in small blocks, with the last line regularly
centred.
The position of the translation in a parallel column is not attested in
early Oriental Jewish bibles, but is rather a feature of multilingual
Christian manuscripts. Going back to Origen’s Hexapla, this layout
appears notably in the twelfth-century polyglot bibles which include
Hebrew written by Christians. This is the case of the psalterium quadruplex
(Leiden, Universiteitsbibl. BPG 49a) in which the Hebrew text is juxta-
posed to Jerome’s Hebraica and the Greek Septuagint version to the Latin
Gallicana,41 or the Hebrew-Latin Psalter, Leiden, Universiteitsbibl. SCA
Hebr. 8 (Or. 4725).
The influence of Christian models can be detected also in the layout
reminiscent of the Latin manuscripts of the Glossa ordinaria, in which the
commentary is written not only in the outer lateral margins but also in the
lower and/or upper margins. The bilingual psalter Leiden,
Universiteitsbibl. SCA Hebr. 8 (Or. 4725), written probably in
Canterbury ca. 1150, is an example of such an arrangement: the central
column contains the Hebrew text, accompanied by Jerome’s Hebraica
(incomplete), while the outer margins contain an abridged version of
Pseudo-Jerome, Breviarium in Psalmos.42 This commentary sometimes
overflows into the upper margin, as on f. 3r. The Glossa layout, with the
Pentateuch in the middle and other texts in lateral and lower margins,
appears in BL Arundel Or. 2. Here, however, the marginal columns
contain the text of the Five Scrolls, written in small characters of uncalli-
graphic square script as used for masoretic notes. The Glossa layout in this
manuscript does not play its original role of commentary or translation of
the Pentateuch; rather it is a device to economize parchment by filling the
margins with texts not directly related to the main one.
The Glossa ordinaria layout inspired by Latin manuscripts would
become a standard layout for the Talmud, legal and liturgical commen-
taries in the next centuries. In the twelfth century, however, manuscripts
 Hebrew Books 169
containing halakhic (legal) text and rabbinic commentaries display
a simpler arrangement. The earliest dated Ashkenazi manuscript of
a legal text is a copy of part of the Babylonian Talmud (Florence, Bibl.
Naz. II.I.7) from 1177. It is laid out in two parallel columns per page. This
arrangement is found in other early Talmuds, such as BAV Vat. ebr. 109
and 113. It is also found in Oxford, Corpus Christi Coll. 6, written ca. 1200,
probably in England, containing Rashi’s commentary on the Prophets and
Hagiographa, and in a copy of the Mahzor Vitry, New York, Jewish
Theological Seminary 8092, dated to 1204.˙ 43 The layout of the Talmudic
and rabbinic texts in two parallel columns in Ashkenaz is reminiscent of
manuscripts produced in Italy in the eleventh and twelfth centuries.
It differs, however, from the early Oriental tradition, as represented by
the fragments from the Cairo Genizah, in which Talmudic manuscripts,
rabbinic commentaries and legal monographs tend to be written in long
lines, in one block of text per page.
The page layout in long lines is also attested in our twelfth-century
Ashkenazi corpus, where it appears in dictionaries and prayer books.
Among such manuscripts are the copy of the biblical dictionary Mahberet
of the Spanish lexicographer Menahem ben Saruq (ca. 920–70), ˙ BL
˙
Arundel Or. 51 written in 1188/9, as well as the prayer books Oxford,
Corpus Christi Coll. 133, the fragment Cambridge, Pembroke Coll. 59
and the earliest extant copy of the Mahzor Vitry, Klagsbald (olim Sassoon
˙ measuring respectively 145 mm ×
535). All three are rather small manuscripts
133 mm, 148 mm × 298 mm (size of the incomplete fragment consisting of
part of a bifolio) and 195 mm × 160 mm. To facilitate navigation through
the text, a hierarchy of script sizes has been employed; the new lexical
entries or new sections of the prayers or their commentaries are written in
larger, more calligraphic, square script.44 The Mahberet curiously employs
the forms of the letters normally reserved for the ˙ scrolls, such as the pe
lefufah, as a decorative element, which singles out the headings.

Conclusions
This brief presentation of Hebrew manuscripts of the twelfth and early
thirteenth centuries indicates a diversified tradition of book-making, in
which ancient practices inherited from Oriental models coexist with ele-
ments which evidently rely on contemporary Christian visual arts and
books. New page layouts of Latin and vernacular manuscripts, as well as
Gothic script-type, considerably influenced Hebrew Ashkenazi layouts and
scripts from the late twelfth century onwards.45 All this, of course, implies
170 judith olszowy-schlanger
Jewish knowledge of Christian books. Such a familiarity is not, however,
self-explanatory, and merits discussion.
Contacts between Christian and Jewish scholars in the twelfth century
are amply documented. In Paris, Andrew of St-Victor consulted Jews
when working on his Bible commentaries. In England, scholars such as
Herbert of Bosham46 and the author and the anonymous copyist of the
Odonis Ysagoge in theologiam (Cambridge, Trinity Coll. B. 14. 33)47 dis-
play an adequate knowledge of Hebrew, and had evidently studied the
Hebrew Bible and rabbinic biblical commentaries, especially those of
Rashi, for a better understanding and correction of the Vulgate. Contacts
between Jews and Christians were by no means a one-way traffic, deliver-
ing the ‘Hebraica veritas’ to Christians. Jews too were attracted to the
learned culture of their Christian surroundings. They were admittedly
excluded from formal education in its official language, Latin, the
domain of ecclesiastical circles (in this respect, the Jews did not differ
from the majority of the lay Christian population). They spoke the local
vernacular as their mother tongue, while learning to read and write in
Hebrew. However, recent research indicates that twelfth-century Jews
were better Latin users and readers than had been previously thought.
A good working knowledge of the Latin script and language was
a necessity for Jewish trade and monetary transactions, since these were
often accompanied by written documents, as were trials and decisions of
Christian courts. Ignorance of the prevalent legal and administrative
language would have left the Jews rather helpless in coping with the
legal intricacies of their daily life and business.48 Moreover, some Jewish
scholars actually mastered Latin and even used it in their scientific works,
as was the case of Abraham Ibn Ezra (ca. 1089–1164), of Spanish origin
but active in Normandy and England.49
That granted, the acceptance of Christian scribal models goes beyond
such intellectual contacts between Jews and Christians, or the question of
fluency in Latin. It implies more specifically that Jewish scribes had access
to actual books written in Latin script, and that they were able to consult
them and to appreciate their layout and composition. Colette Sirat has
suggested that even when Jewish individuals lacked formal knowledge of
Latin, they could still handle Latin books, even if they did not necessarily
read them. Books were, for instance, frequently pawned to guarantee
monetary loans, and many Latin volumes bear Hebrew inscriptions to this
effect.50 Bookish contacts were probably easier in circles within which
individuals engaged in work or study in common. For instance, the
production of bilingual Hebrew-Latin Bibles, which began in the twelfth
 Hebrew Books 171
century and continued through the thirteenth, witnesses to a good knowl-
edge of Hebrew script by Christian Hebraists, but also to a close collabora-
tion between Jewish and Christian scribes in the making of their books.51
In any case, while literary accounts of Jewish communities in
twelfth-century Europe tell a story of persecutions, exclusions and spatial
segregation from their Christian environment, other historical sources such
as material culture, including of course books, show on the contrary a high
degree of proximity. Hebrew manuscripts are an excellent example of shared
techniques and aesthetic models, and a mark of Jewish familiarity with
Christian books and book-making.

Notes
1. For a recent overview of Jewish settlements in northern France and Germany,
see Toch 2013, 66–74.
2. Incidentally, this work is a mine of information about Hebrew books as it
contains more than 200 rules and prohibitions designed to preserve the books
from profanation: see Sirat et al. 1996.
3. Some sources stemming from the school of Rabbi Eleazar ben Judah of
Worms (ca. 1176–1238) attributed the beginnings of Talmudic learning to
the tenth-century arrival in Mainz of a family of Italian rabbis and cantors, the
Kalonymos, from Lucca. Although legendary, these accounts echo the fact
that Italy and the Rhineland were connected by commercial routes, and
travels of scholars were attested: see Grossman 1975.
4. Sirat 2000; Isserles 2014.
5. For a description of the main features of the Ashkenazi Gothic script and its
similarities to Latin scripts, see Sirat and Dukan 1976; Engel 2014.
6. Kwakkel 2012.
7. Grossman 2001, 136.
8. For a detailed codicological description of the dated manuscripts, see
Sfardata, the database of the Israeli team of Hebrew codicology under the
responsibility of Malachi Beit-Arié, at http://sfardata.nli.org.il/sfardatanew/
home.aspx.
9. See Rosenthal 1972; Beit-Arié, Sirat and Glatzer 2006, 46–61.
10. Beit-Arié, Sirat and Glatzer 2006, 76–81.
11. Previously Sassoon collection, n° 282. See Sassoon 1932, 1.16–18; Beit-Arié
1985; Olszowy-Schlanger 2003, 238–42; Beit-Arié, Sirat and Glatzer 2006,
82–7.
12. This unit is composed of two volumes kept at the Biblioteca Universitaria di
Bologna. MS 2008 (Hagiographa) follows MS 2009 (Prophets). The initial
volume of this three-volume bible, originally kept at the convent of San
Salvatore, the Pentateuch, was lost when the manuscripts of the Bologna
religious institutions were transferred to the Bibliothèque Nationale in Paris,
172 judith olszowy-schlanger
by the order of Napoleon Bonaparte. See Weil-Guény 1991, 293–5; Beit-Arié,
Sirat and Glatzer 2006, 108; Perani and Corazzol 2013, 44–50.
13. Goldschmidt 1966, 63–75; Isserles 2012, 2. 320–7.
14. Mortara Ottolenghi 1985, 149–56; Richler et al. 2008, 406–7.
15. Richler et al. 2008, 417–8.
16. Margoliouth 1899, 41–2 n° 68.
17. Entin-Rokéah 1985. The handwriting of this siddur is not very proficient, but
˙
there are no grounds to suspect that it was copied by a Christian, as did Ta-
Shma 2004, 32.
18. The script of this codex of the Pentateuch resembles that of Valmadonna 1,
and it is possible that they were the work of the same scribe: see Glatzer 1985,
28–9 n° 22.
19. This manuscript is most probably the earliest known copy of the Mahzor
Vitry. The calendars it contains suggest a date between 1145 and 1164. Stern˙
and Isserles 2015 argue for an even earlier date of 1123.
20. Olszowy-Schlanger 2003.
21. Haran 1985, 54–5.
22. Grossman 2001, 131.
23. Ta-Shma 2004, 171–81.
24. See Nahon 1966; Lévy-Willard 2008, 44.
25. Several extant psalters written ca. 1200, probably in England, may have
been written for Christian scholars, e.g. BnF hébr. 113 or Bodl. Libr.
Or. 621.
26. The Pentateuch disappeared during Napoleon’s confiscation of the ecclesias-
tical properties in Italy, when the manuscripts were temporarily transferred to
the Bibliothèque Nationale in Paris: Perani and Corazzol 2013, 45.
27. Montfaucon 1702, 406–7.
28. Hebrew script included originally only consonants. In order to preserve
the tradition of the correct reading of the Hebrew Bible, the language of
which had not been the spoken vernacular for centuries and pronuncia-
tion threatened by oblivion, three main systems of the notation of the
vowels (niqqud) were created in the East, probably in the early Islamic
period: Babylonian, Palestinian and Tiberian (the latter still used today).
Scholars, known as masoretes (‘transmitters of tradition’) added vowels
represented by series of small graphic signs (mainly dots, strokes, stylized
letter shapes), above and/or below the lines of the consonantal text in the
Bible codices. Liturgical scrolls became an object of strict normative rules,
and continued to transmit the consonantal text only. Any additions or
corrections to this sanctified text would render the scroll unfit for ritual
use (pasul). Most medieval Bible codices contain vowels signs. In addition
to the vowels, some more elaborate bibles which were used for study or as
models for the copy of other books (known as ‘masoretic codices’) often
contained signs of cantillation or ‘accents’ (te‘amim). These te‘amim
record the intonation of the reader’s voice, the melody of the liturgical
reading. Corrections and textual annotations appear as critical apparatus,
 Hebrew Books 173
known as masora. Short corrections, reading variants and numbers of the
occurrence of specific rare words or forms are written as abbreviations in
lateral margins and are called the small masora, often referred to in Latin
as masora parva. Upper and lower margins contain longer textual annota-
tions, often containing biblical verses illustrating a specific textual issue.
These annotations are known as the great masora or masora magna.
29. Richler et al. 2008, 406–7 (BAV Vat. ebr. 468) and 417–8 (Vat. ebr. 482).
30. BT, Baba Batra 14b–15a.
31. For this term, see Stern 2012, 290–301.
32. This could be a function of small portable codices. In several thirteenth-century
small volumes the Pentateuch is copied alongside a prayer book and a number
of other texts of the traditional Jewish bookshelf, such as BnF hébr. 633,
measuring only 90 mm × 70 mm or the magnificently illuminated ‘Northern
French Miscellany’, BL Add. 11639, 165 mm × 125 mm. On the reading of the
parashah and Targum from personal books during the synagogue service, see
Isaac of Vienna’s mention of his masters R. Judah he-Hasid and R. Abraham
ben Moshe, who ‘read twice the Bible in Hebrew and once ˙ the Targum during
the public reading from the Sefer Torah by the cantor’, Sefer Or Zaru‘a, part 1,
Hilkhot Qeriat Shema, par. 11.
33. See Olszowy-Schlanger 2012, 28.
34. Sirat et al. 1996, 26.
35. BT, Berakhot 8a: ‘Rav Huna ben Judah says in the name of Rabbi Ammi:
A man should always complete his portions together with the congregation,
reading the Hebrew text twice, and Targum once.’
36. As recorded in the Mahzor Vitry, 158, n° 166, and Goldin 1995, 21.
37. ˙
Sefer ha-Roqeah by Eleazar ben Judah of Worms (ca. 1176–1238) (Hilkhot
˙
Shabbat, par. 53) stipulated that one must read (privately rather than during
the synagogue service) the weekly parashah twice in Hebrew and once in
Aramaic for the Shabbat morning prayer. Eleazer of Worms considered that
a real scholar must know the Targum because it helps him to understand the
Hebrew Bible: Urbach 1963, 4. 111, and esp. Kanarfogel 1992, 88. Isaac ben
Moshe of Vienna (1189–1250) implied that elementary teachers should
instruct their pupils in the weekly parashah together with the Targum or
Rashi’s commentary; see Sefer Or Zaru‘a, part 1, Hilkhot Qeriat Shema, par.
12. The study of the Targum together with the Hebrew Bible is also described
in the ideal model curriculum in the Sefer Huqqei ha-Torah: Kanarfogel 1992,
Appendix A: The origin and orientation˙ of Sefer Huqqei ha-Torah, ibid.,
101–15. ˙
38. www.schoyencollection.com/palaeography-collection-introduction/aramaic-
hebrew-syriac/4-6-6-hebrew-square/ms-206.
39. See Olszowy-Schlanger 2012, 34–5. For Bible codices with interwoven
Sa‘adya’s translation, see Vollandt 2009.
40. See Peretz 2008, 58; Attia 2014, 110.
41. See Olszowy-Schlanger and Stirnemann 2008.
42. Lieftinck 1955, 97–104, at 98.
174 judith olszowy-schlanger
43. For a two-column layout in thirteenth- and fourteenth-century Ashkenazi
Talmudic manuscripts, see Kogel 2014, 121–2.
44. See Isserles 2014.
45. For a description of the main features of the Ashkenazi Gothic script and its
similarities to Latin scripts, see Sirat and Dukan 1976; Engel 2014.
46. De Visscher 2014.
47. Landgraf 1934; Von Mutius 2006.
48. Mundill 1998, 28; Olszowy-Schlanger 2011, 233–50.
49. Smithuis 2006.
50. Sirat 1999.
51. Olszowy-Schlanger 2003, 58–66.
 chapter 10

Liturgical Books
Nicolas Bell

As in so many other respects, the twelfth century was a period of sub-

stantive change in the nature and appearance of books intended for use in
public worship. Several of the rites and ceremonies recorded in liturgical
books were established in Carolingian times and remained largely constant
into the later Middle Ages, even in their finer details.1 However, there were
many changes to the ways in which their texts were presented on the page,
and a general expansion in the number and variety of materials used in
liturgical celebration. Probably more parchment was expended in the
twelfth century on liturgical manuscripts than on any other type of
book. All churches were required to have several different books in order
to fulfil their liturgical obligations, and the variety in content and scope of
these volumes changed considerably over time, leading to them frequently
being amended and replaced.

New Books
In parallel with the diversity of new texts, the twelfth century saw increased
standardisation in the books central to organised worship. Changes took
effect at different rates dependent on geographical location and under
the influence of different monastic orders and diocesan contexts.
The introduction offered here is restricted for the most part to those
patterns of change which can be said to hold across Western Europe, an
approach which comes at the expense of particular details which could be
observed in more specific circumstances. Viewed through this wide lens,
the most notable development in the design and content of liturgical books
over the course of the century was the consolidation of material previously
spread across several different books into one main book for the Mass and
one for the Divine Office: the missal and breviary, respectively.2
The missal was intended to be the book for the celebrant of the Mass,
but was a relatively late invention. In the earlier Middle Ages, each officiant
175
176 nicolas bell
in the liturgy was provided with a manuscript containing only the texts
required for that role. The book for the priest officiating at the Mass was
the sacramentary, which contained only the prayers recited by the cele-
brant: the collect, preface, canon, secret and post-communion prayer, and
some prayers of blessing. Other books contained the texts required by
other participants in the liturgy: the Bible readings prescribed for each
service were provided in a lectionary (or a separate epistolary and evangeli-
ary for the subdeacon reciting the epistle and the deacon reciting the
gospel), while the Psalm verses and other sentences from Scripture which
formed the repertory of chants sung by the choir were contained in the
gradual, sometimes with a separate book for soloists known as the canta-
torium. Instructions for how the various ritual elements required in the
Mass were put together would be recorded in an ordinal, and details of the
ceremonial actions required were presented in a customary.3
From the late eleventh century onwards it gradually became normal for
the celebrant to be required to participate in all aspects of the Mass, reciting
to himself the chants as they were sung by the choir, and following the
Bible readings on the page rather than simply hearing them read by the
deacon and subdeacon. There was also an increase in the popularity of
private Masses said by the priest without choir or separate readers,
a practice which would become much more widespread in the thirteenth
century.4 For both these purposes the celebrant required more information
than was provided in the sacramentary alone, and it thus became com-
moner to merge the contents of all of the separate books of the Mass.
A missal therefore contains the central Mass prayers from the sacramentary
alongside Bible readings and chant texts (and sometimes also their melo-
dies), together with extensive rubrics to explain how everything should be
done, and by whom. The combined contents of the missal enabled the
priest to follow the words of other participants, but also to say them
himself, either privately or in a smaller setting.
This comprehensive form of missal, known to scholarship as the plenary
missal or missale plenum, thus contains almost all the liturgical information
required to reconstruct the Mass.5 The sacramentary became effectively
redundant as a separate book, as to a large extent did the ordinal and
customary. Churches were nevertheless still required to possess some of the
other books which continued in use alongside it. Bible readings could, of
course, be read from a complete Bible, but in practice it remained normal
for the prescribed readings to be copied out in lectionaries. The missal did
not often include musical notation for chants other than those sung by the
celebrant, although examples of fully notated missals are known from the
 Liturgical Books 177
late eleventh century onwards. The singers responsible for the chants sung
throughout the Mass therefore continued to use the gradual, which
remained in use as a discrete book throughout the Middle Ages. There
were major changes in the appearance of musical notation through the
twelfth century, discussed in what follows; one immediate consequence
was that the gradual can often be a physically larger book than the missal,
despite its much more limited contents.
Outside the Mass, the services of the Divine Office were catered for in
a directly comparable way by the breviary, which conflated all the required
texts that had earlier been found in separate volumes: readings from the Bible
and the Church Fathers in an office lectionary, collects and other prayers in
a collectar, chants in the antiphoner and hymns in a hymnal. Central to the
Divine Office was the recitation of the Psalms, and a psalter was therefore
often bound with the breviary (though the psalter as a stand-alone volume
marked up for liturgical use also remained common throughout the Middle
Ages, sometimes intended for lay readers). The services of the Divine Office
follow either the monastic or the secular cursus, the primary difference being
in the services of the Night Office, which were far more extensive in the
monastic cursus.
All the services of the Divine Office consist of prayers, Psalms and
readings, interspersed with chants. As with the missal, there was generally
not enough room to include musical notation in the breviary; it usually
provided the words without the music, while the chants were transmitted
separately in an antiphoner. Moreover, the readings prescribed in brevi-
aries were often much shorter than in full lectionaries. This demonstrates
either that the option was open for a much briefer rendition of the Office
than would result from the full readings or that the breviary was not
intended completely to supplant the earlier books, but merely to act as
a means of clarifying the connections between them.6
The earliest breviaries are generally small and of low grade, which raises
questions about their liturgical function. The larger volumes of later
centuries were clearly designed to rest on a lectern in the centre of the
choir, so that successive officiants in the liturgy could come forward to
recite from the book. But it cannot be presumed that this shared use was
always the case, and the relatively smaller page size of the earliest breviaries
suggests that the initial purpose of the compilation of these volumes was to
enable the book to be easily portable. Members of a monastic community
could then say the Office privately while travelling or infirm, and secular
clergy could carry the book so as to fit the Office between their other
duties. The smaller breviary continued in use alongside larger volumes in
178 nicolas bell
later centuries, sometimes reduced to pocket size and clearly intended for
personal use.7
Modern scholarship has tended to give greater prominence to the
breviary than the other books of the Office, primarily because it helpfully
provides all the liturgical information in one place. (A note of caution
should be expressed here, since the readings copied into portable breviaries
are often shorter than those in corresponding lectionaries.) The evidence of
book-lists, parish visitation returns and other inventories shows, however,
that even by the later Middle Ages the breviary was not always required by
a church, the older books retaining their function for individual compo-
nents of the service.

New Feasts
As well as being the period when the contents of diverse books were
consolidated into the missal and breviary, the twelfth century was the
time in which another structural aspect of liturgical books came to be
standardised to an extent not previously seen, in the separation of
temporal and sanctoral feasts. In previous centuries, no normal pattern
had been decided upon, but it was common in many service books
for feasts falling on fixed dates in the calendar, such as saints’ days, to
be fitted in between the moveable feasts of the Christian year, which
were dependent on the varying date of Easter. Such a scheme was
possible in the earlier Middle Ages because the practice of commem-
orating saints’ feast days through a specially composed Office and Mass
was restricted to only a small number of major saints. The majority of
these festivals, moreover, fell during the summer months, thereby
avoiding conflict with the major moveable feasts of Easter, Ascension
and Pentecost. Other feast days were concentrated around Christmas,
when the temporal feasts in the life of Christ were themselves on fixed
dates (the Nativity on 25 December, Epiphany on 6 January and
Presentation on 2 February). It was inevitable that some of the move-
able temporal feasts would become misaligned with the fixed sanctoral
feasts on occasion, but this apparently did not cause undue complica-
tion for users of the books. The calendar would effectively act as an
index to the fixed feasts, and sometimes included notes of the first and
last possible dates of Easter, which would aid navigation of the rest of
the book.8 The ordinal, meanwhile, explained at some length how to
cope with clashes of date, such as when Good Friday falls on 25 March,
the feast of the Annunciation.
 Liturgical Books 179
By the eleventh century, the proliferation of new saints’ feasts, each
with its own proper chants and lections, meant that the intercalation of
fixed and moveable feasts had become far more complex than before.
Therefore it quickly became normal to segregate feasts into two categories
of temporal and sanctoral, the former charting events in the life of Christ
and the latter commemorating saints and other festivals on fixed days of
the year. There remained a few anomalies, such as St Stephen and St John
the Evangelist, who were generally retained in the temporal section as
they formed part of the Christmas cycle of feasts on 26 and 27 December.
But by the twelfth century most books of the Mass and the Office were
divided into these two complementary cycles. It became standard for the
temporal feasts to begin with Advent Sunday, whereas earlier books had
sometimes begun the year with Christmas, and the sanctoral feasts were
arranged in the order of the calendar, either starting or ending with St
Andrew on 30 November, which falls within Advent in certain years but
not others.
As the number of feasts increased, it became gradually less possible to fit
all of the required texts into a single codex. Books were therefore some-
times split into two volumes, either with separate volumes for the temporal
and the sanctoral cycles or by separating the summer and winter, a pars
aestivalis and pars hiemalis, each volume containing separate sections of
temporal and sanctoral feasts for the relevant portion of the year. Division
of this sort first became necessary with the breviary, and became more
normal only in later centuries for other books such as the missal, anti-
phoner and gradual.9
New feasts demanded new compilations of texts, and the twelfth century
saw the flourishing of the versified Office, in which a coherent series of
chant texts was composed in verse to present the life of the saint, following
a formulaic pattern.10 Chants of the Office had sometimes been written in
verse since the ninth century, but it was only from the eleventh century that
an entire series of chants for a feast would be presented systematically in the
same verse form, and in the twelfth that the system became firmly
embedded. New Offices were integrated into antiphoners and breviaries,
but also circulated independently in self-contained libelli, or as an appen-
dix to a manuscript of a saint’s vita.11 An example is shown in Figure 10.1
(p. 187), a page from a fascicle containing the materials for the Mass and
Office of the feasts of the deposition and translation of St Cuthbert,
appended to a manuscript of his vita. Written in Durham in the
mid twelfth century, this page (measuring 230 mm × 148 mm) displays
the chants and texts for the Office and the Mass of the deposition in as
180 nicolas bell
concise a format as possible, giving only the incipits of any texts which were
more widely familiar.
Alongside these new developments, some of the older books remained in
use with largely unchanged contents, but with certain alterations in
appearance. For example, the liturgical psalter, showing the division of
the 150 Psalms for use in the Divine Office over the course of the week, had
taken a standard form since the early Middle Ages, but there came to be
more variety in the additional contents supplied with the Psalms, and the
way in which they were presented on the page.
Psalters had from an early age often included additional texts such as
canticles and the Te Deum at the end of the volume, but came increasingly
to incorporate cues, or first-line incipits, of hymns, antiphons and the short
lessons known as capitula at the end of each Psalm, sometimes as well as
a calendar at the front and a litany at the end. This expansion in contents
effectively led to the psalter becoming complementary to the ordering of
the breviary, where the full texts of lessons and antiphons were accompa-
nied by incipits of the Psalms. On some occasions the full texts of the
appropriate Psalms were combined with the full texts of particular Offices,
forming a new category of book for a particular time of day: the diurnal for
the daytime hours, the nocturnal for the Night Office, or the matutinal for
the Office of matins alone.
The verses of Psalms had by long tradition been presented each on a new
line, the division of the verse punctuated in accordance with the system per
cola et commata, and this layout generally continued through the twelfth
century. When the psalter was incorporated into a breviary, pressure on
space meant that the mise-en-page was generally altered to a continuous
text, the verse divisions marked by minor initials rather than being set on
a new line. This more concise format also came into use in free-standing
psalters in later centuries when space was a consideration.12

Reform and Renewal

At several periods in the Middle Ages, phases of liturgical renewal have
been associated with an ambition to restore an original, pure and
uniform system of worship to replace corrupted or diversified practice.
The twelfth century saw some of the most sophisticated debates over
liturgical propriety, and the most substantial changes in the presenta-
tion of liturgical manuscripts in this period derive from various pro-
grammes of reform. The early Cistercians were principally motivated
by a desire to return to the purest forms of worship, in accordance with
 Liturgical Books 181
their strict adherence to the Rule of Benedict. Any prayers or chants
not mentioned in the Rule were stripped out, and what remained was
generally shortened or simplified to conform with the principle of
austerity. Under the leadership of Bernard of Clairvaux, a further
programme of reform took place, this time not motivated by authentic
adherence to Benedictine principles so much as a desire for uniformity
of practice. Texts were normalised, chant melodies deprived of any
extravagance, and an air of restraint and consistency was imposed over
the liturgy.13
The Cistercian reforms had unexpected consequences for the appear-
ance of liturgical books outside the confines of the Cistercian Order.
In order to maintain strict uniformity of practice across the Order, new
books were copied precisely from their exemplar. The success of the
Order’s expansion led to its books having a very similar appearance and
content throughout Europe. The unprecedented uniformity was especially
notable in the appearance of the musical notation, which was always
written on a four-line stave in order to show pitches of the melody very
clearly. The use of the ruled stave had become normal in France, but in
parts of Germany it remained customary to use the older system of neumes,
which showed the articulation and direction of the melody but not the
precise pitches. The Cistercians were among the first to bring stave nota-
tion to these lands, and as the influence spread to secular houses, the system
gradually usurped local practices, leading in the thirteenth and fourteenth
centuries to the near-universal adoption of square notation on a four-line
stave.
Imposition of standardised practice was not solely a Cistercian initia-
tive. The principles of uniformity, as well as some aspects of the appear-
ance of the books, were adopted by the Carthusians and by the
Premonstratensian canons, and spread as those orders expanded. In the
thirteenth century the Dominicans likewise followed the Cistercian
model in some respects, and latterly also the Franciscans. Other more
localised reform movements, such as the influential renewal of the
Benedictine Rule centred on the monastery of Hirsau in the Black
Forest, likewise led to uniformity in this area. The collective result of
these varied attempts at standardisation was not true uniformity of
practice, of course: it rather led to a growth in liturgical diversity as
a single neighbourhood might witness the discrete rites of an increasing
number of orders of monks, canons and friars alongside a diocesan
liturgy with various parochial interpretations. But it could reasonably
be claimed that the combined effect of these diverse reforms was that the
182 nicolas bell
major liturgical books were substantially less varied across Europe by the
end of the twelfth century than they had been at the start.14
Despite these programmes of uniformity, inconsistencies remained, and
it is often difficult from a modern standpoint to be sure of how to translate
the prescriptions in liturgical books into actual practice. A common exam-
ple of such inconsistency can be seen in a close comparison of the calendar
at the start of a book of the Mass or Office with the sanctoral section.
The calendar will list the names of saints and the rank of each feast, which
should correspond exactly with the information found at the appropriate
place in the sanctoral section. It is surprisingly common for the two not to
match up. Various reasons may be posited, depending on the precise
circumstances. Throughout the Middle Ages, and especially from the
twelfth century onwards, changes were frequently made to the calendar
of saints: new feasts were promulgated, ranks of feasts were altered, local
customs either were proscribed or came to be sanctioned over a wider area.
It is very common for a liturgical calendar to be peppered with amend-
ments in later hands, representing the adoption of new saints. Any addition
to the calendar above a certain rank should be complemented with new
chant texts and readings for the corresponding portion of the sanctoral, but
inserting a substantial body of text into the tightly packed pages of the
sanctoral was a much more complicated task than adding a single name to
the calendar. Various solutions were found to this problem: a flap of
parchment could be sewn onto the relevant page, a new bifolium or
more could be sewn into the binding, an appendix of new feasts could be
added to the end of the volume, or the problem could be left unsolved,
with no new material added. The resultant mismatch between calendar
and sanctoral could be perpetuated if the book became an exemplar for
copying without careful prior thought being given to reconciling
anomalies.15
The frequency with which these liturgical inconsistencies are found has
additional consequences. Since colophons are very rare in liturgical manu-
scripts, most evidence for their date and origin is derived from the saints
found in their calendars, Offices or litanies. The inclusion of a group of
unusual saints in a calendar or litany can point to a particular locality for
which the book was originally intended. In a tightly controlled order such
as the Cistercians, new feasts were promulgated by the General Chapter
with the expectation that every daughter house would promptly amend
their service books. This ought to enable very accurate dating of
a manuscript, since there should be a cut-off date for any manuscript
after which feasts do not appear in the original phase of writing but only
 Liturgical Books 183
in later additions. In practice the situation is often much messier: if a feast
promulgated in 1180 is present, but not one from 1170, this could mean that
the earlier feast was omitted, either by accident or by design, or that the
later feast was included before gaining official sanction.
Reconciling the statutes of a monastic order with the evidence of
liturgical books throws up many such difficulties, but these are multiplied
in secular uses, where veneration of local saints was sometimes officially
encouraged, sometimes merely tolerated, and sometimes took place despite
prohibitions. In practice, therefore, all liturgical information must be
treated with great care before drawing conclusions as to the date and
place of origin of a particular book, and must take its place alongside the
palaeographical and codicological evidence on which such decisions are
normally based for other manuscripts.

Other Books
Together with the missal and breviary and the related books for the Mass
and Office which remained in use alongside them, various other liturgical
books are often found in both monastic and secular contexts.
The processional is a small, portable book, often written out by a singer
for his personal use, containing the antiphons and responsories sung in the
processions of Candlemas, Palm Sunday and the Rogation days preceding
the feast of the Ascension.16 The manual is literally a handbook, containing
the occasional services of baptism, marriage, visitation of the sick and
burial of the dead, and was therefore normally the property of a parish
priest. In general, neither of these books was particularly ornate, and any
basic illumination had the principal purpose of enabling easier navigation
of the texts. Their contents did not change significantly over the course of
the twelfth century. The pontifical contains the special services of the
bishop: confirmation, ordination, the dedication of a church and some-
times also coronation. Episcopal blessings were sometimes incorporated
into the pontifical, but sometimes were collected separately in
a benedictional. Both of these bishops’ books could be large, grand and
ornate, but as with the processional and manual, the pontifical was mainly
used while moving around, for which a portable format was preferable.17
With all of these books for occasional use, it is very difficult to make general
comments about development in format and presentation through the
twelfth century. Although their function remained essentially unchanged
throughout the century, there was a wide range in their content,
184 nicolas bell
proportions and layout, dependent entirely on the requirements of the
intended user.
Figure 10.2 (p. 188) is an English pontifical of the second quarter of the
twelfth century, with pages measuring 242 mm × 165 mm. This shows the
first page of the order for the consecration of a church, an elaborate liturgy
which occupies the next seventeen folios. The larger text size is used for all
the words which the bishop is required to recite, and for the rubrics which
instruct him in the required actions and movements. A smaller size of text
is used for the incipits of psalms following the introductory rubric and for
the antiphon with notation at the bottom of the page, since these were not
performed by the bishop himself. The presentation of the music warrants
separate discussion in what follows.

Music
The principal books of chant, the gradual for the Mass and the antiphoner
for the Office, remained broadly similar in content throughout this period:
the core repertory of chants was expanded by the addition of new saints’
Offices, but otherwise remained very stable, with a largely fixed body of
chants consisting of musical settings of biblical verses appropriate to the
feast day. Bigger changes were seen in the ancillary repertories of tropes and
sequences, and new compositional approaches to these genres led to new
experiments in page layout.
Tropes are interpolations to an existing chant. It was the practice in many
monasteries, and occasionally in secular churches, for the chant sung by the
choir at Masses on major feasts to be expanded upon by a solo singer who
effectively provided a commentary on the main chant. While the chant was
a Bible verse sung to an ancient melody, the words and music of the trope
elements interrupting it were in a new style, often composed locally and
thereby adding solemnity to the chant in a way that was unique to that
institution. The practice of troping flourished throughout Europe in the
tenth and eleventh centuries. By the twelfth the tradition had begun to die
out in some quarters, with other practices such as new processional liturgies
providing the means to add solemnity to feast days. But in others the
practice of troping continued, with new words and melodies composed in
new styles. In many places the new tropes related closely to theological
developments, inserting texts into Mass chants such as the Gloria and
Sanctus which reinforced the debates presented in sermons of the time,
often with florid new melodies.18
 Liturgical Books 185
The sequence originated at the same time as the earliest tropes, initially as
a means of attaching words to the florid textless melodies known as
sequentiae which followed on from the singing of the alleluia in the
Mass. The form and style of the sequence changed radically in the late
eleventh and twelfth centuries: where originally the verses had used a free
metre in order to fit to pre-existing melodies, new styles came to adopt
regular rhythmic metres, and words and music were composed hand in
hand, often following strophic patterns. The sequence became the
preeminent genre of liturgical poetry, and new repertories of sequences
quickly developed across Europe.19
Since both tropes and sequences were local repertories of new composi-
tions, there was a greater imperative for them to be supplied with musical
notation than in the case of older repertories of plainchant, which were
already well known to all singers. Tropes and sequences often circulated
together in a troper, a small book which effectively collected together all of
the new music to be sung alongside the established chants contained in the
gradual or cantatorium. The earliest tropers often took the elongated
format which had been used for the cantatorium, a shape sometimes
designed to accommodate ivory covers. By the twelfth century this format
came to be superseded by more conventional proportions.20 The earliest
tropers, from the tenth century, are among the first books to be provided
with musical notation throughout: earlier graduals generally supply the
words of chants without any notation. By the end of the eleventh century,
all graduals were fully notated too. In both cases the notation in this early
period consisted of neumes which displayed the shape and direction of the
melody without specifying the pitch.
By the end of the eleventh century the need had grown to find a way for
notation to display pitch accurately. In the region of Aquitaine, simple
notes were plotted graphically on the dry-point lines ruled for text, but this
scheme had the double disadvantage of lacking the nuances which neumes
conveyed and taking up a great deal of space. From the start of the twelfth
century experiments were made in using one or two ruled horizontal lines
to enable the relative pitch to be shown more precisely, and by the end of
the century it had become standard through much of Europe to rule four
or sometimes five lines in red ink, with each pitch allotted to a line or the
space between them. In other words, the system which underpins Western
music notation to this day was formed in the twelfth century.
The earliest inscriptions of neumes on a four-line stave tended to retain
the basic shape of the earlier staffless neumes, stretching them where
necessary to cover the intended range of pitches. In parallel with the
186 nicolas bell
development of Pregothic script into the more angular forms of Gothic
Textualis in the thirteenth century, the shapes of note forms became more
angular and stylised through the course of the twelfth century, eventually
assuming the forms of square notation which became standard in later
periods. However, it was not until the thirteenth century that a separate,
wider pen nib came to be used as a norm for square notation, which
emphasised the contrast between wide horizontal strokes and hairline
verticals.
The adoption of the four-line stave was a gradual process, and took
effect at different rates in different places. Recent surveys are enabling the
construction of a more systematic account than has been possible
hitherto.21 In some regions, especially in France, it was common for four
lines to be ruled with a dry point, the lines for the notes F and
C subsequently ruled in red and yellow. Elsewhere lines were ruled free-
hand in red ink, without a dry-point ruling to guide the pen. Sometimes
a four- or five-line stave was employed systematically throughout a volume,
but on other occasions the number of lines varied depending on the range
of notes to be accommodated. Except when lines were coloured to show
pitch, a clef was required at the start of the line to show which line
corresponded to which pitch. In the earlier period the position and choice
of letter used for the clef varied considerably, but by the thirteenth century
it became most common for the clef to show the note C or F.22
Staves were generally ruled at a later stage than the pricking and ruling of
the parchment, and there are perhaps as many cases of the staves being
ruled after the text had been written as vice versa. It is inevitably very rare
for the notation to have been written before the stave-lines were drawn
(except in cases of red and yellow lines being applied to existing dry-point
lines), but it is common for the words to be added after the notation, so
that syllables can be placed precisely under the relevant notes. It seems not
to have been until the fourteenth century that the rastrum or four-nibbed
pen came into use for ruling several parallel lines simultaneously, and
individual line-ruling remained the norm for some time thereafter.
Figures 10.1 and 10.2 show two early examples of notation on a stave.
In the pontifical in Figure 10.2, probably from the second quarter of the
twelfth century, the shape of the notes is very close to that found before the
advent of the stave. A ‘natural’ sign is used as a clef to show that the middle
space is a B. In Figure 10.1, from the middle of the century, the notes are
a little squarer, and therefore slightly less close to the shape of the neumes
from which they derive. Here the stave has three or four lines depending on
the number of notes required, and D, A, C and F are all used as clefs.
 Liturgical Books 187

Figure 10.1 Office and Mass of St Cuthbert, Durham, mid-twelfth century.

Cambridge, Trinity College, O. 3. 55, f. 54r.
188 nicolas bell

Figure 10.2 Pontifical, perhaps from Ely, ca. 1125–50. Cambridge, Trinity College,
B. 11. 10, f. 50r.
 Liturgical Books 189
Both of these figures demonstrate the problems which affected any scribe in
planning the layout of a page with a musical stave. Where previously neumatic
notation could be added selectively between normally spaced lines of text, in
the manner of an interlinear gloss, ruling a stave required at least as much
space as a text line. The result was that writing music on staves could use twice
as much space as writing music without them, and the scribe needed to know
exactly where the music would occur before planning the shape of the page.
Once the stave had settled into regular use, in the second half of the twelfth
century, standard formats came into being. In a gradual, where most of the
texts are supplied with music, the default pattern was to rule the whole page
with stave-lines, leaving gaps only at those points where rubrics or other texts
without chants were to be written. In the missal, notation was generally
restricted to very small portions of the canon of the Mass, which meant that
stave-lines could only be added at a late stage, once the precise length of other
texts had been calculated. In a hymnal, where only the first verse of a hymn
needed to be notated, the subsequent verses following the same melody, the
stave would be ruled for the length of a single stanza, leaving space for the
remaining verses of text to be written afterwards. Sequences were generally
composed in paired strophes with each line of music repeated, and this was
often displayed on the page by writing the melody once only, with the two
strophes of text spaced below it to align with the notes. All of these various
formats were planned in the interests of economy, but also of elegance.
Although minor errors of judgement and planning are easy to find, it is
surprisingly rare for a book to have empty stave-lines ruled in vain, or for
notation to be forced onto extra lines ruled outside the main textblock.
Concern to save space may also have been a primary motivation for
adopting a two-column layout in many liturgical books. It is very difficult
to make any generalised comments about presentation in two columns in
twelfth-century liturgical books, beyond the general observation that most
liturgical books of the eleventh century are in single-column layout,
and most of the thirteenth century in double columns.23 A two-column
mise-en-page can save space and can aid legibility, but neither of these is
universally the case. The size of the page and the size of the text are
inevitably the major determining factors, as well as the size of the text
relative to the page. The presence of music on a stave complicates these
decisions, since some types of chant benefit more than others from being
laid out in regular lines across the page, rather than split into narrow
columns. The decision of whether or not to adopt a columnar layout was
ultimately one of a number of interrelated questions which the compiler of
the manuscript had to pose at an early stage of planning.
190 nicolas bell
Conclusions
This brief overview of developments in the design and content of liturgical
books has not sought to cover the many other ways in which liturgy was
understood in the twelfth century. The tradition of liturgical commentary
reached new layers of sophistication during this period, in keeping with
new scholastic methods. The ninth-century treatise of Amalarius of Metz
remained the fundamental study of liturgy, but his work received new
layers of exegesis, most notably in the Gemma animae of Honorius of
Autun (d. 1154) and the Summa de ecclesiasticis officiis of Jean Beleth, which
was composed in the 1160s and became widely read in the thirteenth
century. Sermons are intrinsic to church services, but have generally been
treated separately from liturgy in later scholarship, to the detriment
of both.24 Liturgy is also closely reflected in literary and other texts
throughout this period.25 Recent work has started to reintegrate liturgical
studies into the wider context of medieval life and thought, and liturgical
books have a central place in this discussion.

Notes
1. McKitterick 1993, for the context of the Carolingian reforms.
2. Harper 1991.
3. For these categories of book, see Palazzo 1998.
4. Jungmann 1951, 1. 212–33; Vogel 1986, 105–6.
5. The term missale had on occasion been used in earlier centuries to denote
what is here termed the sacramentary, hence the need in modern times to
distinguish the composite book as missale plenum: Palazzo 1998, 27–34.
6. Leroquais 1934; Salmon 1967; Tolhurst 1932–42, esp. vol. 6.
7. Van Dijk and Walker 1960, 528–42.
8. In general, see Harper 1991, 45–57; Grotefend 1891–8 on liturgical calendars.
9. Salmon 1967, 44–85.
10. Jacobsson and Haug 2001; Jonsson 1968.
11. Huglo 1988, 64–75, on libelli.
12. Leroquais 1940–1; Parkes, Pause and Effect, 103–5, on the layout of Psalm
verses.
13. Marosszéki 1952; Waddell 2007.
14. King 1955; Heinzer 2008, 85–405, on Hirsau.
15. Chadd 1986. See also the ‘Excursus: On Ascription of Liturgical Books to
Specific Churches’ in Pfaff 2009, 192–9.
16. Huglo 1999–2004.
17. For catalogues of pontificals and benedictionals, see Kay 2007; Rasmussen
1998.
18. Iversen 2010.
 Liturgical Books 191
19. Fassler 1993.
20. Huglo 2001. A catalogue of tropers is provided in Husmann 1964.
21. Recent projects to survey manuscripts with musical notation include Hartzell
2006; Colette et al., covering manuscripts up to 1200 with French neumes in
the BnF; Meyer et al. 2006 (eight volumes to date); and Klugseder et al. 2014.
22. Haines 2008 and several of the contributions to Haines 2011.
23. For the use of double-column formats more generally, see Chapter 1 of the
present volume. The relative sizes of selected antiphoners are tabulated in
Huglo 1988, 95; for graduals, see Huglo 1957.
24. Reconciliation of this division is evident in Morand 2008, and other con-
tributions to the same volume.
25. See especially Zieman 2008.
 chapter 11

Books of Theology and Bible Study

Lesley Smith

Even in a volume dedicated to the creativity of the books of the long

twelfth century, those made for the study of the Bible and the burgeoning
discipline of theology are notably rich in innovation. In part, this is
a response to the centrality of the Bible in so much of medieval life; but
it is also the corollary of a change in setting for Bible study in this period,
which went from being the preserve of monasteries to a subject taught in
clerical schools, mostly attached to cathedrals. Paris in particular boasted
a variety of establishments: the important school at the cathedral of Notre
Dame was joined by those at the abbeys of St-Victor and Ste-Geneviève,
and a number of others run by individual masters. By the end of the
century, these had consolidated to become a proto-university, and con-
stituted the most important centre for biblical research in Europe.
Moreover, the schools underpinned the establishment of a secular com-
mercial book trade in the city. In a symbiotic relationship, the core
business from masters and students allowed the book trade to flourish
in other – perhaps more frivolous – subject-areas. The trade was also
predicated on, and reflected, a shift from predominantly oral teaching to
types of instruction that expected students to have access to books.
We see this movement mirrored in the form of some of the books
discussed in this chapter.1
Because of its central place both in the study of the Bible and theology and
in the commercial book trade, almost all the types of book I discuss here were
products of Paris. Even when not literally made in the city, their spiritual
home was Paris; copying texts originating or circulating in the schools, they
mimic (or try to) Parisian innovations in design. Paris books found their way
into the hands of scholars and libraries across Europe. Other schools in the
twelfth century continued to offer biblical and theological courses, but as yet
we know comparatively little about the particular books and materials they
produced. And not all monastic writers were eclipsed by the movement to
the schools. Bernard of Clairvaux, and in particular his Sermons on the Song
192
 Books of Theology and Bible Study 193
of Songs, was a twelfth-century bestseller known by schoolmen as well as
monks. However, since the form in which these texts were produced was
relatively conservative, I have chosen not to deal with them here. Instead,
I want to focus on innovation.
We begin by looking at the most authoritative medieval book: the Bible.
Study of the Bible was responsible for probably the most creative phase of
twelfth-century book production, both in the texts produced to facilitate
such study and in their layout and design. A typical twelfth-century monastic
volume – whether a patristic text or part of the Bible, such as the gospels or
the Pauline Epistles – was a plain book, with about a half-folio page
size. While there are spacious margins, the text, written with little
abbreviation, takes up the whole written space, often in a single column.
The page has little differentiation to aid the reader: generally, for instance,
there are no running headers with the names of the biblical books; chapter
division was still not standardised, though some bibles have numbered
capitula lists at the beginning of each book, linked to numbers in the margins
by the text; decoration to mark out sections is often minimal. A book is to be
read from beginning to end, or else finding one’s way requires prior
knowledge of the text. The usual form of a biblical commentary in this
period was lemmatised: a short section of biblical text (= lemma; plural
lemmata) is followed by the commentary. The lemma itself is marked out
either by underlining (sometimes in red; occasionally the whole lemma is
written in red) or by quotation marks in the margin. In this form of
commentary, the whole biblical text is not present, only the parts that
have been chosen for exposition, and the level or register of the individual
comments is not distinguished, so that short etymological notes, for exam-
ple, sit alongside more complex spiritual interpretation. One point simply
follows after another.
However, from the beginning of the twelfth century, another form of
biblical commentary began to appear – the Glossed Bible (Glossa, later
Glossa ordinaria) – and by the second half of the century, it was the text of
choice for any scholar.2 Students and libraries no longer wanted plain Bible
texts; they looked instead for the Gloss. The Gloss story seems to have
begun at the cathedral school at Laon, run by the renowned master
Anselm, aided by his brother Ralph and Gilbert ‘the Universal’ of
Auxerre. Anselm was known as a teacher of the liberal arts, but some
time around 1100 he seems also to have offered lectures on the Psalms
and the Pauline Epistles, taking the teaching of the Bible from the mon-
astery to the ‘secular’ (that is, not monastic, though still clerical) school.
Nevertheless, in starting with these two books, Anselm was following the
194 lesley smith
monastic pattern: the Psalms were the backbone of the daily liturgy, and
the Epistles set out Paul’s formulation of Jesus’ teaching, constituting the
framework of Christian theology. In addition to these two texts, it seems
likely that the team at Laon also prepared a representative selection of Old
and New Testament books.
The most immediately apparent novelty of the gloss was the layout
of its pages (Figure 11.1), a format probably borrowed from glossed
liberal arts texts, which Anselm also taught. Each page was divided into
three columns of largely unvarying widths, with the inner column
narrower than the other two. The central column presented the
whole text of the biblical book, with the outer and inner columns
reserved for commentary. Only the central column was fully ruled.
The other columns were ruled only as and when needed for the
comments or ‘glosses’ on the text; and they were ruled more densely
than the central column, usually with two or three lines of gloss for
every line of biblical text. The scriptural text was written in a script two
or three times larger than the glossing hand, on alternate lines, with
room for short glosses in between the lines. The start of each gloss
was generally marked by a paraph sign, but not with a biblical lemma,
so that often only positioning gives a clue as to which gloss refers to
which part of the scriptural text. Few glossed books were highly
decorated, although they may have coloured initials to begin chapters
or sections of Bible text.
The glosses themselves were drawn from the Church Fathers and from
‘modern’ masters such as Bede, Rabanus Maurus and Isidore. Each biblical
book uses a different selection of authorities. At one extreme, Job relies
entirely on Gregory the Great’s Moralia in Iob as the basis of its glosses.
Genesis, on the other hand, employs a range of resources from Augustine,
Jerome, Bede, Isidore, Rabanus and others. Some glosses are paraphrases,
some verbatim extracts. All are somewhat contracted and rely on the reader’s
existing knowledge to make sense of what is on the page. At this stage of its
history, the gloss seems to have been a tool for the lecturer to use in the
classroom. Sitting with a glossed book on a reading stand or on his knees, he
could read the biblical text aloud, using the individual glosses as the founda-
tion of his sequential commentary. It would not be enough for him simply to
read out the glosses, which often need to be put in context and/or amplified
in order for them to be useful to students. The gloss in its beginnings was an
aide-mémoire for an expert user.
Like almost all cathedral schools, the school at Laon was reliant on the
drawing power of its masters to attract pupils; indeed, it was rare for the
 Books of Theology and Bible Study 195
fame of a school to outlast one particularly respected teacher.3 After the
death of Anselm, Laon declined in importance, and it would not have been
surprising if the glossed book format had also fallen away. However, by
some agency not yet understood, around 1140 the idea of a Glossed Bible
was taken up by scholars at Paris (perhaps specifically by Peter Lombard,
master of the cathedral school at Notre Dame); the remainder of the
biblical books (minus a few stragglers) were glossed and the existing
glosses were revised and expanded. By the middle of the century, a set of
glossed books typically ran to twenty or twenty-one volumes, with some
books, such as the Wisdom literature, commonly copied as a group, or
gospels bound together in pairs.4 As the twelfth century progressed,
scholars began to comment not simply on the Bible, but on the Glossed
Bible, with ‘Glossa’ quoted as an authority alongside Scripture and the
Fathers.5
Around the same time as this development in the status of the gloss came
changes in the layout of the page. The density of glossing for any one
biblical book could vary enormously, depending on the interest and
importance of particular parts of the text, and it is generally true that the
beginning of a book attracted much more comment than the end. In these
terms, the ‘simple’-format gloss, described earlier, with its invariable col-
umn widths, was inefficient and wasteful of space: in some parts,
glosses had to be crammed into the margins above and beyond the outer
columns; in others, large areas of the page remained untouched. Scribes
began to vary the width and height of the central (scriptural) column from
page to page, depending on how much gloss they needed to accommodate,
and they ran the glosses into more than one parallel column when the
volume of commentary demanded it. These experiments gradually reduced
the amount of unused space on each page until, from around 1170, the
layout reached its apogee, and the whole available written area was fully
utilised (Figure 11.2).
It is hard to overpraise the achievement of page design that these
complex-format glossed books represent. No two pages are alike.
The layout of a particular gloss was not worked out and then slavishly
repeated from copy to copy: the variation in the size of script between
any two scribes means that each glossed manuscript worked out the
layout anew. It is no wonder that Abbot Simon of St Albans is written
of as employing a special scribe (scriptor specialis) to do this work: the
spatial awareness and control it requires are breathtaking. Unlike the
simple-format copies, the whole page was ruled, at the height of the
glossing script, but once that had been done, the scribe could use any
196 lesley smith

Figure 11.1 Layout of the Glossa ordinaria. Oxford, Bodleian Library, E. D. Clarke
35, ff. 53v–54r.

part of the page at will, for either Scripture or glosses. The biblical text
was recognisable by a hierarchy of script, since it was twice as big as the
glossing hand and written on alternate lines; the interlinear glossing
was retained.
 Books of Theology and Bible Study 197

Figure 11.1 (cont.)

In addition to this revision of the layout, the gloss text was also revised
about 1170, and in particular the prefatory material to each book was added
to and reordered; but we do not yet know enough about the various
redactions of the gloss to be clear how this was done, and whether there
was an overarching agenda for all the biblical books.6 What seems clear is
198 lesley smith

Figure 11.2 Layout of the Glossa ordinaria. Oxford, Bodleian Library, Auct. E. inf. 7,
ff. 118v–119r.
Books of Theology and Bible Study 199

Figure 11.2 (cont.)

200 lesley smith
that the closely woven pages of the complex-format gloss could not have
been used for teaching in the same way as the simple-format copies. It was
simply too difficult to find one’s way through so much densely packed
material for it to be viable for the immediacy of the classroom. Instead, the
complex layout must represent a stage where the gloss was used for
consultation and reference, probably by students as much as by their
teachers. It reflects the almost entirely oral teaching of the early twelfth
century giving way to a more text-based model.
The Glossed Bible became the central text of twelfth-century Bible
study. Part of the reason we know this is the sheer number of copies
which survive – thousands of extant manuscripts, even now, and many
more lost copies visible in medieval library catalogues. The text that
started as a classroom tool in Laon spread across Europe, and although
the majority of copies date from the twelfth and thirteenth centuries,
some were still being made in the fifteenth, even after the first printed
edition produced by Adolph Rusch of Strassburg in 1480/1. Moreover,
although each page of each copy is individual, the text they preserve is
largely identical, allowing for the differences in the one or two revisions
of each biblical book. This is an extraordinary feat of the mass production
of texts – an ability which is an important part of the twelfth-century
book story.
Alongside the gloss, the first half of the century saw the production of
two further commentaries on the Psalms and the Pauline Epistles which
experimented with layout and design, and both enjoyed a measure of
success – although one much more than the other. The first was the
work of Gilbert de la Porrée (Gilbert of Poitiers).7 Gilbert was one of
Anselm’s pupils, who taught at Chartres and Paris before becoming bishop
of Poitiers in 1142. We are still unsure when and where Gilbert wrote his
commentary, and of its relation to the Laon Psalms and Epistles glosses.
However, what does seem to be clear is Gilbert’s own involvement in
devising the layout of his text, since this same format (‘cum textu’) is found
in his other, non-biblical commentaries such as his commentary on
Boethius’ Opuscula sacra. Each page is divided into two unequal columns,
with the commentary in the broader outer column and the biblical text in
the narrower inner. The biblical lemmata in the commentary are distin-
guished, generally by underlining. Unlike the gloss layout, in Gilbert’s
design the commentary text is written first and the biblical text added
afterwards. This is because, more often than not, the amount of commen-
tary outweighs the scriptural text, so the scriptural column has to pause so
that the commentary can catch up. The scribe achieves this by adding
 Books of Theology and Bible Study 201
decorative line fillers to the scriptural column and spacing out the writing.
Less often, it is the commentary that gets a little ahead, and the biblical text
has to be scrunched together to fit the page. Judged by layout alone,
Gilbert’s commentary would be usable by a teacher in the classroom, in
the same way as the simple Gloss texts; but since the author of a study of his
Psalms commentary notes that his method can appear to be chaotic, this
might not have been an option for anyone other than Gilbert himself, and
it is more likely that the commentaries were used for reference. Certainly
Gilbert’s were much less successful than the second set of Psalms and
Epistles commentaries with their own experimental layout – those of
Peter Lombard.
Peter was head of the cathedral school at Notre Dame in Paris – a very
influential teacher who was made bishop of the city at the end of his life.8
Peter certainly taught using Glossed books, though his own Psalms and
Pauline Epistles commentaries employ another layout, now known as
intercisum or intercut. In this, a block of biblical text, taking up about
half the width of the column of script, is surrounded and followed by the
text of Peter’s commentary. The biblical script is twice the size of the
commentary hand, and written on alternate lines; in some more expensive
copies, it is underlined, or even written entirely, in red. As with the Gloss
and Gilbert, the whole scriptural text is present, although divided into
small portions. The intercisum layout is easier to produce than Gilbert’s, as
each section of commentary simply follows its portion of Scripture, so
there is no need to judge how much space should be left, and no stretching
or contracting of the writing.9 It was also directly usable in the classroom
by a teacher. This layout was so associated with Peter Lombard that it
appears in incunable copies of his works, such as the edition of his Psalms
commentary printed in Nuremberg in 1475 by Johannes Sensenschmidt
and Andreas Frisner.
The format remains workable even when a copy was very smart and very
academically complicated, such as the manuscript of Peter’s Psalms com-
mentary made for Thomas Becket (Figure 11.3).10 Here, the commentary
runs alongside two of Jerome’s translations of the Psalms, the Gallican and
the Hebrew. The two translations are written in slightly different sizes of
script to reflect their relative importance (the Gallican the bigger, in the
middle of the columns; the Hebrew the smaller, to the left), but with the
whole retaining the intercisum layout. The three sets of commentaries on
the Psalms and Pauline Epistles – the Gloss, Gilbert and Peter Lombard –
came to be known as the parva, media and magna glosaturae, and their three
separate formats meant that a twelfth-century scholar could open a book
202 lesley smith

Figure 11.3 Peter Lombard’s Psalms commentary made for Thomas Becket.
Cambridge, Trinity College, B. 5. 4, f. 135v.
 Books of Theology and Bible Study 203
and know immediately, simply from the look of the page, which of the
three he was dealing with.11
For students of the Bible, one further work was even more useful than
these commentaries, especially for beginners. The Historia scholastica of
Peter Comestor was a kind of abbreviated Bible paraphrase.12 Much easier
to read than the Bible itself (and shorter), it was extremely popular. Peter’s
text also demonstrates experiments in layout, incorporating offset text
boxes, split columns, explanatory diagrams and lists of distinctiones, often
by way of marginal tree diagrams, to illustrate different readings of
a particular word according to the four ‘senses’ or methods of scriptural
interpretation. These techniques are also visible in manuscripts of Peter’s
own biblical exegesis. Moreover, these manuscripts take us directly to the
classroom, since they may survive only as reportationes – approved and
corrected transcriptions of Peter’s oral teaching. The immediacy of this
world is preserved when the student reporter notes Peter’s own words,
beginning: ‘The master himself says . . . ’ (‘Dicit etiam magister . . . ’).
The Bible, then, was the bedrock of teaching about God and the
Christian faith; but it was not the only way that the elements of belief
were discussed. Anselm of Laon lectured from a glossed Bible, but he also
taught by considering individual issues – concepts and problems arising
from the biblical text, which took on an identity independent of
Scripture: a theological identity. Consideration of questions such as the
nature of God’s creation, or of Christ’s Incarnation, was not new. This
sort of systematisation of Christianity was what the Apostle Paul was
striving to achieve, and it had been continued by patristic writers, but the
twelfth-century schools developed the organised teaching of these sorts of
issues. Anselm proceeded by means of the ‘sentence’ (sententia) – a short
statement of a single theological problem, supported by varying opinions
on the topic drawn from patristic authorities. Whereas Bible teaching was
done by lecture, theological questions seem to have been discussed in
something of a seminar setting, using sentences as a starting point.
Many collections of sententiae still exist, some with only a few scattered
questions, others much bigger and better organised; some sentences are
common across a number of collections, others exist in only one. Modern
scholars have described these groupings disparagingly as haphazard and
jumbled, but what is more interesting is the picture they conjure up of
teachers and students picking and choosing among a larger pool of
materials and sources to suit their own purposes. The best-known collec-
tion of sentences is associated with Anselm and the Laon school, the Liber
pancrisis – ‘All Gold’.13
204 lesley smith
The sentences method was taken up by Peter Lombard, who taught with
it at Notre Dame. He reordered and organised the issues for consideration,
producing his own Four Books of Sentences.14 Book 1 covered the Trinity;
Book 2 the Creation and Sin; Book 3 the Incarnation and Virtues; Book 4
the Sacraments and the Eschaton. Peter’s compilation was wildly popular:
just as Peter Comestor was known as the Master of the Histories, Peter
Lombard became the Master of the Sentences. At the beginning of the
thirteenth century, Alexander of Hales made expounding the Sentences
a compulsory part of the theology degree, and from then on no one wishing
to become a Master in Theology could graduate without producing his
own commentary on the text, with the result that hundreds of copies of the
Sentences, and Sentences commentaries, are still extant.15 Sentences are the
archetypal twelfth-century method of theological teaching, which were
transformed into the theological summae of the thirteenth century.
Alongside the speculative theology of the sentences were volumes whose
focus was the more practical questions of Christian life. This literature of
pastoral care only really came into its own after the Fourth Lateran Council
of 1215, but it has roots in the twelfth century, building on Hugh of
St-Victor’s pioneering work, On the Sacraments of the Christian Faith (De
sacramentis christianae fidei).16 The twelfth-century scholar most famous
for working in this vein was Peter the Chanter who, like the Lombard, was
head of the Notre Dame school.17 His Summa of the Sacraments and
Counsel for the Soul (Summa de sacramentis et animae consiliis) deals with
hundreds of questions of practical and pastoral morality, from the effects of
baptism, to the extent of allowable taxation, to the binding power of oaths.
The layout of the text in these books is more conventional than that of
the materials for Bible study – broadly speaking, a heading followed by
a paragraph of continuous text. Nevertheless, although their text space may
be less inventive, the books incorporated other sorts of innovation, which
we can gather together under the heading of reader aids. In the
simple format versions of the glossed Bible (Figure 11.1), aside from the
three-column layout of the text, the books themselves are very plain.
Chapter and verse divisions were not settled until the end of the twelfth
century, so at best the books begin with numbered lists of headings (capitula)
which link to numbers in the margins of the page. There are generally no
running headers or book titles, and often little or no colour to mark textual
divisions. Since the glosses rarely begin with biblical lemmata, it can be
difficult to be sure which glosses belong with which part of the scriptural
text – and even, at times, which interlinear gloss should be attached to which
word in the line; each gloss simply begins with a paraph mark. These early
 Books of Theology and Bible Study 205
glosses are plain, brown books, which are not easy to navigate. The reader
must already know a lot about the text and its interpretation to get the most
out of using them. They are books for cognoscenti.
Individual readers appear to have noted the inadequacies of the layout,
adding their own lines or signes de renvoi to tie gloss to text, for instance, or
putting in chapter divisions and running headers. As the twelfth century
progressed, some but not all of these gaps were filled: the reader needed to
know less in order to be able to find his way around, but it is interesting
that some very simple improvements to the gloss layout, such as a means to
tie gloss and text, were never really instituted. When the simple gloss layout
evolved into the more complex versions, on pages with multiple columns
of writing, scribes added tiny symbols to link text that spread over more
than one column, so that the user knew where to carry on reading. Yet the
basic question of which gloss belonged where remained unaddressed, and
indeed became more difficult, since it became harder to maintain the
proximity of text and accompanying gloss than it had been in the simple-
format copies.
Nonetheless, some masters were continuing to think about the presenta-
tion of their work on the page, and the creation of an apparatus to aid
scholarship. Once again we encounter the central figure of Peter Lombard.
Along with the intercisum format, the most careful manuscripts of
Lombard’s commentaries preserve a wide range of readers’ aids. They have
book titles and chapter or capitula numbers as running headers and chapter
or capitula numbers in the margin and can use colour and a hierarchy of size
of initials to guide the reader. Most intriguingly, Lombard’s manuscripts
have a form of footnote – references to the authorities Peter is quoting.
Indeed, the placement of the references tells the user which part of the magna
glosatura he is reading: in the Psalms commentary the references are generally
in the top margin; in the Epistles commentary they are in the side margins.
Figure 11.3 illustrates the most elaborate version of the system, with vertical
red lines to indicate the presence and extent of quotation from the individual
authorities, who are named (here in red) in abbreviated form – AG
for Augustine, Amb for Ambrose, and so on.18 Above each name is
a pattern of dots and dashes (for instance, two vertical dots for Augustine,
two horizontal dots for Ambrose) peculiar to each. The dot-dash pattern is
used within the text of the commentary to mark the beginning of the
quotation from that authority; a single dot above the last word commonly
marks the end. A reader who knows the key to the dot-dash symbols can read
through the text knowing instantly who Peter is quoting, as well as precisely
how much of the text is quotation and how much is Peter’s own.
206 lesley smith
The planning involved in the reference system can be seen from the rubri-
cator’s guideletters at the outer edge of the page: the abbreviations and
symbols have all been carefully copied from an exemplar.
It is a characteristic of medieval manuscripts that the scholarly worth of
a book is not necessarily related to the expense with which it was made. As we
have seen, Cambridge, Trinity Coll. B. 5. 4 (Figure 11.3) is a richly decorated
version of Peter’s Psalms commentary, and the whimsical variety of the
imaginary creatures on its pages might lead us to imagine that it was the
medieval equivalent of a coffee-table book. Nothing could be farther from
the truth. Cambridge B. 5. 4 contains, in the intercisum format, two different
versions of Jerome’s translation of the Psalms as well as Peter’s magna
glosatura commentary, all made accessible by a hierarchy of script and
decoration which guides the reader through the three different texts on the
page. In addition, it has two sets of added notes, biblical cross-references and
Psalm numbers in running headers along the top of the page. The start of
each Psalm is numbered in the margin. Peter’s reference system appears in
alternating blue and red ink, with the dot-dash symbols carefully copied
within the text. Mixing pedantry with invention, the book even has messages
from the mouths of the Church Fathers: in the top right-hand corner (of
f. 135v, Figure 11.3), Augustine points to a quotation supposedly from his
works, but says (through the ‘speaking’ scroll by his side) ‘Non ego’ – ‘not
me!’19 Although it looks like a picture book, this highly ornate and costly
volume is stuffed full of Lombard’s particular reader aids; it would be
impossible for a twelfth-century book to be more scholarly.
Not all copies of Lombard’s text have the reference system, and not all
have it quite so fully as we see here, but it is common enough for us to know
that it was meant to travel with these texts; it represents a highly scholarly
and useful addition to the commentary on its own. But we cannot claim it to
be Peter’s own invention. Manuscripts of commentaries by Bede and
Rabanus Maurus sometimes note in the margins the (abbreviated) names
of the authorities whose opinions are cited. A similar system is also evident in
the commentary manuscripts of Gilbert de la Porrée, whose innovative page
layout we have already considered. Gilbert’s marginal references are not as
elaborate as Peter Lombard’s, but the idea is certainly the same, as is the
scholarly intention behind it. And Gilbert has another trick up his sleeve.
Following Cassiodorus, Gilbert indexed each Psalm according to its main
theme (On the Natures of Christ; On Love; On Lamentation and so on)
using a series of twelve symbols, and these can be found in many manuscripts
of the Psalms commentary.20 One early copy (Oxford, Balliol Coll. 36,
datable before 1166), which is written out continuously with lemmata, not
 Books of Theology and Bible Study 207
in Gilbert’s cum textu format, illustrates its usefulness. At Ps. 27, for instance,
a symbol (a backwards S) to the left of the initial A tells us that this is a Psalm
about the Passion and Resurrection; the Roman iii, above the indexing
symbol, marks the Psalm as the third of this type; and the Roman numeral
xxx below the symbol points forward to the next Psalm with this theme.
Sure enough, Ps. 30 is the next Psalm with the backward S index, and we can
see from the surrounding roman numerals that it is the fourth Psalm of this
type, and that the next is Ps. 56. This is a different sort of readers’ aid than the
references provided in the cum textu Gilbert manuscripts, but no less useful
for that. Whereas the references would help fellow commentators make
sense of the ingredients that went to make up the text, allowing them to
reuse them in their own exegesis, the indexing symbols would perhaps be
most helpful to preachers looking for themes for sermons and links between
biblical texts. Both types of readers’ aid take us a long way from the plain
Bible text with which we began. This is the Bible used for more than just
continuous reading; this is a text manipulated by scholars and preachers
according to their needs.
Another series of readers’ aids was much more visual in nature. For
beginners in the biblical classroom, Peter of Poitiers’ Compendium historiae
in genealogia Christi provided a series of useful diagrams to help make sense
of the text.21 These sometimes take the form of long parchment rolls –
a kind of medieval wallchart or poster – presumably to be tacked up on the
classroom wall; but they can also be found as a few folios in a manuscript
with other contents, often preceding one of the other texts we have met so
far. Some of what Peter provides are maps or plans, such as those for
Jerusalem and its Temple; but the core of the work is a series of biblical
genealogies, in the form of linked circles that explain who in the Bible is
related to whom. This genealogical interest was more than just pedantry or
a means of better following a complicated story. It was, for instance, an
important biblical argument that Jesus was part of the family of King
David, and the genealogies show proof of those links. Similar diagrams and
genealogical chains can be found in manuscripts of Peter Comestor’s
Historia scholastica, where they are within or alongside the text, rather
than separately produced. The diagrams – of Noah’s Ark, for example – in
Comestor’s Historia manuscripts are an instance of what we might call
visual exegesis, that is, of pictures employed not as decoration but as
illustration – sometimes even as part of the exposition of the text. It is
a practice that seems to be linked to the important school run
by Augustinian canons at the abbey of St-Victor in Paris; and certainly
the most extensive extant twelfth-century example is found in Richard of
208 lesley smith
St-Victor’s commentary on the Vision of the prophet Ezekiel. Richard is
trying to explain how Ezekiel’s notoriously complex description of the
rebuilding of the Temple in Jerusalem, destroyed by the Babylonians, can
be understood to describe an actual viable construction. His text includes
a series of large diagrams, common across copies of the text, and generally
accurately reproduced, in which he attempts to draft the proportionate
sizes and shapes Ezekiel provides. He is at particular pains to explain how
the building could be sited on the sloping side of the Temple Mount.
The diagrams are an integral part of the exegesis, and Richard refers to
them in the text. His point is a theological one: if Ezekiel’s figures are
simply impracticable, then the Bible becomes in some sense unreliable; but
if Richard can demonstrate that the text is correct in its detail as well as in
general meaning, then he has confirmed his belief in the literal truth of
Scripture.22
Both Richard and Peter Comestor were associated with St-Victor, whose
most famous scholar was the influential Master Hugh. One of Hugh’s
treatises, The Mystical Ark of Noah (De arca Noe mystica), is the subject of
much debate in the matter of visual reader aids. The text is a kind of
ekphrasis, a verbal description of a picture.23 It has at its centre a world
map, overseen by Christ and two seraphs. Some scholars believe that the
treatise must have been accompanied by a visual creation of the image
described, either on parchment or on the walls of Hugh’s classroom, so
that he could refer to the drawing as he taught the text. For others, however,
the text is intended to prompt students to create the image in their own
minds, as a portable aid to meditation, and it was never an actual artefact.
No copy of a suitable image has been found; but as we have seen, in the
mid-twelfth-century context of Paris and St-Victor, the use of images in
teaching was not uncommon.
Finally in this section, we should note some visual material that com-
bines pure decoration with the exegetical purpose of Richard of St-Victor’s
Ezekiel drawings or the illustrative diagrams of Peter of Poitiers and others.
Some Glossed books and commentaries begin with historiated initials that
fulfil both functions. An Oxford manuscript of the Gloss on Romans
(Bodl. Libr., Auct. D. 1. 13, f. 1r) is a particularly nice example.24
The initial P[aulus] marks the beginning of the biblical text – the first of
Paul’s Letters – and within the page-high letter are a number of scenes from
the Apostle’s life: his preaching; his escape from Damascus by being let
down from the walls in a basket; and his traditional (but not biblical) death
by beheading in Rome. The historiation sums up the whole context of the
Letters before the text begins, acting as a preview for new readers and an aid
 Books of Theology and Bible Study 209
to memory for others. It is also a wonderfully inventive and enjoyable
picture.
Decoration and historiation certainly continued beyond the confines
of the long twelfth century, but we cannot say the same for some of the
other elements of the books we have examined in this chapter. Although
glossed Bibles continued to be made in the thirteenth century in great
numbers, they were presented in exactly the same way as their twelfth-
century exemplars; there are no innovations in the later period.
The reason for this might be that, although the gloss continued to be
used, by the second quarter of the thirteenth century it was no longer at
the cutting edge of biblical research. Thirteenth-century (and later)
copies were either made outside Paris or were made there for use else-
where. Instead, in the 1230s Hugh of St-Cher and the Paris Dominicans
produced a new commentary on the whole Bible that updated the gloss
and took commentary in a slightly different direction, including
sentence-type material among the exposition.25 Hugh’s commentary
did not use the gloss page design, but went back to a two-column
lemmatised format. It was also known by a new name – postilla(e) –
which served to distinguish Hugh’s type of commentary from the Gloss
layout. But despite the new name, the layout of Hugh’s postillae was in
some ways a return to the past. The creativity of the twelfth century seems
to have fallen away, perhaps because the scribal effort involved in produ-
cing glossed books was too difficult and expensive: the Dominicans were
a new order dedicated to poverty and without a history of book collecting
to fall back on. Anything it produced had to be easily copied anywhere in
the Order.
But although the textual layout of postillae was more conventional than
that of the Gloss, the Dominicans did not entirely revert to previous forms.
The Order had its own innovations. Figure 11.4 illustrates a typical manu-
script of one of Hugh’s postillae. The textual layout is a lemmatised
commentary, written in very plain style, with the lemmata underlined in
the ink of the text; but although the text layout is plain and, seemingly,
old-fashioned, other readers’ aids have been retained, with running head-
ers, marginal subject headings and points in the argument numbered in the
margin. There is an index of subjects – and the ‘e’ at the top of the page is
part of the reference system it employs. Hugh and the Dominican team in
Paris were also responsible for a biblical concordance and list of corrections
to the scriptural text – two innovative projects that served the Dominicans’
need to be able to find their way quickly around the text, to use it for
preaching and teaching, rather than simply for their own private reading.
210 lesley smith

Figure 11.4 Hugh of St-Cher, Postilla on Revelation. Oxford, Bodleian Library,

Bodley 444, ff. 102v–103r.

They wanted to be able to find theological themes and problems, as well as

biblical quotations, and they had to be able to copy their books quickly and
accurately: the older, lemmatised commentary form allowed them to do
this in a space-efficient way.
 Books of Theology and Bible Study 211

Figure 11.4 (cont.)

Late twelfth- and thirteenth-century scholars show an increasing inter-

est in the hebraica veritas – the original language of the Bible. A group of
English manuscripts preserves Hebrew Bible texts with Latin translations
in a two-column page layout strongly reminiscent of Gilbert de la Porrée’s
212 lesley smith
commentary format.26 Similarly, the twelfth-century awareness of other
faiths included concern about Islam. Peter the Venerable, abbot of
Cluny, commissioned Robert of Ketton to translate the Qur’ān into
Latin.27 Copies of Robert’s translation, such as the earliest, twelfth-century
manuscript (BnF, Bibl. de l’Arsenal 1162), are laid out as a text with
marginal and interlinear glosses. Moreover, in the same way that paraph
marks were used to mark new glosses in the Glossed Bible, these texts
employ a special paraph sign to mark the beginning of a new Qur’ānic
section or sura. The similarity of this to a glossed book of the Bible is striking.
What we see in these twelfth-century books is not a straightforward
history of ‘progress’ and change: some of the innovation sticks and
some does not. We should perhaps not be surprised that the compli-
cated page design of the Glossed Bible proved too difficult – perhaps
too expensive – to be generally employed for new commentary in the
thirteenth century. Even the early printed copies of the gloss do not
manage to reproduce the intricacies of the complex format. It may also
be the case that the move from oral to written culture in the classroom
meant that books were being used in different ways that required
different layouts. Nevertheless, the experimentation and creativity of
the makers of twelfth-century books of Bible study and theology
remain a high point in the history of book design.

Notes
1. The classic study is Smalley 1983. For the development of the schools in
general, see Southern 1995, 2001. Smith 2001 illustrates many of the texts
discussed here.
2. Smith 2009. De Hamel, Glossed Books; Smalley 1935–6, 1937, 1984. Biblia cum
glossa ordinaria represents a modern facsimile of the 1480/1 editio princeps.
3. For Anselm and his school, see Smith 2009, 17–33; Giraud 2010, ch. 1; Southern
2001, chs. 1–4; Flint 1976.
4. In the Introduction to the facsimile, Biblia cum glossa ordinaria, Gibson states
that the gloss was often found as a nine-volume set, following Cassiodorus’
description of one of his bibles at Vivarium; but in practice, the situation was
much more varied. There was no single way of organising and gathering the
books. The five books of Moses, for instance, can be found together, in groups
or separately. Similarly, the gospels are often found in pairs, as are the four
Major Prophets. The book of Ruth is sometimes regarded as part of Judges, and
included in a volume with Joshua, but it can also be found with the historical
books about individuals – Tobit, Judith and Esther (and sometimes Ezra). Few
 Books of Theology and Bible Study 213
people seem to have owned a matching set of glossed books, but rather to have
collected as they could, which might include some duplicates, with a single
biblical book found in more than one volume: Smith 2009, 179–80.
5. See, for example, Peter Comestor and Peter the Chanter in Smith 2009,
209–15.
6. Smith 2009, 73–6. The revision is noted by the few scholars who have edited
parts of the Gloss, for example, Dove 1997, introduction; Andrée 2005; Zier
1993.
7. Gross-Diaz 1996; Van Elswijk 1966; Smith 2009, 121–34 and 195–9.
8. See the Prolegomena to the edition of Peter Lombard’s Sentences, and Colish
1994. For the layout of the texts, see Smith 2009, 130–7 and 200–4.
9. The commentary itself often begins with or includes short biblical lemmata,
so that in some cases the words of Scripture appear twice – once in the
intercisum block and again in the commentary column.
10. Cambridge, Trinity Coll. B. 5. 4 has been digitised at http://trin-sites-pub
.trin.cam.ac.uk/james/viewpage.php?index=453. Morgan and Panayotova
2015, no. 32.
11. Smith 2008.
12. Apart from the book of Genesis (ed. Sylwan), the Historia is still unedited.
The text is printed in PL 198. 1053–1722.
13. Giraud 2010 and Lottin 1959, vol. 5, discuss sententiae in general, as well as
printed collections by Anselm and others. Collections of sentences are among
the edited works of the theologians Robert of Melun, Oeuvres, vol. 3, and
Peter of Poitiers, Sententiae.
14. Peter Lombard, Sententiae in IV Libris Distinctae.
15. Alexander of Hales, Glossa in IV Libros Sententiarum Petri Lombardi.
16. Hugh of St-Victor, De sacramentis christianae fidei.
17. Peter the Chanter; Baldwin 1970.
18. For diagrams of the intercisum format and on the reference system, see Smith
2009, 130–7, and Smith 2001, no. 2.
19. The figure would appear to be correct. It seems to be pointing to a comment
on the titulus to Ps. 52 (Dixit insipiens), Pro Melech, which is not part
of Augustine’s exposition of the Psalm.
20. For a full description (and drawings) of the symbols, see Gross-Diaz 1996,
51–65 and Appendix 1: Cross Index. Smith 2008, figs. 3a and 3b, show Oxford,
Balliol Coll. 36, ff. 24v and 26r.
21. Moore 1936, 108, notes a contemporary mention of Peter’s invention of these
‘wallcharts’. See also Worm 2012.
22. Delano-Smith 2012, 2013.
23. Rudolph 2004 has argued extensively in support of the Ark as an actual
picture. The opposite view, that the picture was never meant to exist, is
taken by Carruthers 2008. For the text, see Hugh of St-Victor, De archa Noe.
24. For an image, see http://bodley30.bodley.ox.ac.uk:8180/luna/servlet/view/all/
what/MS.%20Auct.%20D.%201.%2013?os=0&pgs=50.
214 lesley smith
25. There is no modern edition of Hugh’s Postilla in totam bibliam, although
there are many incunable and early printed editions (often now digitised),
such as that by Johann Amerbach for Anton Koberger in Nuremberg, 1498–
1502.
26. Olszowy-Schlanger 2009; Smalley 1983, chs. 3 and 4.
27. Burman 2007, esp. ch. 3 and fig. 1.
 chapter 12

Logic
John Marenbon and Caterina Tarlazzi

Among twelfth-century books, logical manuscripts constitute a coherent

field of investigation.1, 2 Logic was studied in the schools attached to
cathedrals and, especially, in the many schools that grew up in Paris after
about 1120.3 The course in logic was fairly narrowly set, based on a small
number of logical texts and commentaries from the ancient world.
Twelfth-century authors wrote their own commentaries on these texts
and independent logical works; in both will be found some of the most
important philosophical thinking of the century.
The first section of this chapter sets out the ancient and late antique texts
used by logicians. There follow studies of the twelfth-century manuscripts
containing this material (Section 2), of commentaries by twelfth-century
logicians (Section 3), and of independent logical treatises from the period
(Section 4).

1 The Logical Curriculum in the Twelfth Century

In his Dialectica, probably written in 1110–5, Peter Abelard listed the
textbooks around which the study of logic focused in his time. They
comprise, he says, just seven books (codices), the work of three authors.
Aristotle himself wrote only two of them: the Categories and
On Interpretation, Porphyry one – an introduction to the Categories
(usually known as the Isagoge), while four by Boethius are generally in
use: his De divisione, ‘Topics’ (De topicis differentiis), and his ‘categorical
and hypothetical syllogisms’ (De syllogismis categoricis, De syllogismis
hypotheticis).4 To judge from the number of surviving manuscripts and
commentaries, the Isagoge, Categories, On Interpretation, and De topicis
differentiis were somewhat more widely and carefully studied than the
other three. Other logical works of Aristotle’s – known collectively as the
logica nova – were also known in the period: the Prior Analytics, Sophistical
Refutations, and, later in the century, the Topics and the Posterior Analytics.
215
216 john marenbon and caterina tarlazzi
But of the logica nova only the Sophistical Refutations gave rise to a lively
twelfth-century commentary tradition.5
Two further Boethian texts – the Introductio ad syllogismos categoricos,6
and the pseudo-Boethian De definitione (in fact by Marius Victorinus),
were copied to a lesser extent, and did not receive commentaries.7 Some
copies also continued to be made of the works which mainly constituted
the logical curriculum from ca. 800 to the late tenth century – textbooks
of logic such as the pseudo-Augustinian Categoriae decem (a paraphrase of
the Categories), Apuleius’ Periermenias (an introduction to syllogistic),
Cicero’s Topics with Boethius’ commentary and also Alcuin’s Dialectica
(based on the Categoriae decem and encyclopaedic accounts) – what can
be called a ‘Roman’ curriculum, because the works are either popularisa-
tions of Greek logic written for Latin-speaking readers in the Roman
Empire or, in the case of Alcuin’s Dialectica, derived directly or indirectly
from them.8
For studying the texts by Porphyry and Aristotle, Boethius’ commen-
taries were of great importance. Early on he wrote a commentary on the
Isagoge in dialogue form, using Marius Victorinus’ translation, and then
later a commentary using his own translation. A single commentary on the
Categories survives, and for On Interpretation Boethius produced a double
commentary – an introductory, literal prima editio, and a lengthy secunda
editio, which preserves much material from Porphyry’s lost long commen-
tary on the same work.9

2 Logical Textbooks and Boethius’ Commentaries

Ninety surviving twelfth-century logical manuscripts contain works from
the curriculum described in the previous section. They are listed in the
Appendix. In their content, these manuscripts are of four main kinds: [A –
Appendix I, II, III] those which contain some or all of the seven central
texts; [B – Appendix IV and V] those which contain Boethius’ commen-
taries; [C – Appendix VI] those which contain texts of the logica nova; and
[D – Appendix VII] a few logical manuscripts which reflect the archaic,
pre-eleventh-century ‘Roman’ syllabus. There is some overlap between
these groups, especially between [A] and [B], but in general they are
surprisingly distinct.10
In the Appendix, [A] is divided into three categories: (I) longer collec-
tions (containing all or all but one of the seven central texts); (II) shorter
collections (containing three or more of the central texts); (III) isolated
texts (containing one or two of the central texts). (I) is the biggest category,
 Logic 217
with fourteen MSS; there are nine in (II) and eleven in (III). But most of
the MSS in (II) have at least four of the central texts, and they may well
have lost leaves or quires which included the others. Most of the MSS in
(III) are fragments, which most plausibly came from complete or near-
complete collections of the central texts, since there is little evidence for
their circulation except in this way.11 The evidence of commentaries made
at the time suggests that Boethius’ textbooks on categorical and hypothe-
tical syllogisms and on division were less studied than the other four works,
but there is only a small variation in the total numbers of copies of all seven
central texts: Porphyry, Isagoge: twenty-six; Aristotle, Categories: twenty-
five; Aristotle, On Interpretation: twenty-two; Boethius, De topicis differ-
entiis: twenty-six; Boethius, De divisione: twenty-one; Boethius, De syllo-
gismis categoricis: twenty; Boethius, De syllogismis hypotheticis: fifteen.
The manuscript evidence therefore bears out Abelard’s description of the
core curriculum; and, although the dating of manuscripts within the
twelfth century is often imprecise, there is no suggestion from the evidence
presented in the Appendix that the core curriculum had changed by the
end of the century. But whereas Abelard speaks of seven codices, the norm
seems to have been a single volume, containing all seven of the central texts,
or most of them (see Appendix, section A).
These volumes were not normally glossed, and they very rarely contain
commentaries or other logical texts from the twelfth century.12 They are in
general more carefully presented and written than the cramped booklets
which do contain twelfth-century commentaries. Consider, for example,
the nine surviving complete or near-complete collections listed under [I] in
the Appendix, other than the five special cases which contain logica nova
texts too (see later in this chapter). The books are of medium size (aver-
aging about 250 mm × 100 mm; the smallest (8: Orleans, Bibl. mun. 265) is
151 mm × 95 mm, the biggest (6: London, Lambeth Palace 339) 260 mm ×
170 mm). All except one have some elements of decoration, in some cases
(2, 6, 7, 9) quite elaborate. No. 2 (Darmstadt, Landesbibl. 2282), in
particular, includes not only elaborate, monochrome initials, but
a frontispiece (Figure 12.1), depicting Dialectica, with her customary
snake in one hand and a model of Porphyry’s tree in the other.13 At each
corner is a famous philosopher: Plato, Aristotle, Socrates, and (depicted in
the same way) Master Adam, that is to say Adam of the Petit-Pont (in Paris,
where he had his school) or of Balsham (his home village). This volume
therefore presumably is connected to his school, one of those which
flourished in the second half of the century, when the manuscript was
written.
218 john marenbon and caterina tarlazzi

Figure 12.1 Depiction of Dialectica. Darmstadt, Universitäts- und

Landesbibliothek, 2282, f. 1v.
 Logic 219
Although the seven central texts form a distinct group, because of their
popularity, they are sometimes joined in the same collections by other logical
texts – usually examples of the ‘Roman’ curriculum studied in earlier
centuries – Cicero’s Topics in 1, 3, 4, 10, 14; Apuleius’ Periermenias in 5
and 14, and Boethius’ commentary on Cicero’s Topics in 23. The pseudo-
Boethian De definitione was sometimes added (1, 4, and 6),14 as was Boethius’
own Introductio in syllogismos categoricos (1, 3, 4, 20, 24) – regarded as
a supplement to, not a replacement for De syllogismis categoricis.
The texts of the ‘Roman’ syllabus, which had begun to go out of fashion
by the beginning of the eleventh century, were copied occasionally not just
with the central texts, but in what might be called archaic logical collec-
tions. The list in the Appendix (VII) omits copies of the pseudo-
Augustinian Categoriae decem included in collections of Augustine’s
works, and also copies of Cicero’s Topics and Boethius’ commentary
included in otherwise non-logical manuscripts. Of the five manuscripts
listed, one (87: Charleville, Bibl. mun. 187) combines two central texts with
archaic ones.
There was considerably more enthusiasm for the newly available texts of
the logica nova than for the ‘Roman’ ones. With an important exception,
these texts were usually grouped together (75, 80–4) or appear singly
(76–9, 85), but are not copied with the central texts (but 75 – Bologna,
Bibl. Univ. 4228 – includes a copy of On Interpretation). The exception
consists of no fewer than five ‘longer collections’ (1, 5, 11, 12, 13) which
unite three or more logica nova texts with a full collection of the central
texts. Two of these manuscripts (grouped together as a single collection, 1:
Chartres, Bibl. mun. 497, 498) belong to the assemblage which was
deliberately made by Thierry of Chartres in the mid-century of textbooks
on the seven liberal arts – his ‘Heptateuchon’.15 The other four are all
manuscripts from late in the century which should be considered as the
earliest examples of what would become a standard type of university
logical manuscript, collecting together the works of Aristotle’s Organon,
including Porphyry’s Isagoge, and, like them – and unlike most twelfth-
century collections – adding the Liber de sex principiis, a twelfth-century
composition on the six categories, a subject treated only fleetingly by
Aristotle himself. The manuscript evidence therefore bears out the idea
that, for most of the twelfth century, in the schools the logica nova was
studied apart from the main curriculum. But whereas the evidence of
commentaries and treatises might suggest that, of the logica nova, only
the Sophistical Refutations were widely studied before 1200, the evidence of
the manuscripts suggests otherwise. The other texts of the logica nova are
220 john marenbon and caterina tarlazzi
nearly as popular (Sophistical Refutations, fourteen copies; Topics, eleven;
Prior Analytics, nine; Posterior Analytics, six; Liber de VI principiis, three),
although only one commentary on the Prior Analytics is known, and no
commentary on the Topics or Posterior Analytics before the 1230s.16
The existence of university-type complete Organon manuscripts from the
last decade of the twelfth century or earlier (5, 11, 12, 13) also suggests that,
even if the twelfth-century schools with their characteristic interests con-
tinued to flourish up to and beyond 1200, the logical curriculum which
would characterise study in the universities was already becoming
established.17
The texts of the logical curriculum were difficult to approach without
a guide. Pre-twelfth-century manuscripts of the Categoriae decem and
Porphyry’s Isagoge quite often contain an extensive marginal
commentary.18 In the twelfth century, by contrast, the aids for study
took the form of independent commentaries, either by contemporaries
(see Section 3) or by Boethius (only for the Isagoge and the two works by
Aristotle). In the case of the contemporary commentaries, they are almost
always in separate manuscripts. Those by Boethius usually are. Thirty
twelfth-century manuscripts contain one or more of Boethius’ commen-
taries without any of the central texts – almost as many as the manuscripts
of the central texts, although the number of copies of each Boethian
commentary is considerably less: Commentary on Isagoge 1: seven; on
Isagoge 2: ten; on Categories: eighteen; on On Interpretation 1: eight; on
On Interpretation 2: sixteen. Perhaps because of the length of the commen-
taries, no manuscript collects all of them; indeed, most contain just
one, with an especially large number containing the most complex and
challenging of them, the second commentary on On Interpretation.19 It is
interesting that, although Boethius wrote his two commentaries on
On Interpretation as a pair, they are copied together in only one manuscript
of the period (59: BnF lat. 6400 F, ca. 1100).
More rarely, Boethius’ commentaries are combined with texts.
Surprisingly, in nearly half of the cases (70, 72–4), the logical texts do not
match up with the Boethian commentary, and in another two MSS (68–9)
there is only a partial match. The commentaries are written out separately
from the texts, except in one manuscript: (67: Bamberg, Staatsbibl. Class 15
[HJ.IV.9]) contains a text of the Isagoge and, separately, Boethius’ first
commentary, but it also has copies of Boethius’ commentary on the
Categories and his first commentary on On Interpretation with Aristotle’s
texts written in the margins (which are unusually wide, suggesting that this
presentation was planned in advance).
 Logic 221
3 Twelfth-Century Logical Commentaries
As mentioned earlier, twelfth-century logical commentaries are rarely found
in the same manuscripts as the ancient and late antique textbooks or
Boethius’ commentaries.20 Over the past few decades, a number of scholars
have tried to identify and catalogue manuscripts containing twelfth-century
logical commentaries.21 Although the majority is unpublished, most twelfth-
century logical commentaries have been transcribed by Yukio Iwakuma (of
Fukui Prefectural University), who has generously shared them with other
scholars. The method of the commentaries has been scrutinised and
a distinction has been made between literal commentaries (word-by-word
exegesis of the letter of the commented text), problem/question commen-
taries (discursive treatment of the problems raised by the commented text),
and composite commentaries (combining both aspects).22 It has been noted
that some commentaries are collections of glosses meant to supplement a pre-
existing commentary,23 and the features of commentary prologues have also
attracted attention.24
The dissemination of twelfth-century logical commentaries exhibits five
main characteristics. First, they tend to circulate in twelfth-century manuscripts
only. Second, the manuscripts containing them stand out for their poverty: lack
of decoration, poor parchment and minute and abbreviated handwriting are
fairly common. Third, the manuscripts are few – totalling well below 100.
Fourth, the commentaries are generally anonymous. Finally, and related to the
third point, most of the commentaries are usually found in a single copy. When
found in more than one copy, commentaries have two special characteristics.
On the one hand, their text tends to vary greatly from one manuscript to the
other. On the other, their manuscripts tend to be interrelated with respect to
content. All in all, the dissemination of logical commentaries seems to have
relied on limited material and human resources. Below the surface, however,
the matter is more complicated – probably, as we shall see, as a result of the
special sort of authorship pertaining to this sort of writing. In what follows, we
try to look more carefully at the fourth and fifth aspects: anonymity (I) and
single- versus multiple-copy commentaries (II). We then adumbrate the second
feature, the poverty of manuscripts (III), concluding with some considerations
on the best method for analysing this material (IV).
1. Logical commentaries are usually anonymous in their manuscripts. There
are a few exceptions, of which Peter Abelard stands out as the most
prominent.25 Several commentaries are attributed to him explicitly;26 not
all of them authentic.27 Otherwise, given the anonymity of twelfth-century
logical commentaries and frequently similar incipits, scholars today find it
222 john marenbon and caterina tarlazzi
best to identify each commentary using alphanumeric tags: ‘P’ stands for
Porphyry’s Isagoge; ‘C’ for the Categories; ‘H’ for On Interpretation (Peri
hermeneias); ‘SE’ for the Sophistici Elenchi; ‘B’ for Boethius’ De differentiis
topicis; ‘D’ for De divisione; ‘SC’ for De syllogismis categoricis; and ‘SH’ for
De syllogismis hypotheticis; the various commentaries on the Isagoge are thus
named ‘P1’, ‘P2’, and so on, and similarly for the other works.28
II. Each logical commentary usually survives in one single manuscript.
Exceptions to the rule are rare, and none of them is disseminated in more
than a handful of manuscripts. For instance, of the eighteen commentaries
on De differentiis topicis dateable to the period from before 1100 to 1200, only
two (B8 and B12) are transmitted by more than one manuscript (one by three
and one by two).29 Commentaries transmitted by more than one manuscript
seem to have been particularly influential pieces of writing dating from the
early twelfth century, such as P3, C8, and H11.30 As mentioned earlier, such
‘standard’ commentaries have two special characteristics, which warrant
further discussion. First, when a commentary exists in more than one
manuscript, it frequently has a significantly different textual form in each
(II. 1). Second, when a commentary is found in more than one manuscript,
such manuscripts are usually themselves interrelated (II. 2).
II. 1 The fluidity of logical commentaries has been observed at different
levels.31 On the one hand, when transmitted by more than one manuscript,
significant sections of what is otherwise the same commentary cannot be
collated. On the other, sections of a certain commentary are found in what
is otherwise a distinct commentary – as comments either on the same
passage, or on a different passage of the same primary text, or on a different
passage of a different primary text. The same text therefore seems to exist
in different manifestations. The very notion of the same text becomes
problematic. Are two pieces of writing which share significant sections
two versions of the same commentary or two separate (but related) com-
mentaries? The simple and elegant way of referring to each commentary by
means of its alphanumeric tag can face difficulties. We give two examples;
the first is related to the Isagoge commentaries, the second to the Categories
commentaries. P15, prima facie, seems distinct from both P3 and P16.
Closer inspection, however, reveals that P15 has significant sections in
common with each, to the point that it was employed in editing P3.
In turn P3 exists in four manuscripts with significant variants among
them. What is more, P3 also shares two passages with P14.32 The same
happens in commentaries C7 (BnF, lat. 17813, ff. 19bis–54) and C14 (Assisi,
Bibl. del Sacro Conv. 573, ff. 15–48) with respect to C8. In turn C8 exists in
 Logic 223
four manuscripts with significant variations. Such variations are unevenly
distributed (they are concentrated in the second half of the commentary,
from chapter 6 of the Categories onwards).33 Textual fluidity may also shed
light on the anonymity of commentaries. Far from being due simply to
a missing inscription, anonymity seems to be a result of this very fluidity.
Some commentaries might have been reworked by different masters
(whose views might not always coincide). In some cases, the text appears
to be layered, with new interpretations added to an existing commentary.34
II. 2 Manuscripts of multiple-copy commentaries also tend to be inter-
related. By ‘interrelated manuscripts’, I mean manuscripts that transmit
similar collections of twelfth-century logical texts (commentaries, inde-
pendent treatises, or shorter notes). Their relationships are as follows. BAV
Reg. lat. 230 transmits a version of C8, a commentary also found in BnF lat.
13368; BL Royal 7 D. xxv; BSB Clm 14458; and, in a slightly different form
(= C14), Assisi, Bibl. del Sacro Conv. 573.35 The Vatican manuscript also
transmits B8, also found in Orleans, Bibl. mun. 266, and Paris, Bibl. de
l’Arsenal 910. This is followed by H11, again in common with BnF lat.
13368. In turn, BnF lat. 13368 transmits (in addition to single-copy texts and
to the multiple-copy texts that have been mentioned) a treatise De gene-
ribus et speciebus and one on modals, which are also found in the afore-
mentioned Orleans, Bibl. mun. 266; along with D8, also found in the
aforementioned Assisi, Bibl. del Sacro Conv. 573; and P3, again found in
the Assisi manuscript as well as in Bodl. Libr. Laud. lat. 67, and St Gallen,
Stiftsbibl. 134 (and in Dublin, Trinity Coll. 494, where it has a slightly
different form, P15, which has sections in common with commentary P16
of the aforementioned BSB Clm 14458). Assisi, Bibl. del Sacro Conv. 573
has already been mentioned for several relationships. In addition, it trans-
mits H9, found in Orleans, Bibl. mun. 266. Orleans, Bibl. mun. 266 is an
extremely rich collection of twelfth-century logical texts. Some of them are
found in this manuscript alone. Of those also copied elsewhere, some have
been mentioned already: in addition, the Orleans MS transmits SH3, also
found in BSB Clm 14458 and St Gallen, Stiftsbibl. 134 (and, in fragmentary
form, SH9, in BSB Clm 14779). The Orleans MS also transmits D1, also
found in St Gallen, Stiftsbibl., 134, and in Paris, Bibl. de l’Arsenal, 910.36
Such complicated relations show that, when manuscript a transmits
a commentary x which is also found in manuscript b (that is, a multiple-
copy commentary), a and b often share other texts as well; and/or they are
related to a single third manuscript c that brings together material found
separately in a and b; and/or a shares content with c and b shares different
224 john marenbon and caterina tarlazzi
content with d, and c and d are themselves related either directly or
indirectly. The pattern is multiplied, resulting in a high number of rela-
tionships. This all seems to suggest that interrelated manuscripts transmit
texts originating from the same milieu.
III. Some of the manuscripts that transmit twelfth-century commentaries
are homogeneous and of fairly high quality.37 Generally speaking, however,
they are non-homogeneous codices of rather inferior quality.38 As a rule,
manuscripts have no decoration and are crafted with great parsimony.
The handwriting, for instance, is minute; almost every word is abbreviated
(sometimes highly);39 and the same hand is rarely responsible for more
than one quire or one text. The most interesting feature of manuscripts
transmitting twelfth-century commentaries, however, is their non-
homogeneity, to which we now turn.
Different codicological units are identified by features such as change in
the script, mise-en-page, and text; lack of catchwords at the end of quires;
and blank pages.40 Manuscripts transmitting twelfth-century logical com-
mentaries can be non-homogeneous in two different ways. In some cases,
a codicological unit transmitting logical texts is found with entirely unrelated
units, differing with respect to both subject matter and date. An example is
BnF lat. 3237, transmitting P17. P17 is read on a quaternion (ff. 123–130,
f. 130v blank). The manuscript is made of thirteen other codicological
sections dating from various centuries, the subjects of which range from
theology to liturgy to logic.41 Given the dating of certain sections, this
manuscript must have been assembled much later than the twelfth century
when the section on logic was written.
In other cases, however, more than one (or even all) codicological units
contain twelfth-century logical texts (and date from the twelfth century, as is
usual for this material). Most interrelated manuscripts mentioned earlier
belong to this category. Probably the clearest examples are Orleans, Bibl.
mun. 266, and BnF lat. 13368. The Orleans manuscript transmits twenty-
eight twelfth-century logical commentaries and treatises, written by several
hands dateable to the middle and second half of the century.42 At least three
codicological units have been identified (pp. 5–120; 121–290; 291–300).43
Their number, however, may be higher. For instance, pp. 153–70 seem to be
a distinct section (a quaternion plus a singleton). The content of pp. 153–70
is found in exactly the same order in the Parisian manuscript, ff. 168r–177r,
also a distinct codicological unit. The Parisian manuscript is, in fact, an
example of both sorts of composition. The manuscript has several codico-
logical units, and ff. 128–231, containing logical texts, even have their own
 Logic 225
independent foliation. But this logical section, in turn, is composite.
Iwakuma identifies five codicological units in it, dating from slightly differ-
ent periods between 1130 and 1160.44
Non-homogeneous manuscripts of the second sort are extremely inter-
esting because they share some features with the entirely homogeneous ones.
The case can be made for such material having been bound together already
in the twelfth century. Interrelated logical manuscripts are never related to
one another in their entirety. They are always related partially, via a given
codicological unit (as in the Orleans-Paris example cited earlier). A way of
explaining the partial relationships could be that such units were booklets,
which enjoyed some form of independence before being bound with the
units accompanying them today.45 And if a unit transmitting twelfth-
century logical texts circulated independently for some time and is now
bound with units also transmitting twelfth-century logical texts, the fact
seems meaningful. Of course, as Erik Kwakkel has highlighted, there are
limitations to what can be united in a composite manuscript, particularly
with respect to page height.46 However, twelfth-century logical commen-
taries are, as has been said, quite rare. Their content and authors are also
difficult to identify, as rubrics and decoration are frequently lacking. It is
hard to see how somebody much later than the twelfth century would have
the skill or will to identify material as logic from this period and collect it
together. So non-homogeneous manuscripts containing logical commen-
taries in more than one codicological unit may result from an intentional act
of collecting this material at a time close to when they were written. Careful
analysis of the content frequently supports such a hypothesis. For instance, it
has long been noted that the texts transmitted by the Orleans manuscript in
its different codicological units all report the views of Joscelin of Soissons and
William of Champeaux. Unfortunately, too little is known of the purpose of
manuscripts transmitting twelfth-century logical commentaries, their pro-
duction, or their intended readership for firm conclusions to be drawn.
We do not, for instance, know whether the composition of a particular
manuscript stems from the way in which an author assigned his writings for
reproduction, or from an act of collecting a certain kind of material per-
formed by an interested reader, or both.47
IV. Such observations prompt a final remark about the best method to
investigate this kind of material. Scholars investigating twelfth-century
logical commentaries are confronted by a peculiar lack of unity.
The text, existing in different versions, is not a single entity; in many
cases, there is probably more than one author; and the dissemination of
226 john marenbon and caterina tarlazzi
many of these texts took place through booklets rather than homogeneous,
higher-quality manuscripts. Researchers can try to accommodate these
features into their approach to the subject in two ways. On the one
hand, they can try to follow such lack of unity by adopting a ‘modular
approach’.48 Texts abound with lists of contemporary opinions on given
problems, and often indicate the name of the magister supporting a certain
opinion (or at least his initials). Scholars can therefore follow a problem-
by-problem approach (collecting all the opinions on a certain problem) or
a master-by-master approach (collecting all the opinions of a certain
master).49 Another way of studying these writings, we suggest, is to place
weight on the manuscript itself as an important element of unity within
this material. Indeed, the juxtaposition of texts in homogeneous manu-
scripts is usually meaningful. BnF lat. 17813 (Figure 12.2) transmits four
texts all close to the teachings of William of Champeaux on realism. Milan,
Bibl. Ambrosiana 63 sup. is a collection of Abelard’s and Abelardian
commentaries. Non-homogeneous manuscripts collecting several codico-
logical units of twelfth-century logic may also have been gathered in the
twelfth century following a certain purpose, as has been said. Thus the
manuscript, it can be argued, stands as an important unity to which
scholars should resort in their investigations of this fluid and multifaceted
material.

4 Twelfth-Century Treatises and Independent Works

Twelfth-century logicians did not just write commentaries. They produced
independent works, ranging from longer treatises, which attempt to cover
the whole of the subject. Only a few survive: Garlandus’ Dialectica (prob-
ably ca. 1100–10), Abelard’s Dialectica (ca. 1110–5), Adam of Balsham’s Ars
disserendi (1132); from the mid-century the ‘Compendium logicae
Porretanum’, the ‘Ars meliduna’, and the ‘Introductiones montane maiores’;
and from the 1180s William of Lucca’s Summa dialetice artis. Except for
Garlandus’ Dialectica, of which there are two extant copies, all these
treatises each survive in just one twelfth-century manuscript. Some did
not survive at all: for instance, Abelard very probably wrote a Grammatica,
which – despite the title – was an important contribution to logic, but no
copy of it remains.50 These works were just a little more likely than the
commentaries to be explicitly attributed to an author in the manuscript –
as in the case of Abelard’s Dialectica, one of the manuscripts of Garlandus’
Dialectica, and the Ars disserendi. The other works, though, are anonymous
(and the titles given in inverted commas are ones made up by modern
 Logic 227

Figure 12.2 Example of phytomorphic initial (commentary P14). Paris, Bibliothèque

nationale de France, lat. 17813, f. 14v.
228 john marenbon and caterina tarlazzi
editors). These treatises also differ from the commentaries in that there are
no close relations between them. Where they do not occupy the whole of
a manuscript (or an independent part of one), they are not normally
combined with other logical material.
There was also a different sort of, usually shorter, logical treatise, called
‘Introductiones’, devoted to explaining the art of making and evaluating
arguments, especially on the basis of Aristotle’s On Interpretation.51 Yukio
Iwakuma has listed twelve such treatises. In one case at least, Introductiones
in two different manuscripts are closely connected;52 and one set, the
Introductiones montane minores, is copied in two manuscripts.
The manuscripts transmitting them often contain other logical treatises
or commentaries: for example, one of the two manuscripts containing the
Introductiones montane minores contains another Introductiones text and
four other short logical works, and the other contains seven further logic
works or notes and three commentaries.
There are, as previous comments will have hinted, many short
logical notes and works from the twelfth century in manuscripts of
the time, ranging from what are hardly more than jottings about
masters’ opinions to treatises on particular themes, such as Abelard’s
De intellectibus or the De generibus et speciebus. Usually, a number of
such pieces will be found together in a single manuscript, often
together with Introductiones and/or commentaries. The manuscripts
of the Introductiones montane minores are one example. Another is
Orleans, Bibl. mun. 266, discussed extensively in Section 3: it contains
twelve commentaries, four collections of notes on a textbook, and
twelve twelfth-century logical works or passages.53
It is common to all this material, from the comprehensive treatises to the
brief notes, that it was almost never copied after the beginning of the
thirteenth century. The one exception is Adam of Balsham’s Ars disserendi,
which survives in a twelfth-century manuscript and, in a different (and
partly more authentic form) in a manuscript from ca. 1300 and its fifteenth-
century copy.54

5 Conclusion
Twelfth-century logical manuscripts fall into three largely distinct groups.
There are the manuscripts of the ancient and late antique textbooks, and
Boethius’ commentaries; those transmitting the few long, comprehensive
treatises on logic by twelfth-century authors; and those which contain
twelfth-century commentaries and/or shorter twelfth-century logical
 Logic 229
writings, ranging from Introductiones to notes. Within the first group there
is a distinction between manuscripts containing the central texts of the
logical curriculum, and those with texts of the logica nova (including
Aristotelian texts without a twelfth-century commentary tradition). And,
although a few manuscripts contain the central texts along with one or
more of Boethius’ commentaries, the usual pattern is for Boethius’ com-
mentaries to be transmitted separately from the texts with which they are
concerned.
A similar and even sharper division can be seen between the twelfth-
century commentaries and the textbooks on which they comment,
which are hardly ever contained in the same manuscript. The explana-
tion for this surprising separation of linked material probably lies in the
way the manuscripts functioned in the schools. But there are as yet no
clear answers to the question of who used the manuscripts, how they
used them, and what were the interrelations between lecturing, learning
by heart, writing on wax tablets, and copying on parchment.55 Indeed,
one of the important contributions which study of the twelfth-century
book can make to understanding the intellectual life of the time is in
providing evidence to respond to such queries. This chapter, therefore,
is not so much a finished piece of work as a collection of data, along with
some suggested interpretations, intended to contribute to the wider
project of understanding the methods and aims of the twelfth-century
schools.

APPENDIX
This appendix lists the surviving twelfth-century manuscripts of ancient
and late antique logical texts and commentaries (‘relevant material’). Only
twelfth-century manuscripts are included (so e.g. ‘after 1150’ means
1150–1200), although I have included manuscripts listed as being from
the turn of the twelfth century (I use the designation ‘ca. 1100’), and the
turn of the thirteenth century (I use the designation ‘ca. 1200’). For MSS
dated to early in the century, I say ‘early’, and for those late in the century,
‘late’. Where no date is given, then the best dating available to me is simply:
twelfth-century. Where booklets not containing relevant material, or not
from the twelfth century, are bound with manuscripts listed, they have
been ignored. I have tried to indicate where a manuscript is made up of
more than one distinct booklet containing relevant material (see 71 and
80). In most other cases, the relevant material is not in distinct booklets,
although the catalogues on which I rely are not always clear enough to rule
230 john marenbon and caterina tarlazzi
out the possibility. My main sources of information have been the Codices
volumes of Aristoteles Latinus (AL) and the four volumes so far published of
Codices Boethiani (CB), the datings of which I have preferred in cases of
disagreement. I have also used the lists of manuscripts in Boethius, De
syllogismo categorico and Introductio ad syllogismos categoricos, and
Thomson, Catalogue, I and II. I have been able both to check and to
supplement this information from the very valuable collection of material
which Yukio Iwakuma makes available to those working in the field, and
a number of items (as indicated in what follows) are noted by him alone.
I and all scholars in this field owe an immense debt to Professor Iwakuma’s
vast and detailed work.

Abbreviations
Central Works
P: Porphyry, Isagoge
C: Aristotle, Categories
H: Aristotle, On Interpretation
B: Boethius, De topicis differentiis
D: Boethius, De divisione
SC: Boethius, De syllogismo categorico
SH: Boethius, De syllogismis hypotheticis
S: all of the previously listed seven works (P, C, H, B, D, SC, SH)
ISC: Boethius, Introductio ad syllogismos categoricos

Boethian Commentaries
InP1/ InP2: Boethius, Commentary on Isagoge, prima editio (dialo-
gue)/secunda editio
InC: Boethius, Commentary on Categories
InH1/ inH2: Boethius, Commentary on On Interpretation prima edi-
tio/secunda editio

Works of the ‘Roman’ Curriculum

Alcuin: Alcuin, De dialectica
Apuleius: Apuleius, Periermenias
CatX: pseudo-Augustine, Categoriae decem
Defin: Marius Victorinus, De definitione
InTC: Boethius, Commentary on Cicero’s Topics
TC: Cicero, Topics
 Logic 231
Logica nova
lsp: Liber de sex principiis
pa: Aristotle, Prior Analytics
posta: Aristotle, Posterior Analytics
se: Aristotle, Sophistical Refutations
ta: Aristotle, Topics

A Central Texts

I Longer Collections (Containing All or Almost All of the Seven

Central Texts): 14
1. <AL474,465> Chartres, Bibl. mun. 497, 498 [ca. 1150] (destroyed), S,
ISC, Defin, TC, se, pa, ta
2. <AL848> Darmstadt, Universitäts- und Landesbibl. 2282 [late], S
3. <AL909> Erlangen, Universitätsbibl. 191 [late], S, ISC, TC
4. <AL1388; CBIII.121> Florence, Bibl. Laurenz. S. Marco 166 [after
1150], S, ISC, Defin, TC
5. <AL51; CBII.Austria23> Klosterneuburg, Stiftsbibl. 1098 [1170/90],
S, TC, Apuleius, lsp, pa, posta, se, ta
6. <AL287; CBI.146> London, Lambeth Palace 339 [after 1150], S,
Defin, TC
7. <AL1452; CBIII.200> Milan, Bibl. Ambrosiana M 62 sup. [ca. 1100],
P, C, H, B, D, SC
8. <AL435> Orleans, Bibl. mun. 265 [12 ex./13 in.], P, C, H, B, SH
(fragments), SC
9. <AL1511; CBIII.260> Padua, Bibl. Antoniana 553 scaff. XXII [ca.
1150], S
10. <AL765> Tours, Bibl. mun. 676 [ca. 1200, perhaps later] (destroyed),
S, TC
11. <CBIII.413> BAV Borgh. 131 [ca. 1200], S, TC, lsp, se, pa, posta
12. <AL1871; CBIII.539> BAV Vat. lat. 2978 [ca. 1200], P, C, H, B, D,
lsp, se, pa, posta, ta
13. <AL1875; CBIII.543> BAV Vat. lat. 2982 [ca. 1200], P, C, H, lsp, se,
pa, posta, ta
14. <AL1645; CBIII.380> Venice, Bibl. Naz. Marciana lat. Z 273 (=1574)
[ca. 1200], S, TC, Apuleius
232 john marenbon and caterina tarlazzi
II Shorter Collections (Containing Three or More of the Central Texts): 10
15. <AL1386; CBIII.118> Florence, Bibl. Laurenz. S. Marco 125, C, H, B,
SC, SH
16. <AL147; CBII.Netherlands15> Leiden, Universiteitsbibl. BPL 1925,
P, C, H, B, D
17. <AL2044; CBI.87> BL Add. 18342 [after 1150], P, C, H, D
18. <AL379; CBI.237> Oxford, Trinity Coll. 47 [before 1150], P, C, H,
B, ta
19. <AL1526; CBIII.269> Padua, Bibl. Universitaria 1688 [late], C, P, B,
D, SC
20. <Iwakuma> BnF lat 6400 G, B, D, SC, SH, ISC
21. <Iwakuma> BnF lat 12959, B, D, SC, SH
22. <Iwakuma> BAV Ottob. lat. 1974 [ca. 1100], P, C, H, B
23. <CBIII.485> BAV Reg. lat. 1649 [ca. 1100], B, D, SC, SH, InTC
24. <CBIII.526> BAV Vat. lat. 1722 [before 1150], B, D, SC, ISC

III Isolated Texts (Containing One or Two of the Central Texts): 11

25. <BIII.12> Assisi, Bibl. del Sacro Conv. 573, P (fragment)
26. <AL2123> Bern, Burgerbibl. 702, P (fragment)
27. <AL258; CBI.25> Cambridge, Fitzwilliam Museum Maclean 165,
C, B
28. <CBI.155> London, private owner [ca. 1200], B (fragments)
29. <Thomsen Thornqvist> BSB Clm 14503, ISC (fragment)
30. <Iwakuma> BSB Clm 29246(1) [late], P (fragment)
31. <Iwakuma> BSB Clm 29246(2) [late], P (fragment)
32. <Iwakuma> BSB Clm 29384(23) [late], D (fragment)
33. <AL1493> Naples, Bibl. Nat. Vindob. 47 (Martini 7) [late],
C (fragment)
34. <Iwakuma> Pommersfelden, Schlossbibl. 16/27664, SC
35. <AL2116> Trier, Bistumsarchiv. 6, P

B Boethius’ Commentaries

IV Boethian Commentaries (Manuscripts Containing One or More

of Boethius’ Commentaries, but None of the Central Texts): 30
36. <AL2087> Bamberg, Staatsbibl. Class 15 (HJ.IV.9), InP1, InP2
37. <CBII.Switzerland22> Bern, Burgerbibl. 332, InH2
 Logic 233
38. <AL2132; CBIV.Spain17> Córdoba, Bibl. de la Catedral 153, H2
39. <AL1943; CB1.57> Durham Cath. A. II. 11 [ca. 1150], InC
40. <AL2097> Erfurt, Universitätsbibl. Dep. Erf. Cod. Amplon.
8° 66 InC
41. <AL1317; CBIII.60> Florence, Bibl. Laurenz. Plut.71, 21, InH2
42. <AL1382; CBIII.114> Florence, Bibl. Laurenz. S. Marco 102, InP1,
InP2
43. <AL1383; CBIII.115> Florence, Bibl. Laurenz. S. Marco 113, InP2,
InH2
44. <AL1384; CBIII.116> Florence, Bibl. Laurenz. S. Marco 114 [after
1150], InC
45. <AL1385; CBIII.117> Florence, Bibl. Laurenz. S. Marco 124, InH2
46. <AL1387; CBIII.119> Florence, Bibl. Laurenz. S. Marco 130, InP2,
InC, InH2
47. <AL933> Göttingen, Universitätsbibl., Apparat. Diplom. 10E.
Mappe IV.1, InC (fragment)
48. <AL1947> Hereford Cath. O. VIII. 6, InC (fragments)
49. <AL2095> Kues, St. Nikolaus-Hospital 190 [ca. 1100], InP2
50. <AL293; CBI.98> BL Arundel 348 [before 1150], InH2
51. <CBI.121> BL Harley 2713, InP1
52. <AL1018> BSB Clm 331, InH1
53. <AL434> Orleans, Bibl. mun. 80 [ca. 1100], InH2
54. <AL436> Orleans, Bibl. mun. 269 [ca. 1100], InP1, InP2, InC
55. <AL537> BnF lat. 1954 [ca. 1100], InC
56. <AL539> BnF lat. 2858, InC
57. <AL579> BnF lat. 6400C [ca. 1100], InH2
58. <AL580> BnF lat. 6400D, InH2
59. <AL582> BnF lat. 6400 F [ca. 1100], InH1, InH2
60. <AL652> BnF lat. 15104, InH2
61. <CBIII.284> Pistoia, Archivio Capit. C. 77 [ca. 1200], InP2
62. <AL190> Prague, Bibl. Cap. Metr. L. 54, InH1
63. <Iwakuma> Tournai, Bibl. mun. 74 bis [ca. 1100] (destroyed), InP2
(fragments)
64. <AL2189; CBIII.522> BAV Vat. lat. 566, InH2, InC
65. <AL23> olim Vollbehr, Dr. O. (Washington, DC), present where-
abouts unknown, 9 [late], InP1, InC
234 john marenbon and caterina tarlazzi
V Boethian Commentaries with Texts (Manuscripts Containing One
or More of Boethius’ Commentaries, and One or More of the Central/
Peripheral Texts): 9
(a) With Central Text(s)
66. <AL406> Avranches, Bibl. mun. 229, P, C, H, InP1, InC, InH1
67. <AL790> Bamberg, Staatsbibl. Class. 10 (H.J.IV,7), C, H, InC,
InH1
68. <AL267; CBII.Switzerland32> Cologny-Genève, Bibl. Bodmeriana 9
[ca. 1100], C, InC, H
69. <AL43; CBII.Austria20> Klosterneuburg, Stiftsbibl. 671, C, InC,
B, D
70. <AL581> BnF lat. 6400E [ca. 1100], D, SH, Defin., InH2
71. <CBII.Switzerland62> St Gallen, Stiftsbibl. 831 [two parts, the first
ca. 1100, the second twelfth-century], P, InP1; TC in first, earlier part
of MS
72. <CBIII.499> BAV Rossi 537 [ca. 1150], B, SC, inP1, InP2, InC

(b) With Peripheral Text

73. <AL575> BnF lat. 6398, InC, InH2, Apuleius
74. <AL747> Rouen, Bibl. mun. 932, InH2, Alcuin

C Other Ancient Textbooks

VI Logica nova (Manuscripts Containing Only Texts of the Logica

nova): 11
75. <AL2144; CBIII.27> Bologna, Bibl. Univ. 4228 [ca. 1150],
H (fragment), se, pa, ta
76. <AL243; CBI.52> Cambridge, Trinity Coll. O. 7. 9 (1337) [ca.
1200], se
77. <AL470> Chartres, Bibl. mun. 92 (destroyed), posta
78. <AL473> Chartres, Bibl. mun. 190 (destroyed), se
79. <AL1330; CBIII.95> Florence, Bibl. Laurenz. Ashburnham 1459
(1382), se
80. <AL1361; CBIII.84> Florence, Bibl. Laurenz. Plut. 11 9, se, ta; pa
[two parts, part one twelfth-century, part two, containing pa, ca.
1200]
 Logic 235
81. <AL1420; CBIII.148> Florence, Bibl. Riccardiana Ricc. 126 (L.IV.6)
[late], se, pa, posta, ta
82. <AL488> Laon, Bibl. mun. 435 ter [ca. 1200], se, ta
83. <AL1450> Milan, Bibl. Ambrosiana I. 195, se, pa
84. <AL1065> BSB Clm 16123 [ca. 1200], se, ta
85. <Iwakuma> BSB Clm 29384(25) [late], ta

VII Archaic Collections: 5

86. <AL38> Admont, Stiftsbibl. 742, Defin., Apuleius
87. <AL2056> Charleville, Bibl. mun. 187, B, D, Defin. CatX (also,
Augustine, De dialectica)
88. <Iwakuma> BSB Clm 22292, Apuleius, Defin. (also, Martianus
Capella, De nuptiis, Book IV [on logic])
89. <AL2190; CBIII.523> BAV Vat. lat. 567, CatX, InTC
90. <Iwakuma> Vercelli, Bibl. Capitolare CXCIX (35), CatX

Notes
1. John Marenbon is the author of Sections 1, 2, 4, and the Appendix, Caterina
Tarlazzi of Section 3.
2. Often twelfth-century authors call the subject dialectica, but sometimes this
word is used to refer specifically to topical inferences. For this reason, and
because of the various contemporary connotations of ‘dialectic’, ‘logic’ is
a better term to use.
3. For a basic introduction with bibliography (to which should be added Rosier-
Catach 2011), see Marenbon 2007, 131–66.
4. Abelard, Dialectica, 146: 10–7.
5. See later in this chapter for the question of how widely the logica nova was
studied in the twelfth century.
6. The Introductio is probably Boethius’ reworking and extension of Book I of De
syllogismis categoricis, and it is a narrower, more advanced work. The standard
title used here has no manuscript authority: indeed, De syllogismis categoricis is
often called Introductio ad categoricos syllogismos, and the Introductio called
Antepraedicamenta.
7. There are, however, glosses to the copy of De definitione in BSB Clm 22292, ff.
6v–15r (Iwakuma).
8. On the ‘Roman’ curriculum, see Marenbon 2013a, 176–9.
9. On Boethius’ commentaries, see Ebbesen 2009, and Magee and Marenbon
2009, both in Marenbon 2013, the second at 305 (with full details of
editions).
236 john marenbon and caterina tarlazzi
10. I touched on the issues discussed here in Marenbon 2013a, 181–2.
The complete survey on which the present chapter is based enables the rather
qualified conclusions there to be presented more firmly and, in some cases,
corrected.
11. The exceptions are 34 (Pommersfelden, Schlossbibl. 16/27664), which, unu-
sually, combines a central text with medieval commentary (which is, however,
most probably late eleventh-century: see Hansen 2005, 45–7), and 35 (Trier,
Bistumsarchiv 6), where an incomplete copy of the Isagoge appears along with
various texts and passages connected with the liberal arts.
12. There are a few exceptions. Besides the early 34 (Pommersfelden, Schlossbibl.
16/27664), mentioned in n. 11, 25 (Assisi, Bibl. del sacro conv. 573), which
includes a fragment of the Isagoge, is one of the important collections of
twelfth-century commentaries, discussed in Section III, and 27 includes,
besides copies of the Categories and De topicis differentiis, twelfth-century
commentaries on the Isagoge (P6) and Categories (C6). 15 has extensive glosses.
13. See Figure 12.1; images of the whole manuscript can be found at http://tudigit
.ulb.tu-darmstadt.de/show/Hs-2282/0004.
14. The status of De definitione is hard to define: in its content, it belongs to the
Roman curriculum, but it became known only along with Boethius’ genuine
textbooks, in the late tenth century: see Marenbon 2013, 185–6.
15. See Jeauneau, ‘Prologue’.
16. On the Topics, see Green-Pedersen 1984, 87; but Iwakuma has discovered
a fragment of a Topics commentary in Worcester Cath. Q. 12, ff. 393r–4v,
dated to ca. 1200. On the Prior Analytics, see Thomsen Thornqvist 2010.
17. SE15 (now dated to 1204 or later: Thomson 2013, 2. 160) shows that the
twelfth-century schools continued after the beginning of the thirteenth
century: cf. Ebbesen 2011.
18. See Marenbon 1981, 116–32 and Appendix 3; 90 (a twelfth-century glossed
MS, Vercelli, Bibl. Capitolare CXCIX [35]), should be added to the list.
19. The MSS containing the Boethius commentaries only break down as fol-
lows: – InP1: 1; InP2: 3; InC: 7; InH1: 1; InH2: 9; InP1 and 2: 2; InH1 and 2: 1;
and one each for InP1+InC; InP2+InH2; InP2+InC; InP1 and 2+InC;
InP2+InC+InH2.
20. The texts commented on are present only as short lemmata (underlined or else
identified by paraphs).
21. See Green-Pedersen 1984, 419–27, for commentaries on Boethius’ De diffe-
rentiis topicis. B1 to B3 are commentaries dating back to before 1100, B4 to B12
date back to the first half of the twelfth century, and B13 to B17 to the second
half of the century. All are transmitted by twelfth-century manuscripts, with
the exception of B2 and B3 (their manuscripts are dated XI/XII cent.) and B6
(XII/XIII cent.). Marenbon 1993, 98–122, contains a ‘Working Catalogue of
Commentaries on the Isagoge, Categories and De interpretatione from c. 875 to
c. 1150’. A ‘Supplement’, which updates information and extends to commen-
taries up to 1200, was published in Marenbon 2000. The catalogue of
Categories commentaries was then revised again in Marenbon 2013a.
 Logic 237
Ebbesen 1993, 148–73, contains a ‘List of Latin Commentaries on the
Sophistici elenchi and Treatises De fallaciis, c. 1125–1300’. Unpublished cata-
logues of commentaries on De divisione, De syllogismis categoricis, and De
syllogismis hypotheticis have been drawn by Yukio Iwakuma.
22. See Marenbon 1993, 85–92; Marenbon 1997, 31–2; Marenbon 2013a, 141–3.
23. For instance, the Supplementa Notularum super Topica Boethii, in Orleans,
Bibl. mun. 266, pp. 194b–204b (Green-Pedersen 1974 and 1977).
24. See Hunt 1948; Iwakuma 1993, 1999, 94; 2008, 50–1; Marenbon 1997, 92–3;
Mews 2005, 97–8.
25. P3 (f. 215ra: ‘Rabanus super Por.’, note added in the upper margin) and H11 (f.
225rb: ‘Rabanus super Terencium’) are attributed to Rabanus in BnF lat. 13368,
but the attribution is deemed unacceptable. C27 is attributed to a Ros.,
probably Roscelin, in the only manuscript Milan, Bibl. Cap. Ambros. M2,
f. 1r (‘Incipiunt Ros. Glossulae categoricarum, quae auree gemme uocantur ’);
again, the attribution is questionable (see Marenbon 2013, 150).
An attribution to Joscelin of Soissons is found in Orleans, Bibl. mun. 266,
p. 149a, rubric: ‘Notule de diuisionibus secundum mag. Gosl.’ (D2). ‘Glose super
Porphirium a magistro W. collecte’ in Erfurt, Universitätsbibl. Dep. Erf. Cod.
Amplon. 8° 5, f. 1r (P13) may refer to the act of collecting, rather than
authoring, the glosses.
26. Each Logica ‘Ingredientibus’ commentary is attributed to Abelard: P10 in Milan,
Bibl. Ambrosiana M 63 sup., f. 15vb, rubric: ‘Pretri (!) Abaelardi Palatini edicio
super Porphirium explicit’ (attribution repeated by a different hand on f. 1r,
upper margin: ‘Incipiunt G<lose> secundum magistrum Petr<um>
Abaelard<um> super Porphirium’); C10 in Milan, Bibl. Ambrosiana M 63
sup., f. 43vb (copyist’s hand): ‘Expliciunt G<losse> M<agistri> Petri Abaelardi
super predicamenta Aristotelis’ (as in P10, repeated by a different hand on f. 16r,
upper margin, but some letters have now been trimmed away: ‘Incipiunt
G<lose> m<agistri> p<etri> a<baelard>i super predicamenta Aristotelis’); H8 in
Milan, Bibl. Ambrosiana M 63 sup., in a hand different to the copyist’s both on
f. 44r, upper margin (‘Incipiunt G<losae> secundum m<agistrum> Petrum
Abaelardum super librum peryarmenias’) and f. 72r, right margin (‘Expliciunt
G<lose> m<agistri> Petri Abaelardi super librum peryarm<enias>’); B12 in BnF
lat. 7493, f. 168r, rubric: Petri Abailardi super Topica glossae incipiuntur felici
<h>omine (also repeated in the upper margin of the same page by a later hand).
27. P12 (Logica ‘Nostrorum Petitioni Sociorum’), now considered an Abelardian
commentary rather than Abelard’s, is attributed to him in the only manu-
script (Lunel, Bibl. mun. 6, f. 8ra, rubric: Incipiunt Glosule magistri P. Baelardi
super [a.m. Porphirium]). H4, D7, and P5 are all attributed to Abelard in their
manuscript, BnF lat. 13368, by a hand different from the copyist’s (f. 128r,
upper margin, partially damaged: ‘Petri Abae. < . . . > Summi Peripatetici edi
< . . . > Perihermenias»; f. 146, upper margin: ‘Petri Abaelardi iuniori<s>
Palatini summi peripatetici de divisionibus incipit’; f. 156ra: ‘Petri Abaelardi
I. P. S. P. editio super Porphyrium’), but the case for dis-attribution has been
made (Martin; Cameron).
238 john marenbon and caterina tarlazzi
28. See the catalogues of commentaries mentioned in n. 21: B-commentaries are
listed by Niels Green-Pedersen; P- C- and H-commentaries by John
Marenbon; SE-commentaries by Sten Ebbesen, and D- SC- SH-
commentaries by Yukio Iwakuma.
29. Green-Pedersen 1984, 418–27. One of the two copies of B12 is, in fact, a one-
page fragment and, interestingly, it follows one of the three copies of B8.
30. See later in this chapter and Jacobi 2011.
31. Poirel 2011.
32. Iwakuma 1999, 101–22.
33. To mark the similarity existing among these texts, Yukio Iwakuma calls C7,
C14, and the four C8 manuscripts the ‘C8-complex’: see Iwakuma, 2003,
2009, 89–91, 2003; Marenbon 2013a, 146–8, 159–60.
34. See Rosier-Catach 2011, xii–xvi.
35. And, as mentioned, in another version in BnF lat. 17813 (= C7).
36. In addition, Paris, Bibl. de l’Arsenal 910 transmits a fragment of B12,
a commentary also found in BnF lat. 7493.
37. For instance, the Abelardian Logica ‘Nostrorum Petitioni Sociorum’ (P12) and
commentary SC1 on Boethius’ De syllogismis categoricis are written by the same
hand in Lunel, Bibl. mun. 6. Though small, this manuscript, probably from
southern France, has wide margins, one gold-decorated initial, smaller red
and yellow initials, and rubrics. BnF lat. 17813 is also a homogeneous manu-
script. Written in rather elegant handwriting, the text is widely spaced and
there are a few phyto- or zoomorphic initials (Figure 12.2). Padua, Bibl.
Universitaria 2087 lacks decoration of any kind, but it is homogeneous and
its handwriting is clear and professional. In general, homogeneous manu-
scripts tend to be of higher quality than composite manuscripts.
38. See also Ebbesen 1993, 145.
39. To give just one example, f. 71r in BAV Reg. lat. 230 measures 148 mm ×
228 mm and is written in two columns of approximately 53 mm × 19 mm
containing sixty lines each. This means that almost three lines of text are
crammed into a height of one centimetre. Thanks to abbreviations and the
size of the handwriting, the page contains almost 1,800 words. Similar
measurements are found throughout the whole logical section of this compo-
site manuscript, ff. 41r–87v.
40. See Gumbert 1989, 1999, 2004; Munk Olsen 1998; Kwakkel 2002, 2012.
41. See Catalogue général, 398–405.
42. A detailed description is found in Minio-Paluello 1958, xli–xlvi.
43. I owe this information to an unpublished description by Yukio Iwakuma.
44. See Iwakuma 2008, 45–7.
45. On booklets, see Robinson 1980; Hanna 1986; Gillespie 2011.
46. See Kwakkel 2012, 72.
47. A passage in the Vita prima Gosvini throws light on the way in which
commentaries were copied within twelfth-century schools. It mentions
a commentary on Priscian’s Institutiones (probably the so-called Glosulae in
Priscianum), written by a famous grammarian of the time and sought by
 Logic 239
everybody, so that master Azo gave his student Goswin a quaternion at a time
for the copy (the task was then actually accomplished by Goswin’s brother,
a quicker copyist): see Grondeux 2009, 886–90.
48. As suggested by Rosier-Catach 2011, xxiii–xxiv.
49. Marenbon 2013, 151–2. Views (rather than writings, which can be the product
of more than one person) are attributed to a certain magister.
50. See Mews 1985, 92–3, 127–8; possibly De intellectibus is a surviving part of it:
cf. Marenbon 1997, 50–1.
51. See John of Salisbury, Metalogicon III. 4 (115: 10–116: 23); cf. De Rijk 1967,
167–70, who cites this passage.
52. These are the Introductiones dialecticae, attributed in their explicit to
‘Wilgelmum’, in ÖNB lat. 2499 and the Introductiones dialecticae artis secun-
dum magistrum G. Paganellum in El Escorial, Real Bibl. E. IV. 24 (ed.
Iwakuma 1993; on whether they are the work of William of Champeaux, see
Jacobi 2011, 263–7).
53. See n. 42 above.
54. See Minio-Paluello 1958, 1. xiii–xviii.
55. Occasional passages in texts from the time give a glimpse of what happened,
such as that cited in n. 47.
 chapter 13

The Classical Revival

Irene O’Daly

In characterising the Long Twelfth Century as a period of ‘renaissance’,

Charles Homer Haskins placed considerable emphasis on the role played
by the revival of classical texts, a revival that he argued ‘stood for
a harmonious and balanced type of culture in which literature and logic
both had their place, but which was hostile to the professional and
technical spirit that triumphed in the new universities’.1 As Haskins
admitted and others have asserted further, this revival of classical learning
was hardly unique to the late eleventh and twelfth centuries, and was
preceded by another period of interest in Latin texts – what has been
termed the ‘Carolingian Renaissance’.2 However, the relative impact of
the twelfth-century revival is enhanced due to the number of manuscripts
that has survived, high in comparison with those of previous centuries.3
While Haskins tended to romanticise the extent to which medieval
scholars such as John of Salisbury (ca. 1120–80) could be seen as epito-
mising a ‘renaissance’ in learning, the increase in the number of surviving
Latin classical texts from the period speaks of a growing interest in their
content.4 The Latin classics are the focus of this treatment, as they were
the nexus of classical study for much of the period and provide the best
basis for longitudinal comparison. They offered a major resource for
the learning of Latin, both at the level of basic pedagogy and in the
attainment of advanced argumentative and expressive skills, such as the
cultivation of the ars dictaminis, useful for medieval students who were
increasingly destined for administrative careers in religious and secular
chanceries.
Birger Munk Olsen has attested that the ‘classics were produced, like
all other books, according to the nature of the text and the destination of
the manuscript’; that is, that the factors that dictate how a copy of
a classical text was made were the same as those that affect any other
book.5 A twelfth-century classical manuscript should look much the
same as any other manuscript of the period, by this reading. However,
240
 The Classical Revival 241
the very features that Munk Olsen pointed out, while universal aspects of
manuscript production, are also those that serve to make every manu-
script an individual artefact; as Zetzel asserts, ‘how a book is copied is
directly related to why it is copied’.6 To examine, then, what the classics
looked like in the Middle Ages, we must turn to the manuscripts them-
selves and understand them within their immediate, ever-changing con-
texts of production and use, and how these affected their material
makeup. Furthermore, we must recognise that the classics were textually
flexible in this period; they were pillaged for excerpt collections, accreted
glosses and other paratextual accompaniments, and even became the
source of new texts in the form of commentaries. This chapter first
interrogates the content of the classical curriculum in the Long Twelfth
Century, before considering three dynamics of interaction with texts –
compilation, accretion and commentary – with a view towards investi-
gating how the reception of the classics was shaped by the form and
function of the manuscripts that were their carriers.

The Classics: Texts and Contexts

As Munk Olsen’s survey of the surviving Latin classical manuscripts of
the Middle Ages demonstrates, texts varied in their popularity, with some
manuscripts surviving in many copies, and some in few.7 In part, this was
influenced by the role of texts in teaching. Teaching of the liberal arts
depended, at least in its most idealised incarnation, on the knowledge of
a range of texts, many of which were classical. Thus, the Ars lectoria of
Aimeric of Angoulême, written in 1086, compares classical writers to
types of metal – gold, silver and lead or tin. The ‘golden’ authors (‘apud
Gentiles sunt libri autentici, hoc est aurei’) are Terence, Virgil, Horace,
Ovid, Sallust, Lucan, Statius, Juvenal and Persius. Those classed as
‘silver’ (‘in subgradiuo genere, hoc est argenteo’) are Plautus, Ennius,
Cicero, Varro, Boethius, Donatus, Priscian, Servius and Plato translatus –
that is, the Timaeus as translated by Calcidius. Aimeric comments that
were this work in its original language it would count as ‘gold’, but that it
is degraded by being in translation. ‘Catunculus’ (author of the Disticha
Catonis) and ‘Homerulus’ (author of the Ilias Latina) are part of the
‘communi genere’ or common stock (described elsewhere as of ‘lead’ or of
‘tin’), along with the Elegies of Maximianus, and the writings of Avianus
and Aesop.8 Implicit in Aimeric’s division is a sense of progression, from
the basic texts for learning Latin, such as the Disticha Catonis –
a collection of moral and philosophical maxims dating from the third
242 irene o’daly
century – through texts used for the teaching of grammar (Donatus,
Priscian), and rhetoric (Cicero), up to the historians and poets. Aimeric’s
list may be compared to that found in Conrad of Hirsau’s Dialogus super
auctores, which distinguishes between auctores maiores and minores,
recommending students to progress through the latter (starting with
Donatus) before studying the former, as children are weaned from milk
on to solid food.9 The auctores maiores listed by the magister in this
dialogue include those he explicitly describes as ‘Roman’, such as
Arator, Prudentius, Cicero, Sallust, Boethius, Lucan, Virgil and
Horace, but also Ovid, Terence, Juvenal, Persius, Statius and the
Latinised Homer.10
While idealised, certain elements of Aimeric’s curriculum merit atten-
tion. First, it is interesting that authors we may regard as post-classical,
such as Boethius and Priscian, count among Aimeric’s list, and Prudentius
and Arator among Conrad’s; in so doing they point to the breadth of time
that the medieval period regarded as ‘classical’. Secondly, the authors that
Aimeric refers to as ‘golden’ are among those whose works survive in the
greatest number of copies from the eleventh and twelfth centuries.11
The rhetorical works, Cicero’s De inventione and the pseudo-Ciceronian
Rhetorica ad Herennium, although ‘silver’ in Aimeric’s classification, were
the most popular classical works in the medieval period, surviving in more
than 120 copies apiece. Thirdly, the organisation of Aimeric’s list seems to
reflect the way in which some texts circulated in manuscript form; thus the
authors described as ‘tin’ often circulate together in loose compilations
under the title Liber Catonianus, a grouping that would become further
popularised (and augmented by the addition of Statius’ Achilleis and
Claudian’s De raptu Proserpinae) as the ‘sex auctores’ during the course of
the twelfth century.12
Finally, it should be noted that Aimeric’s list of classical works appears
alongside another list that presents a fourfold division of religious works,
again using the metallurgical conceit. Canonical books of the Bible and
liturgical works are ‘gold’; ‘silver’ includes the writings of Ambrose,
Augustine, Jerome and other patristic fathers; ‘tin’ represents authors
such as Bede, Prudentius and Arator, while the ‘lead’ category is used for
Lives of saints and martyrs.13 It must be recollected, therefore, that the
study of classics and the liberal arts in the Middle Ages did not take place in
isolation, but in conjunction with a comprehensive curriculum of biblical
and exegetical study. Thus, in a manuscript of the Bible owned by the
abbey of St-Thierry at Reims (Reims, Bibl. mun. 23, f. 25r) dating from the
early twelfth century, we find ‘Phylosophya’ depicted in the initial ‘O’ of
 The Classical Revival 243
the opening of Ecclesiasticus: ‘Omnis sapientia a Domino Deo est: et cum illo
fuit semper, et est ante aevum.’ Personified as a woman enthroned with
a halo, she is surrounded by three other figures depicted in half roundels.
‘Phisica’ holds grammatica, geometria, musica and astronomia in her hands;
‘Logyca’ carries rethorica and dyalectica [sic], while Ethica has the four
classical virtues of iusticia, temperantia, fortitudo and prudentia emblazoned
on her chest. This division, conflating sapientia with philosophia, departs
from the traditional Platonic–Stoic division described by Isidore by sub-
stituting grammar for arithmetic under the category of physics, and in so
doing incorporates the seven liberal arts into a Christianised philosophical
scheme. The inclusion of such an initial in a bible illustrates how the study
of the trivial and quadrivial arts was literally embedded in a Christian
curriculum. The classics informed the teaching of the liberal arts, but were,
in turn, mediated by Christian perspectives and sources.
Idealised accounts of curricula, of the type found in the writings of
Aimeric and Conrad, provide an insight into the kind of classical texts that
were studied in the medieval period, but not into how they were encoun-
tered in a material sense by medieval scholars. At the start of his discussion
of the liberal arts in Sacerdos ad altare accessurus, Alexander Neckam
(1157–1217), teacher at the schools in Oxford and Paris, comments:
‘A student who is to be educated in the liberal arts should carry a wax
tablet on which anything noteworthy may be written.’14 A further refer-
ence to note-taking in the classroom may be found in Peter of Blois’
letter to the archdeacon of Nantes regarding the study practices of his
two nephews. Peter bemoans the fact that one of the nephews, William,
‘learns dialectic not in books, as is the custom, but from scraps of
parchment and quires [in schedulis et quaternis]’.15 These references to
tablets, notes and quires suggest that students infrequently owned their
copies of classical (or other) texts, but learned from those of the teacher,
or from copies belonging to the teaching institution of which they were
part.16 This is the interpretation suggested too by the glosses that such
texts accumulated. Although some glosses consisted of excerpts from
commentaries, or forwarded new textual interpretations, many simply
provided explanations of word meanings and syntax; their relatively facile
content implies that one of their primary functions was to provide
reference material for teaching use in the classroom.17 We cannot assume
by default that every classical manuscript, glossed or otherwise, was
necessarily used in the classroom, however, but must interrogate the
codicological and textual features of specific copies of a text to determine
its destination and perhaps multiple stages of use.
244 irene o’daly
Codicological Considerations
Munk Olsen’s exhaustive survey of surviving copies of classical texts from the
ninth to twelfth centuries threw up Sallust’s Historia some distinctive
codicological peculiarities of such manuscripts. One notable aspect is the
fact that despite the general move towards the presentation of manuscripts in
two columns over the course of the twelfth century, classical prose manu-
scripts still tended to be produced in ‘long lines’ – lines that extend across the
width of the main textblock.18 Two factors seem particularly significant – the
copying tradition of the text and the size of the manuscript – although others
come into play, such as place of origin (notably house style).19 No single
author was copied exclusively in two columns, nor in long lines, although it
is notable that the majority of manuscripts of Pliny’s Naturalis historia
(usually in large format) were copied in two or even three columns, while
only one manuscript of Sallust’s Jugurtha and Catilines from the twelfth
century (CUL Ii. 6. 20) is in two columns. This would seem to suggest that
certain texts or authors were associated with particular layout traditions, an
association reinforced by imitative copying. The antiquity of the text could
be reinforced by an adherence to its presentation in its oldest copies.20
Oxford, Corpus Christi Coll. 57, a glossed copy of the Satires of Juvenal
and Persius copied by a scribe associated with Bury St Edmunds in the early
twelfth century, measures 385 mm × 300 mm – almost square – imitative of
the format preferred in the late antique period.21 In fact, imitative copying
may also explain copies with a mise-en-page that departs from tradition, as is
apparent from the manuscript tradition of Caesar’s Wars. Although the text
was usually copied in long lines, four extant copies are in two columns.
Of these, BnF lat. 5764 (s. xii2) and ÖNB 95 (s. xii1) share common textual
traits indicating that they were both copied from the same exemplar, now
lost.22 Although we cannot be certain of the format of the original lost
exemplar of these manuscripts, the likelihood is that it too was in two
columns, so the existing copies of it demonstrate how layout decisions persist
from manuscript to manuscript; it was easier for a scribe to emulate mise-en-
page, rather than innovate it.
Size was, no doubt, a further influential factor. A work as long as
Pliny’s Naturalis historia had to be copied on pages of large dimensions to
permit the text to be presented in a single volume; the largest exemplar
from the twelfth century (Luxembourg, Bibl. nat. 138, provenance Orval)
measures 520 mm × 350 mm, with 179 leaves. Large pages seem to have
encouraged the use of a multi-column layout, presumably to prevent lines
from becoming too long for the perceptual span of the reader, and to
 The Classical Revival 245
facilitate vertical ocular motion of the reader down the page. Thierry of
Chartres’ Heptateuchon, a compendium of the liberal arts (Chartres, Bibl.
mun. 497–8, completed ca. 1150, destroyed in 1944), and Wibald of
Corvey’s collection of Ciceronian works (Berlin, Staatsbibl. lat. fol. 252,
ca. 1150) were also presented in double columns on account of their size.
The Heptateuchon is recorded as measuring 525 mm × 365 mm, with its
two volumes originally containing some 1,400 leaves, while Wibald’s
volume measures 485 mm × 329 mm.23 On the other hand, some books
were made deliberately narrow, copied in a tall and skinny format, now
termed ‘holster’ or ‘register’.24 This format, particularly popular for texts
in verse, by its nature precluded the presence of more than one main
column of text. A typical example of this type is BAV Pal. lat. 1685,
a French twelfth-century copy of Lucan’s Pharsalia, measuring 227 mm ×
110 mm, giving a width/height ratio of 0.48 – particularly skinny, given
that the average width/height ratio of books in this period is 0.67–0.72.25
It is likely that scansion was a factor in the selection of this format for
verse texts.26 Munk Olsen notes that in five manuscripts of Horace, the
lyric poems are copied in two columns, while the hexametric poems are in
one, to respond to different verse lengths.27 While two-column verse
manuscripts exist, they are rare; in the twelfth century 98 per cent of
Lucan manuscripts are in one column, as are 87 per cent of Ovidian
manuscripts and 88 per cent of Statius manuscripts.28 Verse texts could
have been copied in two columns on broader leaves, but the facility of the
single column for highlighting the rhyming structure seems to have led to
a preference for a narrow format, a preference that perhaps later became
fashion.

The Dynamics of Compilation

These brief observations on the codicological presentation of the classics in
the Long Twelfth Century only touch on the diversity of ways in which
text, genre and layout could interact. The heterogeneity of the texts in
question (varying in length and form), changing fashions in manuscript
production, and diverse contexts of use affected the appearance of classical
books in this period. To speak in general terms of the ‘classics’ belies the
variety that comprised the corpus, a corpus that includes prose, poetry,
instructional texts, epistolary collections and drama. Composite manu-
scripts of this period, those that contain collections of texts by single or
multiple authors, demonstrate how twelfth-century scholars sought to
make sense of such heterogeneity by pulling together texts on similar
246 irene o’daly
subjects or in similar textual forms. Compilation is an interpretative act,
and affects the way in which classical texts were read and transmitted.
At one extreme, the motivation to combine texts resulted in large-scale
volumes like the Heptateuchon, compiled with a view towards presenting
a comprehensive account of the liberal arts – ‘in unum corpus voluminis’, as
the prologue comments – underpinned by a commitment to the unity of
the trivium and quadrivium.29
At the other extreme, we find florilegia, selections of excerpts culled
from classical texts, arranged by author or by theme, and deliberately
designed to synthesise classical learning.30 The copy of the prose sections
of the Florilegium Gallicum found in Paris, Bibl. de l’Arsenal 711, ff.
182–243, a manuscript associated with the School of St-Victor, exemplifies
this tradition.31 It arranges its material by author and by text, conveying
these divisions to the reader by the use of running titles across the top of the
page, and further distinguishing the opening of each excerpt by the use of
alternating red and blue penwork initials. The text is also apportioned by
rubrics referencing the source of the excerpt and, on occasion, its theme.
The sections excerpting parts of Cicero’s De inventione, for example,
contain short definitions of eloquence taken from the prologues of books
I and II of the text, followed by definitions of the cardinal virtues and their
subsidiary types, each individually rubricated (f. 190rv), with titles such as
‘Quid sit uirtus’ and ‘Quid sit fortitudo’. The purpose of such collections
was to isolate aphorisms, whether for contemplation or for educative
purposes, and they proved popular even for experienced scholars like
John of Salisbury.32 The Florilegium Angelicum refers to such accessibility
in its preface, commenting that it can always be kept at hand [‘ut semper ad
manum habeas’].33 Collections of texts like the Florilegium Gallicum and
the Florilegium Angelicum demonstrate an expedient approach to classical
learning, while the popularity of such collections can be measured by the
fact that some seventy-five florilegia manuscripts survive from the twelfth
century or earlier.34
Compilations give an insight into the way in which an author was read
or perceived in the Middle Ages. A case in point is Seneca the Younger,
a classical author who was treated with particular regard in the twelfth
century due to his alleged association with St Paul. Jerome’s Vita Senecae
(De viris illustribus, XII), often included among the preambles to collec-
tions of Senecan writings, referred to the ‘letters between him and Paul
which are read by many’, contributing to the medieval image of Seneca as
a saintly sage.35 The apocryphal correspondence between Seneca and Paul,
consisting of fourteen short letters, circulated from the fourth century on.
 The Classical Revival 247
Two other works were also commonly attributed to Seneca: Formula vitae
honestae by the sixth-century bishop Martin of Braga, which circulated as
De quattuor virtutibus in the Middle Ages, and De remediis fortuitorum,
a short dialogue between Sensus and Ratio. It is likely that both of these
works were extracted from genuine, now lost works of Seneca, but as
Reynolds memorably wrote, ‘Seneca’s fame grew fat on works which he
had never written.’36 Other patterns of transmission emerged within the
corpus; De clementia and De beneficiis frequently circulated hand in hand,
reflecting their common themes of generosity and kindness and perhaps
also reflecting their value as ‘mirrors for princes’. The pairing of these texts
may also result from the form of their ninth-century archetype, BAV Pal.
lat. 1547, which preserves both texts together.37
Approximately forty single-volume compilations of Seneca’s writings
survive from the twelfth century, indicative of a desire to collect his known
works.38 A typical example of such a single-author compilation is Yale,
Beinecke Libr. Marston 45, a manuscript from the Cistercian monastery of
Igny, near Reims, dating from the third quarter of the twelfth century – one
of four such compilations from the Cistercian milieu.39 It is written in two
columns, a layout typical of Cistercian manuscripts of the period, with plain
decoration of alternating red, green and blue initials. The manuscript con-
tains Jerome’s Vita Senecae (f. 1va), the pseudo-Senecan letters to Paul
(ff. 1va–2vb), the Epitaphium Senecae (f. 3ra), the first eighty-five Epistulae
morales (ff. 3ra–79rb), excerpts from De beneficiis (ff. 79va–99vb) and De
clementia (ff. 100ra–103rb), Formula vitae honestae (ff. 103rb–105va), De
remediis fortuitorum (ff. 105va–107va) and a set of sentences attributed to
Publius Syrus and Seneca (f. 107vb). The final leaves (ff. 107vb–109rb)
contain excerpta from Claudian and De tribus dicendi generibus by William
of St-Thierry, although these texts are written in a distinct, slightly later hand
than the rest of the manuscript, and are not mentioned in the contemporary
contents list found on f. 1r. This collection of 110 leaves, measuring 294 mm
× 204 mm, demonstrates how thematic association, traditions of attribution
and alleged authorial repute motivated compilatory practices, and in turn,
embedded Seneca’s works within a particular textual context, and an increas-
ingly popular manuscript form – the collection.
A final influential form of compilation is the association of works that
are linked by subject. One example from the field of rhetoric demonstrates
this trend in action. Cicero’s De inventione and the pseudo-Ciceronian
Rhetorica ad Herennium frequently circulated together, and were com-
bined with a variety of other texts, including contemporary and late
antique commentaries. A common companion item was the fourth book
248 irene o’daly
of Boethius’ De differentiis topicis, found in seventeen manuscripts dating
from the twelfth century containing both rhetorical texts attributed to
Cicero. York Minster XVI. M. 6, produced ca. 1150 and associated with St
Mary-in-the-Fields, Leicester, contains De inventione and Rhetorica ad
Herennium, and concludes with the fourth book of De differentiis topicis,
written over seven leaves (ff. 113r–120r). In Bamberg, Staatsbibl. Class. 21,
a twelfth-century French manuscript containing De inventione and
Rhetorica ad Herennium – copied in independent codicological units but
designed to be used together – the excerpt from De differentiis topicis
appears at the conclusion of the part of the manuscript containing De
inventione (ff. 71r–80r). In other cases where De inventione and Rhetorica ad
Herennium have been copied as a single production unit, the extract from
De differentiis topicis often appears sandwiched between the two texts. For
example, in Leiden, Universiteitsbibl. VLQ 103, a French manuscript from
the second half of the century, the text is found on ff. 46r–52v, and in BL
Harley 3509, another manuscript from the second half of the century from
France or England, the text appears on ff. 44r–48v. As is clear from the
details of the foliation of these manuscripts, the excerpt is relatively short,
less than twenty pages long in its most recent translation.40 This was of
major practical advantage for copyists; it could be accommodated easily
alongside copies of De inventione and Rhetorica ad Herennium, without
increasing the size of the manuscript as a whole by more than a single extra
quire, at most. While the subject matter is the principal feature dictating
the inclusion of this late antique text alongside the rhetorical works
attributed to Cicero, its length was also a contributing factor that facili-
tated its excerption and circulation.

The Dynamics of Accretion: Textual and Visual Additions

Compilation affected the kind of works that classical texts were read
alongside. However, the texts themselves were also open to change in
this period. Classical texts accumulated a variety of paratextual additions,
such as glosses, introductions (accessus) and illustrated elements. This
flexibility must be borne in mind when considering the materiality of
such texts in the Middle Ages; any treatment must take into account
how the format of a text was adapted to incorporate such additions.
Lucan’s Pharsalia serves as a case in point. One of the most popular texts
in the Middle Ages, it survives, according to Munk Olsen’s estimation, in
more than thirty copies from the eleventh century, and in more than 100
copies from the twelfth century.41 Leiden, Universiteitsbibl. BUR Q 1,
 The Classical Revival 249

Figure 13.1 Lucan’s Pharsalia, Egmond, just after 1050, with later additions and
glosses. Leiden, Universiteitsbibliotheek, BUR Q 1, ff. 1v–2r.

a copy of the text dating from early in the second half of the eleventh
century (provenance Egmond) contains glosses and additions made over
the following 100 years which demonstrate how these accretive processes
worked, and how the manuscript itself was shaped by, and accommodated
in turn, such an apparatus (Figure 13.1).
The text itself is written in a single column; the first letter of each line is
rubricated, while the opening lines of the text on f. 2r are written in
alternating letters of black and red. The book is narrow, measuring
276 mm × 162 mm, giving a width/height ratio of 0.58. Glosses appear
around the text in all four margins written in a small script, at the rate of
approximately two lines of gloss for each one of the main text. The glosses,
linked to the text with a system of alphabetic and symbolic signes-de-renvoi,
250 irene o’daly
are added in two layers by two distinct contemporaneous hands; the second
glossing hand has a more rounded aspect than the first, and is in a lighter
ink. This second hand is often forced to work around the glosses provided
by the first glossator, as seen on f. 7r, where the second layer of gloss is
enclosed in a sinuous curved line to distinguish it from the first.
On occasion, both hands gloss the same word or passage in different
ways, further demonstrating that they represent two stages of reflection
on the text. One function of the glosses is to explain terms that would have
been unfamiliar to the reader. So, the mention of the dwellers by the river
Araxes [I, 19] on f. 1r is glossed ‘de arabia’; the reference to Iulia on f. 4r [I,
113] is glossed ‘uxor pompeii, filia cęsaris’. Another function is to offer
synonyms for words used or syntactical aid – perhaps to facilitate classroom
use. So on f. 6v the lines ‘stridor lituum clangorque tubarum | non pia
concinuit cum rauco classica cornu’ [I, 237–8] are glossed in two fashions,
with ‘non pia’ explained in an interlinear gloss as ‘impia’, while a tie mark
links it to the word ‘classica’ to demonstrate that these words are linked – it
is an ‘impious fanfare’. A third function is to give additional information
about the intention or content of the text; thus on f. 1r, the first glossing
hand has provided in the upper margin a brief note on the difference
between civil war [ciuile bellum] and war that is ‘more than civil’ [plusquam
ciuile] – these definitions are taken from Isidore’s Etymologiae [XVIII. 2–4],
where the same examples of each type are given. The second glossing hand
has also added a note in the inner margin claiming that the first seven lines
of the poem are by Seneca, expressing a common belief. The broad margins
and aerated space between the lines of text facilitate the insertion of
a variety of marginal and interlinear glosses, and were intended to be
used for this purpose.42
In addition to these interlinear and marginal notations, further addi-
tions are made by a number of hands on the flyleaf of the text and at its
conclusion. The flyleaf (f. 1rv), a singleton added to the first quaternion,
contains a number of texts intended to act as an accessus – a brief historical
or biographical prologue, or discussion of the intention of the author – to
the text.43 The main scribe of the text has added two short texts (f. 1r),
a note on the auguries of the conflict and the commonly circulated version
of Lucan’s epitaph.44 The presence of the hand of the main scribe on this
flyleaf demonstrates that it was a contemporary addition to the volume,
serving the dual purpose of protecting the illuminated opening initial and
consciously providing a space for additions. The third text on this page, the
account of Lucan’s life written by Suetonius, is in a hand that does not
appear elsewhere in the manuscript; also dating from the second half of the
 The Classical Revival 251
eleventh century, it exhibits some documentary features, such as distinctive
long descenders.45 On the verso, the two glossing hands have added further
texts. The first is a historical note in the hand of the first glossator.46
The second opens with the intentio of Lucan, but is followed immediately
by another intentio, that of Terence, with a brief list of characters (and their
characteristics) from Andria and Eunuchus.47 The hand that has added this,
the second glossing hand, also provides an extensive note at the end of the
manuscript, wrapping around the text on ff. 150v–151r, and continuing into
the (formerly) blank f. 151v. This is an extract from book VI of Orosius’
Adversus paganos. Taken as a whole, these annotations augment the reader’s
knowledge of Lucan by providing basic bibliographic information, but also
general contextual information on classical Rome. Furthermore, the addi-
tions demonstrate how a text like Pharsalia could become, in Christopher
Baswell’s words, a ‘pedagogical node’ around which the teaching of Latin,
as well as of classical culture and history, could take place.48 The text was
a potential repository of knowledge about the classical world that could be
expanded as glossators saw fit, and so offers material evidence of engage-
ment from the second half of the eleventh century on with the revivified
classical corpus.
As noted, one of the functions of the accessus is its role in situating a text
within its classical context. Another, albeit more unusual intervention that
permitted the ‘classicisation’ of the classical text for the medieval reader was
the addition of illustrations or historiated initials.49 Florence, Bibl.
Laurenz. Plut. 45. 2, a twelfth-century Italian miscellany containing
among other works the apocryphal letters of Paul and Seneca, Sallust’s
Bellum Catilinae and Bellum Iugurthinum, Cicero’s De senectute, De
amicitia and In Catilinam and some commentaries, contains three pen
and wash illustrations among the texts. The first, placed at the conclusion
of Bellum Catilinae, is a battle scene, with combatants dressed in chain
mail, their uplifted spears bisecting the concluding words of the text
(f. 25r). The next two are more explicitly referential of the classical context
of the works. At the beginning of De senectute (f. 57v) is an illustration of
Cicero and Cato seated beneath two porticos. Cicero, on the left, holds an
open book on which is written the opening words of De senectute [‘O Tite si
quid adiuuero’], while Cato, with a rounded back, rests his chin on his
hand, listening intently. Following the prologue to De amicitia (f. 70r),
another illustration depicts Scaevola, Laelius and Gaius Fannius in
dialogue in the garden, with Scaevola sitting ‘according to his custom on
a semi-circular garden bench’ [De amicitia, I. 2], while Cicero is sheltered
under a portico to one side, writing on a roll. The clothing and architecture
252 irene o’daly
are deliberately archaised, while both illustrations convey the immediacy of
Cicero’s compositional role to the medieval reader by depicting him in the
dynamic acts of recording or reading.
While these two illustrations bear a direct relationship to the texts that
follow, two images appended to companion volumes of De inventione and
Rhetorica ad Herennium, probably made in Poitiers in the mid-century,
now Bodl. Libr. Barlow 40 and Lucca, Bibl. Statale 1405, reference instead
what was known of Cicero as a historical figure. Barlow 40, f. 1r
(Figure 13.2), shows Cicero listening to the arguments of Cato and
Caesar on the fate of Catiline and his cohort; he holds a book on which
are written words from In Catilinam [‘Ad mortem Catilina duci’: I.1.2].
Meanwhile Lucca 1405 (f. 1r) depicts Cicero arguing with Sallust; Cicero
holds open the pseudo-Ciceronian Invective against Sallust, while Sallust
holds open the pseudo-Sallustian Invective against Cicero. As the rhetorical
manuals contain rules on argumentation, these images augment the con-
tent of the text by giving examples derived from a classical milieu of
argumentation-in-process, as well as enhancing the reader’s biographical
knowledge of Cicero in a fashion similar to the textual accessus.
Functioning as ‘visual prologues’, such illustrations demonstrate a desire
to make the authors and interlocutors of classical texts immediate to the
medieval audience. In their imaginative depictions of clothing and settings,
they evoke the classical world.50

The Dynamics of Commentary

We have already seen how classical manuscripts came to acquire glosses and
scholia during this period. Further evidence of the place of the classics in
the Long Twelfth Century is their capacity to inspire new texts. Cases in
point are commentaries on the two rhetorical works associated with
Cicero – De inventione and Rhetorica ad Herennium. The most popular
classical works in the twelfth century, they inspired the composition of new
texts, building on an antique tradition typified by the commentaries of
Victorinus and Grillius. John Ward has established that of forty-five
twelfth-century manuscripts containing commentaries on these texts,
only six contain copies of late antique works, with the rest being copies
of new compositions, so demonstrating an increased interest in medieval
interpretations.51 In some instances, text and commentary were produced
as companion pieces. Two manuscripts now in the library of York Minster,
X.VI. M. 6 and X.VI. M. 7, made ca. 1150, serve as an example. As already
described, one contains the texts of De inventione and Rhetorica ad
 The Classical Revival 253

Figure 13.2 Cicero listens to the arguments of Cato and Caesar on the fate of
Catiline and his cohort. Oxford, Bodleian Library, Barlow 40, f. 1r.
254 irene o’daly
Herennium, along with book IV of Boethius’ De differentiis topicis, while
the other contains a number of rhetorical commentaries, attributed to
Manegold of Lautenbach and William of Champeaux. These manu-
scripts were produced contemporaneously and have a shared provenance
(the Augustinian abbey of St Mary-in-the-Fields in Leicester).
The commentaries are distinguished from the main text by having
a different mise-en-page; they are in two columns, while the main text
is in one. The changing layout conveys the status of the commentary as
a new composition – textually and visually distinct from the main texts –
which were, per tradition, presented in long lines.
Other commentaries also convey a story through their format. Leiden,
Universiteitsbibl. BPL 189, contains an extract from the commentary by
Thierry of Chartres (written ca. 1140) on De inventione, copied in the late
twelfth century in France. Written in a cramped hand on a booklet of six
leaves, with more than fifty lines per 120 mm × 70 mm page, the selection
from the commentary (ff. 42r–45v, Commentarius super libros de inuentione
I. 1. 1–I. 5. 7) is followed by a brief extract from the text of De inventione itself
(ff. 46–47r, De inventione I. 1. 1–I. 1. 9). The parchment is of poor quality.
Folia 45–46 have holes that the scribe has written around, while f. 46 has
clearly been cut from the edge of a parchment sheet, with a curved lacuna in
the lower corner, the edge of which shows signs of stretching and has not
effectively taken the ink (notably on the verso). Despite the comparatively
unassuming character of the writing support, the text itself is carefully
executed. One penwork initial has been added on f. 42r, while spaces for
further, unexecuted initials, have been left on f. 46rv. It is likely that the copy
was made for personal use, as some marginal annotations suggest the inter-
action of the scribe with the text. Keywords in the hand of the scribe facilitate
navigation through the thicket of the dense text, and certain points within
the text are also emphasised through the addition of small marginal faces; in
two instances (ff. 44r, 45r) we find faces alongside discussions of ‘calliditas’,
the cunning that is the opposite of wisdom (Figure 8.2). In one of these,
f. 44r, the face is deliberately disapproving, with a downturned mouth,
drawing attention to the discussion in the text of the difference between
wisdom and cunning. In the margin of Cicero’s text itself (f. 46r), a further
face appears alongside Cicero’s musings on the aid that wisdom gives
eloquence in ending wars and facilitating good governance (De inventione
1.1). Considering the extract from the De inventione selected – the opening
prologue that deals with the origin of eloquence and the role of the ‘wise
man’ in the polity – the parts emphasised by these faces show a particular
concern with the morality of the text and, by extension, in the lessons the
 The Classical Revival 255
reader may expect to learn from it. The size, support and nature of the
textual interventions point to the status of this manuscript as an object for
personal consumption of the classics; it demonstrates one of the range of
material forms that classical scholarship took in this period.
Another manuscript of Thierry’s commentary, now BL Arundel 348,
provides a further sense of the context within which the classics were
read. Dating from the late twelfth century, and written in the north of
France, it is now incomplete (ff. 102r–179v, Commentarius super libros de
inuentione, I. 1. 1–II. 2. 170). Several small drawings have been added to
the text by one of its scribes. Most of them appear alongside catchwords
in the lower margins at the end of quires (see ff. 109v, 117v, 131v), although
several drawings do not correspond to the quire divisions. Some are
humorous riffs on the text; a drawing of a crocodile-like animal appears
alongside the part of the text that refers to beasts (bestiae) while a drawing
of a man’s face accompanies the catchwords ‘magni viri studuerunt’.52
Another drawing, found in the lower margin below the opening to
the second prologue to Book I of Thierry’s commentary (f. 128r), shows
a female figure holding a knife-like object in her outstretched hand,
accompanied by a small boy holding an open book. I suggest that this
figure is a personification of the art of Grammar, drawing on the icono-
graphic tradition of the fifth-century writer Martianus Capella, who
described Grammar in On the Marriage of Mercury and Philology as
a woman who bears a ‘pruning knife with a sharp point with which she
could prune the faults of pronunciation in children’.53 The prologue
refers to the state of teaching in the schools; Fame in league with Envy,
criticises Thierry: ‘She [Fame] allows him rhetoric or grammar, as if for
argument’s sake, in order to snatch away dialectic, allowing him anything
rather than dialectic. She alleges now his immoral life-style, now his
negligence in studying, now his long-winded interpretations. Finally,
when all else fails, she objects that he lectures to advanced students, so
that he holds the younger ones back, or rather, corrupts them in such
a way that they cannot make any progress.’54 The use of personification in
the prologue, which is clearly intended to satirise those who resent the
progress of logical teaching in the schools, seems to have prompted the
scribe to sketch this drawing of Grammar. Martianus’ text, which sur-
vives in more than 240 manuscripts from the ninth century on, was very
popular, and its classification of the sciences was particularly influential.
The iconographic attributes of the liberal arts, such as the knife held by
Grammar or the snake held by Dialectic, were parts of a visual vocabulary
that would have been highly accessible to the medieval reader, as apparent
256 irene o’daly
from their contemporaneous employment on the sculptures of the liberal
arts found on the Royal Portal of Chartres cathedral. While this illustra-
tion adds little to the interpretation of Thierry’s commentary, its pre-
sence in this manuscript brings together three important strands of the
history of the reception of the classical tradition in the Long Twelfth
Century, namely, the study of classical texts within the broader context of
the trivium and quadrivium, the reshaping of classical texts by contem-
porary commentators, such as Thierry of Chartres, and the influence on
classical reception of authors of the late antique and early Christian
periods, such as Martianus Capella.

Conclusion
Considering the heterogeneity of the classical corpus, it is a challenge, and
perhaps a mistake, to draw overarching conclusions regarding the history
of its production and reception in the Middle Ages. To do so would
simplify the evidence excessively, and obscure the variety of genres and
contexts within which such texts were read. The twelfth-century cases
examined throughout this chapter provide a number of snapshots of
moments of production and use, and in so doing demonstrate how
established practices, such as compilation, glossing and illustration, shaped
classical manuscripts for reception, both textually and physically. Although
the texts were old, new practices, such as the formation of large-scale
florilegia, the composition of commentary and accessus texts and the
formalisation of glossing traditions, renewed and reshaped their content.
The process appears to have been affected by practical concerns, by
a sensitivity to the demands of the genre of the text and by considerations
of the intellectual environment within which the text is used, whether
communal or individual. Looking in depth at a few manuscripts has shown
clearly that while patterns and trends exist, such as the desire to collect
works by author or theme or the pedagogical practices of annotating and
augmenting the text, each manuscript has an individual character, and
often a unique set of contents. Thus, when considering the nature of
classical scholarship in this period, we must not ask simply ‘what was
read?’ but also ‘in what form and format was it read?’ – whether a scholar
encountered a classical text whole or through excerpts, standing alone or
embedded in a textual context of commentaries and associated literature,
in a large manuscript or a small booklet. Only through such an interroga-
tion will the real character of the twelfth-century classical renaissance be
revealed.
 The Classical Revival 257
Notes
1. Haskins, Renaissance, 99.
2. Brown 1994; Contreni 1995.
3. Munk Olsen, Auteurs classiques; Reynolds 1983; Reynolds and Wilson 1991.
4. Jaeger 2003, 1181–3.
5. Munk Olsen 1996, 17.
6. Kwakkel 2015; Zetzel 2005, 153.
7. Munk Olsen 1996, 17; Munk Olsen, Auteurs classiques, 4/2. 33.
8. Reijnders 1972, 170.
9. Huygens 1970, 79.
10. Ibid., 95–122.
11. Munk Olsen, Auteurs classiques, 4/2. 33.
12. Hunt 1991, 66–70.
13. Reijnders 1972, 169–70.
14. Copeland and Sluiter 2009, 536.
15. Ep. 101: PL 207. 312.
16. See Chapter 5 of the present volume.
17. Reynolds 1996; Wieland 1999.
18. On the changing mise-en-page of manuscripts in this period, see Bozzolo and
Ornato 1983, 318–29; Derolez, Gothic, 37, 58; Ker, English MSS, 42;
Tahkokallio 2015, 143.
19. Munk Olsen 1995, 103–4.
20. Lowe 1925, 207.
21. Thomson 2011, 29–30. Thomson notes that Oxford, Merton Coll. 291,
Martianus Capella, is copied by the same hand in a similar square format.
22. Brown 1972, 33, 44–5. The two other copies in two columns (Leiden,
Universiteitsbibl. BPL 38D and BnF lat. 5783) are later copies of BnF lat. 5764.
23. Burnett 1984, 142.
24. Kwakkel 2015, 71–2; Munk Olsen, Auteurs classiques, 4/2. 158–9.
25. Bozzolo and Ornato 1983, 287–310.
26. Parkes, Pause and Effect, 97–114.
27. Munk Olsen 1996, 14.
28. Munk Olsen, Auteurs classiques, 4/2. 162.
29. Copeland and Sluiter 2009, 441.
30. Munk Olsen 1979, 1980; Rouse and Rouse 1979, 3–42.
31. http://gallica.bnf.fr/ark:/12148/btv1b52501503x
32. Martin 1984, 184–5.
33. Rouse and Rouse, Authentic Witnesses, 129.
34. Munk Olsen, Auteurs classiques, 2. 837–77.
35. Ker 2009, 187–91.
36. Reynolds 1965, 112.
37. Reynolds 1983, 363–4.
38. Munk Olsen 1996, 13.
258 irene o’daly
39. http://brbl-dl.library.yale.edu/vufind/Record/3592235; Munk Olsen 1995,
104–5.
40. Stump 1978, 79–95.
41. Munk Olsen 1982–2014, 4/2. 33.
42. Note the addition of neumes to VIII. 88–98 on f. 106r, again facilitated by the
wide line spacing.
43. Hunt 1948; Minnis 1988, 9–39; Quain 1945.
44. ‘Philosophus iste de hortatur romanos habe ciuliem discor/diam pluribus
modis quia multae et innumerabiles cedes et/alia infinita mala in de praece-
dunt et pluribus signis ante/conflictum et in ipso conflictu manifestatur
omnibus diis displicere’; ‘Corduba me genuit rapuit Nero prelia dixi . . .
haec vere sapiet dictio, quae feriet.’
45. ‘Prima ingenii experimenta in Neronis laudibus . . . diligenter sed inepte
quoque.’
46. ‘Mundus pene totus excepta Parthia et Gallia . . . exsaturatus spiritum
reddidit.’
47. ‘Intentio Lucani est consulere r.p. per dissuasionem ciuilis belli, ciuilem usum
legibus et moribus constitutum perturbantis. Intentio Terentii est per senum
seueritatem iuuenumque leuitatem . . . – auditorem cautum reddere’;
‘Terentius. terens exemplorum. rationibus. emulos. nociue . . . – . . . Thraso
tardus. humanis. rationibus. amore. se oblec<c>tat’ Cf. other examples of
such lists of characters in Munk Olsen, Auteurs classiques, 4/1. 111.
48. Baswell 1999, 136.
49. Munk Olsen, Auteurs classiques, 4/2. 175–205.
50. Sears 2002, 61.
51. Ward 2003, 178.
52. Thierry of Chartres, 68–9.
53. Martianus Capella, 65.
54. Dronke 1988, 363. Thierry of Chartres, 107–8.
 chapter 14

Reading the Sciences

Charles Burnett

This chapter examines common features of the codicology and mise-en-

page of Latin scientific manuscripts in the Middle Ages, paying particular
attention to features which appear to have been taken over from Arabic and
Greek models, and noting variations with respect to the readership of the
manuscripts (monastery, cathedral school, medical school, private library,
university). The manuscripts surveyed will range in subject matter from
mathematics (including astronomy and astrology) and medicine (includ-
ing veterinary science), Peripatetic natural science and metaphysics (the
embryonic Corpus vetustius), to alchemy and magic.

The Introduction of New Sciences

The first evidence for the introduction of new scientific texts into Western
Europe dates from the late tenth century, when texts on the astrolabe,
astrology and the abacus were first written in Latin. The texts on the
astrolabe consist of descriptions of both its construction and use. While
occasional passages can be identified in Arabic texts on the astrolabe, the
majority of the material in these earliest texts appears to be notes that were
meant to be read with a master, and above all, an astrolabe at hand.1 These
manuscripts, therefore, exhibit two notable features: 1) pages of Arabic
texts in transliteration with the equivalent Latin words written immedi-
ately above the Arabic words;2 2) accurate depictions of a real astrolabe and
its parts, even to the point of representing the Arabic alphanumerical
notation and the name of the Arabic maker.3 Most of the astrological
texts were devoted to a particular kind of astrology, of which the distin-
guishing features were the transference of the letters of the names of the
clients into Arabic or Hebrew letters, and the use of the twenty-eight lunar
mansions alongside or in place of the signs of the zodiac.4 The manuscripts
of these texts are distinguished by the presence of tables (of letter-number
equivalents, of Arabic and Hebrew equivalents for the letters of the Roman
259
260 charles burnett
alphabet) and of rotae – of the signs of the zodiac or the lunar mansions (or
both together) – often with the constellations depicted as asterisms.
The earliest abacus texts are associated with Gerbert d’Aurillac (d. 1003
as Pope Sylvester II). In this case, the centrepiece was the depiction of the
abacus itself, which is a board consisting of twenty-seven columns, one for
each decimal place, with columns grouped three by three, and Hindu-
Arabic numerals used to number the individual columns or each set of
three columns.5 A large manuscript format was necessary for depicting the
complete abacus table. In most cases reduced versions were substituted, but
one large sheet of parchment with the abacus markings on it written at the
monastery of Echternach ca. 1000 survives; this is likely to have been
a complete prepared sheep’s skin and is large enough for the apices or
counters to be disposed on it for carrying out calculations.6
Gerbert was well known for using instruments in his teaching of the
quadrivium. The abacus and astrolabe texts joined those on arithmetic,
geometry and astronomy – Boethius’ De institutione arithmetica and De
institutione musica, and the portion of Boethius’ translation of Euclid’s
Elements that had been incorporated into late classical texts on land-
measuring (those of the agrimensores or gromatici).7 Diagrams and instru-
ments were essential for all these texts: the depictions of numerical relations
in the De institutione arithmetica and of proportions in the De institutione
musica, the rhythmomachy board and its pieces for arithmetic, the chess
board and its pieces for geometry and the monochord for music.8 One can
imagine the scene of unbound manuscript pages with brief notes on
instrument construction and use, parchment being used for making instru-
ments (for instance the abacus table or volvelli for demonstrating astro-
logical and astronomical doctrine, and astrolabes) and spheres,
hemispheres and monochords occurring together in the classroom.
The copying, revising and refining of these texts on the quadrivium
continued into the eleventh century, in which Lotharingia (including Liège,
Trier and Echternach), Paris, Chartres, Orléans and Reichenau were parti-
cularly important centres. From Fulbert, bishop of Chartres (1006–28), we
have a set of notes on the stars depicted on the rete of an astrolabe, and
a poem which he composed using these notes.9 On the abacus sheet from
Echternach was added a life-size diagram of the monochord.10

The Scientific Manuscripts of the Long Twelfth Century

This was the scene at the opening of the Long Twelfth Century. From
a scientific point of view, it could be said to have begun when, for the first
 Reading the Sciences 261
time, a large number of Greek and Arabic works on the science of medicine
began to be translated in Salerno and Monte Cassino. Alfano, archbishop
of Salerno (1058–87), himself translated Nemesius’ On the Nature of Man
from Greek, but encouraged a doctor from Qairouan, Constantine the
African, to translate or adapt a large number of medical texts from Arabic
into Latin. The most important of these was the ‘Royal Book’ of ‘Alī ibn al-
‘Abbās al-Majūsī, which Constantine used as the basis for his Pantegni,
a work of ten books on the theory of medicine and ten on its practice. Since
Constantine became a monk of the Benedictine mother house of Monte
Cassino and collaborated with other monks of the monastery, he had the
service of a well-equipped scriptorium. Here scribes wrote in both the
Beneventan and the Caroline scripts, and illustrators were at hand to
ensure the beauty of the manuscripts. Moreover, the important position
of Monte Cassino within the Order, and its Europe-wide network, ensured
that manuscripts produced there could quickly reach other areas of
Europe.11
None of the Arabic manuscripts that Constantine brought with him
from Qairouan has been identified, and the only surviving manuscripts of
his translations that might possibly have been made by or under the
supervision of Constantine himself are Monte Cassino 225 and
The Hague, Koningklijke Bibl. 73 J 6.12 From the similarity in format of
many of the early Latin copies of the works, and from extant Arabic
examples, one can only hazard a guess at how the Arabic Vorlage could
have influenced the mise-en-page of these Latin medical manuscripts.
In the case of the Pantegni, it can be observed that the numbered chapter
titles are given at the beginning of each book, and these numbers are
included within the books themselves, to enable the reader to find his/
her way round the work with facility. While this is not unprecedented in
the Latin tradition, it is a conspicuous feature of Arabic scientific texts, in
which the numbering was usually written out in full (avoiding the ambi-
guity that could arise from the use of numerical symbols). The medical
texts translated or adapted by Constantine did not include diagrams, nor
did their comprehension require the presence of a teacher; rather, the
transmission of this knowledge was strictly textual. This means that textual
accuracy and clear articulation were important desiderata of the transmis-
sion process, explaining why, in at least one manuscript containing the
Pantegni (BL Add. 22719), a whole repertoire of punctuation, colour-
coding, majuscules and minuscules was used to articulate the text.13
The twelfth century itself saw a proliferation of manuscripts with
scientific content. In some twelfth-century catalogues or booklists, separate
262 charles burnett
sections were assigned to the sciences. For example, in the catalogue of
books of Waltham Abbey, added to a bible in the late twelfth century, the
medical texts are listed together at the end, under the title De physica (ten
books, mainly translations by Constantine), while St-Bertin had six ‘libri
medicinales’ (without specifying authors or titles).14 In the twelfth-century
catalogue of the library of St Peter’s Salzburg, books for scholars were listed
separately from theological books. Under the heading ‘Hi sunt scolares
libri istius ecclesie’ are sixty-five books of classical authors and of scientific
works, including ‘Heremanus . . . super astrolabium’ (no. 36), an ‘expositio
super artem Euclidis’ (no. 42) and an ‘Alchorismus’ (no. 63).15 Among the
102 manuscripts listed in the twelfth-century catalogue of St-Amand, aside
from books on computus, rhythmomachy, the abacus, a ‘Sphera de quin-
que zonis’ (no. 14), ‘de arte architectonica et geometria’ (no. 51), two copies
of Boethius’ De institutione arithmetica (nos. 20 and 21) and Plato’s
Timaeus (no. 68), twenty-three books on medicine (with titles and authors
specified) are listed, i.e. nearly a quarter of the whole booklist.16 All these
libraries were in Benedictine abbeys, like Monte Cassino.
New translations from Greek and Arabic had contributed to these
books. The Greek manuscripts from which Burgundio of Pisa (ca.
1110–93), a papal notary who translated theological and scientific works
from Greek,17 made some of his translations can be identified; they include
the annotations of Burgundio and his Greek collaborator.18
The translations from Greek of Euclid’s Elements in BnF lat. 7373 and
Ptolemy’s Almagest in BAV Pal. lat. 1371, retain the ‘Eastern forms’ of the
Hindu-Arabic numerals which were used in the corresponding Greek
manuscripts.19 In the case of the twelfth-century translations of the
Elements made from Arabic, whose manuscripts far outnumber those
made from the Greek, different formats can be observed.20 The earliest
manuscripts of the earliest version (‘Adelard I’) retain not only Arabic
words in the text or the margin but also long lines, and no differentiation
between the enunciations and the proofs. Twelfth-century manuscripts of
the Adelard II version, in which the text was adapted to a Latin audience,
show a greater variety: either the enunciations are in the main text and the
proofs are in the margin or the enunciations are in a larger script than the
proofs, which may precede or follow them. Adelard’s text has been made to
conform to the pattern usual in contemporary theological and legal texts,
in which the lemmata are written in a larger script than the comment.
The main difference is the inevitable presence of the geometrical figures,
whether for proofs or for constructions. A tradition stemming from
Boethius’ partial translation of Euclid retained the proof or construction
 Reading the Sciences 263
lines on the figures, even when they were lacking in the text, thus providing
(perhaps) a mnemonic by which the reader could recall the stages in the
proof or construction which he had presumably learnt from his teacher.
But the deterioration of the figures in the course of copying and the breaks
in the teaching tradition meant that figures were no longer sufficient
without words. So the combination of figures and their corresponding
explanations became the norm in twelfth-century versions of Euclid’s
Elements (copies of the text without accompanying figures are rarely
found, even though the text of Euclid gives all the information the reader
needs to construct his own figures).21

Arabic–Latin Translations Issuing from the

Crusader States, Sicily and Spain
The three areas in which Latin scientific texts were produced from Arabic
models in the twelfth century were the Iberian Peninsula, Sicily with
southern Italy, and the Crusader States. In Antioch in the 1120s–40s
a second, more literal translation of the Royal Book of ‘Alī ibn al-‘Abbās al-
Majūsī was made by Stephen the Philosopher – who also composed
a cosmology based on a book by Ibn al-Haytham – and a translation of
the Almagest was made.22 Stephen’s translation of the Royal Book (Regalis
dispositio) retains the names of the scribes and the date of copying at the
end of each book, and includes the most complete trilingual glossary of
materia medica up to that date. The glossary was not taken from a single
source but was compiled by Stephen himself, using a known Greek glossary
to Dioscorides’ De re medica and an Arabic glossary which has not been
identified, to which he added his own Latin translations where he could.
The glossary covers eighteen pages in the oldest manuscript (the mid-
twelfth-century Berlin, Staatsbibl. lat. fol. 74), but the copyist clearly had
problems in keeping equivalent words on one line, and slanting ink
ligatures have been used to connect the words which have wandered
from their proper places.23 In the case of the Almagest translation
(of which only one fourteenth-century manuscript survives, containing
only the first four books), the tables have been omitted, either because they
can be constructed from the information given in the text or because they
may have been written in a separate fascicle (the same could be surmised for
a set of missing figures referred to in Hugo of Santalla’s translation of
‘Jafar’, On Rains24). The diagrams, however, are fully present, but many of
them are reversed in respect to the original, perhaps because the translator,
in reversing the line of script from Arabic to Latin, considered it
264 charles burnett
appropriate to reverse the diagrams too. Most particular to this manuscript
is the translator’s use of an unparalleled system of numeration by which the
Latin letters of the alphabet are used for digits in the same way as the
Greek, Arabic and Hebrew letters are used, but for the tens, a range of little
signs attached to the digit-letter must originally have been devised, even
though most of them have been omitted or misunderstood by the copyist.
Such an invention for numerals must have been known or taught in the
scientific community to which the translator belonged, but because of the
restricted size or duration of this community, its significance was quickly
lost. This is a hazard of scientific innovation.25
In Sicily, the presence of three languages, Arabic, Latin and Greek, also
encouraged the writing of equivalent texts in these languages in parallel
columns, for example in a psalter, written shortly before 1153 (BL Harley
5786).26 The contiguity of Arabic culture also accounts for the precocious use
of paper, in a codex combining a scientific text of Arabic origin (a portion of
the Book of the Three Judges) with Germanicus’ Aratea, bound into BL
Arundel 268, and apparently written in southern Italy ca. 1200.27
It was in the Iberian Peninsula, however, that the greatest number of
scientific manuscripts was produced, thanks to the translating enterprises
of Hermann of Carinthia, Robert of Ketton, Hugo of Santalla and Plato of
Tivoli in the north-east, and those of John of Seville, Gerard of Cremona
and Dominicus Gundissalinus in Toledo. Among these, John of Seville’s
translations of texts mainly on astrology are the earliest. In his translation
of al-Qabīsī’s Introduction to Astrology, we can see a variety of formats,
˙
probably representing different stages of revision of the text on the part of
the translator and other scholars who were able to consult the Arabic text
directly. One manuscript (BAV Reg. lat. 1285 of the late twelfth century)
distinguishes the words which have no equivalent in the original Arabic by
adding ‘va . . . cat’ at each end of the additional words and proposes
alternative translations of the Arabic text in the margin.28 In another
translation – Abū Ma‘shar’s On the Great Conjunctions – horoscopic charts
are included, but in the format of the chart one version follows one Arabic
manuscript, another version follows another.29
The most prolific twelfth-century Toledan translators, Dominicus
Gundissalinus (ca. 1110–90) and Gerard of Cremona (1114–87), were evi-
dently based in the cathedral itself (Dominicus was an archdeacon resident
in Toledo; Gerard was a canon). But there is no evidence for official
copying and ‘publication’ of the manuscripts in the cathedral. Rather,
the earliest manuscripts of their translations seem to have been copied by
students who visited Toledo from other centres. Dominicus translated
 Reading the Sciences 265
texts on cosmology and psychology, which were unillustrated. Gerard
translated texts on logic, geometry, astronomy, Peripatetic philosophy,
medicine, alchemy, geomancy and the calendar, many of which were
illustrated with diagrams. The carefully drawn illustrations of surgical
instruments in Gerard’s translation of Abu’l-Qāsim al-Zahrāwī’s Surgery
were a model for subsequent depictions.30 The layout of the church
calendar (Liber Anohe) diverges from the traditional layout for calendars
in that it includes depictions of the asterisms of the twenty-eight lunar
mansions presented in boxes, and a ‘Liber de accidentibus alfel’ consists
entirely of a table of the moon in each of the signs of the zodiac, with
a short set of instructions.31 Two manuscripts from the Veneto, written in
a single hand at the beginning of the thirteenth century, BnF lat. 9335 and
BAV Rossi 579, contain the most authoritative texts of Gerard’s transla-
tions. Both are written in two columns, with diagrams incorporated within
the text area and spilling into the margins. Numerous marginal glosses,
evidently by Gerard himself, are attached to the text using a large repertoire
of symbols. Many of these glosses give the Latin translation of a word left in
Arabic in the text, or vice versa. Gerard’s insistence on literal word-for-
word translations of the Arabic text meant that anything not belonging to
the text (such as alternative translations or explanations of the meaning)
had to be relegated to the margin.32 The fact that the two most author-
itative manuscripts of Gerard’s works originate from northern Italy gives
credence to the story that Gerard’s ‘body and books’ were taken back to
Cremona after his death, where the books were available for reading and
copying in the church of Santa Lucia. According to the Cremonese
chronicler Gasapino Antegnati, writing shortly after 1314, ‘these books,
for the most part, are extant today in the sacristy of this church (Santa
Lucia) in the same form as master Gerard translated them in his own hand
on paper,33 although very many are lost because certain people, having
borrowed them to make copies, did not wish to return them, and no one
remembered to ask for them back’.34

Booklets
One of Gerard’s pupils, Daniel of Morley, returned to England with
a ‘precious multitude of books’.35 The very portability of these books
suggests that they were booklets rather than large manuscripts that might
have been intended for the library of a monastery or the incipient Oxford
University. In fact, we find several examples of scientific works being
copied into unbound booklets; sometimes a quire (usually eight folios)
266 charles burnett
would be devoted to each short text. These booklets were often bound later
into larger manuscript volumes. Examples of this are the small, octavo-size
twelfth-century copies of Adelard’s Quaestiones naturales, Eton Coll. 161,
and London, Wellcome Libr. 4, which contain no other text, and consist of
thirty-seven folios and twenty-four folios respectively. Another example is
the medical manuscripts of ‘Magister Herebertus medicus’, who donated
his ‘books’ to the cathedral of Durham in the mid-twelfth century: twenty-
six books are named; several of them were included in one volume (the
phrase ‘in uno volumine’ follows the mention of groups of ‘libri’), so that
the whole donation consisted of only five or six manuscripts. One of these
is now Cambridge, Jesus Coll. Q. D. 2(44), which bears the inscription ‘ex
dono magistri Herberti medici’.36
It was clearly an advantage for someone engaged in practical science to
be able to carry a textbook around with him. The thirteenth-century MS
Brussels, Bibl. royale 8486–91 includes a small booklet of cures for sick
hawks which, as well as being eminently portable, has what looks like blood
stains on one of its pages. Later, doctors would carry around their ‘girdle-
books’ which depicted without any accompanying text the signs of the
zodiac and other astrological features such as the ‘zodiac-man’.37 But
already in the twelfth century scraps of parchment with horoscopes on
them have survived, left by the astrologer himself.38

Compendia
On the other hand, scientific works could be copied into large com-
pendia. We have Arabic precedents of manuscripts written in one hand
containing a large, but coherent, collection of scientific and/or philo-
sophical works, such as Istanbul, Ayasofya 4832 (ca. 1050), containing
sixty-five separate texts on astronomy, arithmetic, geometry and
meteorology,39 and Damascus, Zahiriyya MS 4871 (written between
1155 and 1162 in Baghdad), containing forty-three texts of Hellenistic
and Arabic philosophers.40 Such manuscripts are referred to as
majmū‘āt (literally ‘collections’).41 Such large compendia apparently
did not appear in the West before the late thirteenth century. But
the same spirit must have underlain the collection of twenty-seven
separate works on geometry, astronomy, arithmetic and algebra trans-
lated by Gerard of Cremona, which survives in BnF lat. 9335 and BAV
Rossi 579, both written by the same scribe and with identical layout.42
A similar case, again involving two manuscripts written by one scribe,
is the collection of astrological works translated by Hugo of Santalla (fl.
 Reading the Sciences 267
1145), apparently copied at St Augustine’s Abbey, Canterbury, before
the end of the twelfth century.43 Other translations of Hugo, including
that of the alchemists’ Tabula Smaragdina and a text on geomancy,
survive in twelfth-century manuscripts.44 Later collections of transla-
tions of magical texts (Florence, Bibl. naz. II. III. 214, and Darmstadt,
Landesbibl. 1410) and alchemical texts (the first half of Palermo, Bibl.
comunale 4 Qq. A.10), most of which are anonymous, may reflect
earlier collections which are now lost. The compilation of five magical
books into one volume is specifically mentioned in one of the texts in
the Florence manuscript: Solomon, De arte eutonica et ydaica (De
quatuor anulis), f. 28v: ‘Sciendum quod isti quinque libri artis magice
in unum volumen debent componi et sunt.vii. quaterni de vii. vitulis
nominatis’ (‘Know that these five books of the magical art must be
included in one volume, and [for this] there are seven quaternions
made from the seven calves named’); the ceremony for preparing this
parchment is then mentioned.
BnF lat. 9335 includes a note on the order in which mathematical works
should be read, which does not, however, exactly correspond to the order
of those same works in the manuscript.45 A more thoughtful attempt to put
scientific texts in a rational order can be observed in the Heptateuchon of
Thierry, chancellor of Chartres Cathedral, composed in the early 1140s
(Chartres, Bibl. mun. 497 and 498).46 Thierry’s aim was to include within
the covers of two volumes all the texts necessary for the seven liberal arts
(hence the title ‘Heptateuchon’, implying a secular equivalent to the biblical
Pentateuch). The second volume (49847) includes the older texts of
Boethius (De institutione arithmetica, De institutione musica and ‘geo-
metry’), Martianus Capella (De Nuptiis Mercurii et Philologiae, Book 7,
on arithmetic), several texts from the agrimensoral tradition, Firmicus
Maternus’ Mathesis and the abacus (including Arabic numerals), but also
the newly translated texts of Euclid’s Elements and al-Khwārizmī’s astro-
nomical tables. That these manuscripts were intended as repositories of
learning for the cathedral library, and not for practical use, is indicated
both by their large size (430 mm × 365 mm) and by the fact that the text of
Euclid does not include the proofs and diagrams.
The next step, after translating and collecting scientific texts, was to
anthologise them. This was the procedure of Hermann of Carinthia in
respect to his Arabic authorities on weather forecasting.48 Other examples
are The Book of the Nine Judges (an anthology of chapters taken from nine
different Arabic astrological authorities, and arranged by subject matter)
compiled by Hugo of Santalla (probably with the help of Hermann of
268 charles burnett
Carinthia),49 and Robert of Cricklade’s Defloratio of Pliny’s Natural
History, written for Henry II of England.50

The Mobility of Manuscripts

Thierry of Chartres incorporated into his Heptateuchon new texts – of
Euclid and al-Khwarizmi – that had arrived in Chartres probably through
the agency of their translator, Adelard of Bath. Thierry’s own pupil,
Hermann of Carinthia, advertised to his teacher further texts translated
by his colleague Robert of Ketton and himself, probably in the hope that
they would also be included in the cathedral library; in fact, we have more
evidence for new scientific texts being introduced into Chartres in the
mid-twelfth century than for any other European centre.51 There are
several examples of manuscripts travelling. Henricus Aristippus brought
back a copy of Ptolemy’s Almagest from Constantinople to Palermo on an
embassy for William I, king of Sicily, shortly before 1160.52 Petrus Alfonsi,
when he travelled to northern Europe after his baptism in Huesca in 1106,
brought books with him,53 as did Constantine the African, when he travelled
from Qairouan to Palermo (allegedly losing some manuscripts in
a shipwreck on Cape Palinurus). An anonymous translator endures great
hardship to travel from Salerno to Palermo to read Aristippus’ Greek manu-
script, and Aristippus in turn tries to persuade another scholar, whom he
names ‘Roboratus’ (perhaps Robert of Selby), not to return to England but
to enjoy the libraries of Sicily, where he will find manuscripts of Heron’s
Mechanica, Euclid’s Optica, philosophical works of Anaxagoras, Aristotle,
Themistius and Plutarch and plenty of medical texts.54
I end this short survey of scientific books by looking in detail at five
scientific manuscripts from the twelfth century. This collection of case
studies shows how some of the points raised separately in the survey can be
illustrated in combination in a single manuscript. It will also give an idea of
the variety of manuscripts containing scientific works. In certain cases, it
will demonstrate the relationship between a scientist and a manuscript that
contains his work or that he copied.
1) Bodl. Libr. Auct. F. 1. 9 is a manuscript from Worcester Cathedral
Priory, written in a single hand in the 1140s. It consists of mathematical
texts composed between the early eleventh century and the 1120s to the
1140s. As befits a monastic scriptorium, the manuscript is written very
carefully, with concern not only for appearance but also for the accuracy of
the text (by subpunction, erasure and/or interlinear additions or
 Reading the Sciences 269
replacements, all performed with care not to disfigure the text). One of the
scholars who must have had some role in compiling the volume, Walcher,
prior of Malvern (d. 1135), a Benedictine house close to Worcester, included
a work of his own, in which he emphasises the necessity of using a ruler
(regula) and dividers (circinus) to make the necessary diagrams illustrating
the orbits of the sun and the moon.55 Foreign words (mostly Arabic, but
a few Greek), Roman numerals and fraction symbols, and letters in the text
referring to corresponding letters on the geometrical figures are highlighted
in red.56 Arabic numerals occur in this manuscript, but only in the tables,
which are framed in yellow and green with the grid lines in red.57
The margins are wide, and most of the texts are separated by blank
pages, half pages or columns. In the star table on f. 78v all the Arabic
names of the stars are in red, asterisms are included, and many alternative
names are given in small black script above the rubricated names or, if there
is not room, in the margin.58
2) Cambridge, Trinity Coll. R.15.16 is a small manuscript from the mid- to
late twelfth century consisting entirely of works on arithmetic. The classic
work, Boethius’ De institutione arithmetica, has been beautifully written in
a fine Caroline hand, with lavish use of red and green paint and immacu-
lately drawn diagrams depicting the arithmetical, geometric and harmonic
proportions and other arithmetical consonances. But the mathematician
who owned it (perhaps the ‘Testardus’ whose name appears in the top
margin of f. 1r59) added glosses within the manuscript and notes on the
flyleaves at each end. These notes are necessarily written in a much more
condensed way than the main text, since the scribe evidently wished to fit
in as much information as possible. At the beginning he writes an intro-
duction to arithmetic in the context of the seven liberal arts, which stops
abruptly (f. 3r), after which the page is blank. But we are evidently catching
the author in the act since he adds some notes (in the bottom margin of
f. 2v) indicating what he wants to add about music. He also provides rules
for playing the arithmetical game of rhythmomachy, and draws
a rhythmomachy board at the end of the manuscript (f. 60v) nicely
mirroring the chessboard at the beginning (f. 1r), both being, as it were,
templates for the real thing. In addition there are jottings using Roman and
Arabic numerals, and a key to the Arabic numerals, which suggests that
they are not yet familiar (Figure 14.1).60 Beside ‘Testardus’, other personal
names, Radulfus and <H?>osatus, indicate that the book was probably
used in the classroom or by individual scholars before it ended up in the
library of the Franciscans of Coventry.
270 charles burnett

Figure 14.1 A key to Arabic numerals with their names. Cambridge, Trinity College,
R. 15. 16, f. Av.
 Reading the Sciences 271
3) BL Add. 22719 is a copy of the Pantegni of Constantine the African. This
is a monastic manuscript which belonged to the Priory of St Nicolas in
Exeter but whose script resembles that of manuscripts written at Bath.61
It appears to have been copied in the second or third decade of the twelfth
century. As with the manuscript on arithmetic, so here we have a principal
text to which others have been added. The principal text is Constantine the
African’s Pantegni (the whole of the Theorica and two and a half of the ten
books of the Practica, as is common in early manuscripts of the Pantegni),
but in all the interstices between the books and the works other texts have
been included: before the Theorica there occurs a Christian spell and
glossary of medical terms (largely Greek). Between the Theorica and
Practica occurs a text on the elements taken from the Arabic version of
Nemesius’ De natura hominis and some recipes reflecting local health lore
(with words in Anglo-Saxon). Between the second and the partial ninth
book of the Practica Pantegni is a short work on the humours and more
recipes. After Book IX there come ‘quedam phisica’ covering the use of
magical remedies (Qustā ibn Lūqā’s De physicis ligaturis), metals, foods and
weights. These additions ˙ have been written with the same care as the main
text of the Pantegni. The scribe uses a repertoire of colours and diacritical
marks to articulate the text in the manuscript, as can be observed on a
page of the text on the elements. One can distinguish a descending
hierarchy of articulation: rubricated initial capitals, discrete green paraphs,
half-rubricated capitals, a green dot over the first word of a subsection,
rubricated lower case letters, punctus versus and punctus medius; the super-
script green dots also pick out ‘aqua’ and ‘aer’ as two of the four elements.
4) Bodl. Libr. Digby 51 contains works of astronomy and astrology, mostly
translated from Arabic. One of its scribes knew Arabic – he copies out
a diagram of the four corners of the world, writing Arabic letters on it. He
has supervised the work of at least four other scribes and has filled in
lacunae and made corrections. There is great economy of space, the works
written in two tightly packed columns. But a wide margin has been left at
the bottom of each page, which is sometimes filled with illustrations (as on
ff. 14v, 15r, 81v). Slips of the same thin vellum used throughout the manu-
script have been added between ff. 19 and 21, and ff. 23 and 25, on which the
scribe of the main text has made some additions to the text.62
The manuscript consists entirely of works written in north-east Spain in
the mid-twelfth century, apparently in the circle of Plato of Tivoli, by
authors who evidently knew each other: one work of Rudolph of Bruges,
Hermann of Carinthia’s student, one work of Robert of Ketton, dedicated
272 charles burnett
to the same Hermann of Carinthia, one work of Abraham Ibn Ezra, two
translations usually attributed to John of Seville, but one of which
(Māshā’allāh, On Eclipses) ends with a prayer for Plato of Tivoli, and five
translations of Plato of Tivoli himself. Binding these texts together are the
dedications by Rudolph of Bruges and Plato of Tivoli respectively to ‘John
David’, who can plausibly be identified with a Jewish mathematician
whom Abraham Ibn Ezra also addressed in one of his astrological
works.63 It is tempting to think of the supervising scribe as Plato of
Tivoli himself.
5) Bodl. Libr. Selden supra 24 is a collection of booklets of Peripatetic
natural science, together with a work brought to England from Toledo by
Daniel of Morley, a student of Gerard of Cremona. The first booklet (ff.
1–40) consists of the Metaphysica vetustissima, translated by James of
Venice, and the Ethica vetus (Books 2–3). The second booklet (ff. 41–63)
contains the De generatione et corruptione translated by Burgundio of Pisa.
The third booklet (ff. 64–75) includes only Pseudo-Avicenna, De caelo et
mundo translated from Arabic by Dominicus Gundissalinus and Johannes
Hispanus, while the fourth booklet (ff. 76–84) has only the Liber de causis,
also translated from Arabic, and associated in this manuscript with the
Jewish collaborator of Gundissalinus, Ibn Daud (‘Metaphisica
Avendauth’). The final booklet (ff. 84–117) also consists of one work:
‘Libri Metheororum’, comprising the four books of Aristotle’s
Meteorologica with Avicenna’s chapters on the formation of stones and
minerals as its appendix. The booklets were put together at St Albans
Abbey in the early thirteenth century, evidently because of the similarity
of their subject matter. The St Albans ex libris appears at the beginning of
the collection, where there is a list of contents, and at the beginning of the
final booklet. Other booklets would once have been bound in the same
collection, since there are gaps in the quire numbers, and some works in the
list of contents are missing. Each of the booklets is written in a different
hand, apparently at different times between the late twelfth and the early
thirteenth centuries, and at different places. The first two booklets contain
works translated directly from Greek; the third and fourth contain works
translated from Arabic, and the fifth a combination of books translated from
Arabic and Greek.64 The booklets, however, share certain formal features.
Spaces are left for notes. In the case of the Metaphysica the scribe has left
space in the margins according to the quantity of glosses to be inserted,
indicating that they were already in the exemplar.65 Most notes are preceded
by a paraph. Except in the case of De generatione et corruptione and the
 Reading the Sciences 273
Meteora, titles do not appear at the heads of the individual texts.66
‘Methaphisica’ has been added in dry point on f. 14v as has ‘secundus
liber’ on f. 27v. On the same folio, along the extreme left hand margin
(now partly cut off) are the words ‘incipit primus liber de ethice’.67 Some
works have their title at the end of the text (f. 83v: ‘Explicit metaphisica
auendauth’ – in a different hand from the text; f. 117r: ‘completus est liber
metheororum aristotilis cuius tres libros . . .’ etc.). This is just the kind of
collection out of which would arise the Corpus Vetustius used for the
teaching of philosophy in the incipient European universities.

Conclusion
These five case studies range from the second decade of the twelfth century
until the turn of the thirteenth century. They give examples of collections
of texts on medicine, arithmetic, astrology and astronomy, and natural
philosophy, copied in north-east Spain, in southern Italy and in the
western and eastern regions of England. Each manuscript tells its own
story. As we have seen, the innovatory nature of the texts often gives rise to
innovation in format and layout. Scientific manuscripts cannot be
regarded as forming a genre of their own, independent of other manu-
scripts being written in the same contexts. But they do offer a richness and
variety which makes them most interesting objects to study.

Notes
1. Borelli 2008, 21–2.
2. Millàs Vallicrosa, 175, 290–1 and Plate IX = BnF lat. 11248, eleventh cent., ff.
18v–19r; Schramm et al. 2006–7, 300 (MS Bern, Burgerbibl. 196, early
eleventh cent., f. 8v).
3. Kunitzsch 1998 (BnF lat. 7412, mid-eleventh century).
4. These are fully documented and edited in Juste 2007.
5. Folkerts 1996.
6. This single large piece of parchment (probably originally 680 mm ×
440 mm), now Luxembourg, Bibl. nat. 770, is illustrated and described
in Burnett 2002. For 680 mm × 480 mm as the size of an average skin, see
Gumbert 1993, 236.
7. Full manuscript descriptions are given in Toneatto 1994–5.
8. See the description of Cambridge, Trinity Coll. R.15.16 in what follows.
9. Behrends 1976, 260–1.
10. Falmagne 2009, 2. 376–7.
11. See Chapter 15 of this volume.
274 charles burnett
12. Newton 1998, 367 and pl. 135, and Kwakkel and Newton, in press.
13. See later in this chapter.
14. CBMLC 6, 439; Becker, Catalogi, no. 77.
15. Ibid., no. 115.
16. Ibid., no. 114. For statistics on the number of extant manuscripts on medicine,
the science of the stars and natural science, see, respectively, Chapter 15 of this
volume, Juste and Burnett 2016 and AL.
17. Classen 1974.
18. Vuillemin-Diem and Rashed 1997.
19. Burnett 2002, 252–3.
20. The two earliest versions were probably both the work of Adelard of Bath (ca.
1080–1150), and are conventionally known as ‘Adelard I’ and ‘Adelard II’.
21. Folkerts 1970, 220–7; Burnett 2013.
22. For comparisons between Stephen’s translation and that of Constantine the
African, see Jacquart 1994, 83–9. Dirk Grupe’s editions and English transla-
tions of Stephen the Philosopher’s cosmology (Liber Mamonis) and the
Almagest (translated by a certain ‘Abd al-Masīh of Winchester) will be
published shortly. ˙
23. Burnett 2013a: see illustration on p. 76.
24. Burnett 2007, 162. In the case of On Rains, one of these illustrations is found
transmitted separately from the text in Cambrai, Médiathèque mun. 168,
f. 106v: see Burnett 2004, 79–81.
25. This numeral system is discussed in Grupe’s edition.
26. Burnett 2003–4.
27. Reeve 1980, 511–5; Burnett 2002, 257.
28. Burnett 2001.
29. Burnett 2007, 169, 174–5.
30. Irblich 1981.
31. See Juste 2015, 187 and pl. II.
32. E.g., ‘id est alfabet’ in the margin explaining ‘almuagemmati’ in the text:
Schupp 2005, 20.
33. This is plausible, given the production of paper in Islamic Spain from the
early twelfth century (especially in Jativa), and the use of paper, for example,
in the manuscript of the Leiden Arabic-Latin Glossary, which was probably
written in Toledo in 1175: see Koningsveld 1977 and BL Arundel 268.
34. See Leino and Burnett 2003, 286: ‘Qui libri ut plurimum extant hodie in
segrestia [sacristy] dicte ecclesie sicuti prenominatus magister Girardus pro-
pria manu ipsos translatavit in cartis bombicinis, licet quamplurimi sint
deperditi ex eo quod aliqui ad exemplandum commodati eos restituere
noluerunt et propter oblivionem non fuerunt requisiti.’
35. ‘cum pretiosa multitudine librorum in Angliam veni’: Daniel of Morley,
Philosophia, 212 (preface).
36. Robinson 1980. ‘In uno volumine’ could plausibly refer to a loose wrapper.
37. Carey 2004.
 Reading the Sciences 275
38. E.g., two scrappy pieces of parchment, now bound into BL Royal App. 85, as
ff. 1–2, on which ten horoscopes have been drawn, dating from 1135 until 1160:
North 1987.
39. Sezgin 2010.
40. Ragep and Kennedy 1981.
41. For this genre in general, see Endress 2001.
42. The two manuscripts are the same size (365 mm × 240 mm), use the same
format for catchwords, and consist of quires made up of five bifolia rather
than the more usual four bifolia.
43. Bodl. Libr. Digby 159 and Cambridge, Gonville and Caius Coll. 456/394.
At least one more volume may have once existed, since the scribe of Bodl.
Libr. Savile 15 (s. xv) copied the texts in both these manuscripts, as well as two
further texts by Hugo from a manuscript that has not yet been identified.
44. These are BnF lat. 13951 (Ps-Apollonius, De secretis naturae, including the
Tabula Smaragdina) and Bodl. Libr. Digby 40 (geomancy).
45. The note is attributed to Hunayn ibn Ishaq (Johannitius), the translator of
many of the works from Greek into Arabic. It is a translation of an Arabic text
that survives in Beirut, Université de St Joseph 223 (I owe this reference to
Sonja Brentjes).
46. Jeauneau, ‘Prologue’.
47. Only fragments survived the destruction of the Bibliothèque municipale in
World War II, but a microfilm of the whole MS is available.
48. ‘eam summatim transcurrere diversorumque diversam sententiam sub quo-
dam compendio redigere curavi, ut quicquid verborum numerositas occulta-
bat, aut physicorum dissona multitudo variabat, plerumque etiam inconcinna
scribentium digressio dilatabat, simplicis pagine brevitas absque omni scru-
pulo representet’ (‘I have tried to run through the subject briefly and to reduce
to a kind of summary the different opinions of different men, so that,
whatever the multitude of words used to hide, or the discordant throng of
scientists used to confuse, and especially whatever the inelegant ramblings of
writers used to spin out, might be presented in a single page without
a stumbling block’): Burnett 1978, 123. This work survives in at least one
twelfth-century manuscript: Erfurt, Universitätsbibl. Dep. Erf., Amplon. 4°
365, ff. 50–52.
49. Burnett 1996.
50. Haskins 1927, 169.
51. Burnett 1984.
52. Haskins 1927, 157–65.
53. He mentions them in his conversation with Walcher of Malvern, De dracone
2.2: see Nothaft 2017.
54. See the preface of Aristippus’ translation of Plato’s Phaedo, reported in
Haskins 1927, 168–9.
55. Walcher of Malvern, De lunationibus, 3. 3.
56. Bodl. Libr. Auct. F. 1. 9, f. 78v (Arabic star names in red).
57. Bodl. Libr. Auct. F. 1. 9, f. 56v.
276 charles burnett
58. For a full description of the contents of this manuscript, see Nothaft 2017.
59. ‘Ex cerebro testardi’ (‘from the brain of Testardus’). This name, meaning ‘big
head’, would make us think of Robert Grosseteste, were the manuscript not
too early for him. The hand resembles closely that of the annotator of Soest,
Stadtbibl. 24, who adds the equivalent Arabic terms (in transliteration) to the
terms in Firmicus Maternus’ Mathesis: see Burnett 2004a.
60. For more discussion of the contents of this manuscript, see Burnett 1996,
244–52, and Burnett 1997.
61. I owe this judgement to Michael Gullick and Teresa Webber.
62. In the first instance, he repeats the star table in the text on f. 19v, but gives the
Arabic star names in different spellings and sometimes adds alternative names,
or the constellation in which they occur. In the second example he adds
a phrase which has been omitted in the text.
63. See Burnett 2017.
64. The first three books of Aristotle’s Meteora were translated from Arabic by
Gerard of Cremona, the fourth from Greek by Henricus Aristippus, and
Alfred of Shareshill, who evidently brought both parts together, added his
own translation of the chapters of Avicenna on stones and minerals, probably
considering them also to be the doctrine of Aristotle: see Mandosio 2010,
245–52.
65. The margins are conspicuously wide on ff. 10v and 12v; on f. 5v one note is
placed in a rectangular box (more boxed notes appear on f. 43r et seqq.); other
notes take the form of inverted triangles (ff. 3v, 5v and 8r).
66. The De generatione et corruptione is headed by a now invisible title, but Minio-
Paluello read it as the Greek title ‘peri geneseos keftoras’.
67. This feature, which would allow the reader to leaf through the booklet to find
the beginning of the work, is found elsewhere – e.g. in the margins of Adelard
of Bath’s Isagoge minor in BL Sloane 2030, f. 83r.
 chapter 15

Medical Books
Monica H. Green

Toward the end of the twelfth century, a scholar in northern Italy who
identified himself only by the name ‘Johannes’ drew up a list of the
medical books he owned.1 Distributed across what seem to have been
several different volumes, the twenty-six texts had all been newly
edited, composed, or translated in the previous century. Johannes was
witness to a revolution in medical learning and book culture that had
just recently taken place, a revolution that allowed him and countless
other clerics to claim special learning in a profession that had barely
existed a century before. Actually, as we shall see, there had been three
revolutions, three distinct moments of new text production that
occurred over the course of the period. At the time Johannes was
drawing up his own library’s items, medieval Europe had available
medical books in quantities and with a diversity of content that had
never been seen before.
Aside from his very obvious interest in medicine, we know nothing
about Johannes other than that he was wealthy enough to own so many
books, among which was the extant BAV Vat. lat. 10281, containing his
booklist. Similar personal libraries were formed at about the same time. For
example, Salomon, a monk of St Augustine’s Canterbury, active at the
beginning of the thirteenth century, had many of the same volumes as
Johannes had in north Italy, and so did Master Herebertus ‘medicus’, who
gave his books to Durham Cathedral.2 Salomon owned at least six medical
volumes (containing at least twenty-three medical texts), Herebertus seven
(containing at least twenty-two medical texts). Of the three, Johannes was
most invested in the theoretical aspects of medical education, to judge from
his strong interest in the Articella, a teaching compendium of short texts
that had come together by the final quarter of the eleventh century.
Johannes owned two copies of the collection, two sets of glosses on it,
and a guide to the glosses. Even without biographies of these books’
owners, therefore, the books themselves give us the means to perceive
277
278 monica h. green
changes that medical book culture underwent over the course of the Long
Twelfth Century.
The present chapter has two aims. First, drawing on a decade-long
survey of all 550 surviving Latin medical books from this period, it offers
an overview of the Long Twelfth Century’s new medical texts in the three
chronological phases I have identified: what they were, where they were
produced, and how their patterns of geographic dissemination might be
traced. Second, it demonstrates that there was little that unified medical
books as physical objects other than their specialized content. To the extent
that there are general codicological shifts across this period (in script, page
layout, dimensions, paratext, and decoration), the medical book does not
stand out from developments in other areas of book culture. Importantly,
medical books of this period usually lacked something we now think
normative in the field of medicine: illustrations. Older texts that had
been illustrated kept them, occasionally elaborating them elegantly. But
no new texts created programs of illustration, and only one text translated
from the Arabic imported illustrations from its source. A particularly
unusual manuscript, because it violates these general patterns of topical
specialization and aversion to visual elements, will, I argue, prove these
rules. This manuscript is also unusual in that it is dated, a trait it shares
with only two others in our corpus. Half a dozen more are approximately
datable on the basis of circumstantial evidence, but most have no prove-
nance information before the late medieval period. This survey is therefore
based on paleographical assessments of approximate date and locus of
production.3

Generating a New Body of Texts: Monte Cassino and Salerno,

Antioch and Toledo
At least 550 volumes of Latin medical literature copied between ca. 1075 and
ca. 1225 still survive, in addition to at least 160 catalogue or inventory
witnesses to books owned by particular houses or individuals in the same
period. These extant manuscripts and documentary witnesses testify to active
interest in medicine all across Europe. Salomon, for example, the monk of St
Augustine’s Canterbury in the early thirteenth century, had works by recent
‘masters’ of medicine from the southern Italian town of Salerno, but also
works from the preceding stage of the developing medical corpus, that is,
translations by Constantine the African and the set of the introductory texts
that had formed the teaching curriculum in medicine for the previous
century, the Articella. Northungus, a monk active in Hildesheim probably
 Medical Books 279
from the 1120s to the 1140s, knows only of the great corpus of works that
Constantine translated into Latin from Arabic a few decades earlier, plus
Stephen of Antioch’s retranslation (following Constantine) of ‘Alī Ibn ‘Abbās
al-Majūsī’s Book on the Whole Art of Medicine. Geography was therefore no
determinant of what texts were available across Western Europe in the twelfth
century; most of these works were as readily available in England as in Italy.
Few of the surviving books provide evidence of their place of production.
No single codicological feature unites the 550 extant codices identified thus
far for this project. They range in size from tiny handbooks of approximately
120 mm × 60 mm to 890 mm × 490 mm, the dimensions of the Codex Gigas
(Stockholm, Kungl. Biblioteket A. 148), the largest known book from the
medieval period. Similarly, they are not defined by any single feature of
layout, ruling, or decoration. The ‘holsterbook’ format that Erik Kwakkel
has investigated – a tall, narrow codex that can easily be held in one hand,
making it ideal for instructional settings – is no more characteristic of
medical books than liturgy, philosophy, or grammar.4 Rather, the chief
characteristic of a ‘medical book’ is its exclusively medical content; more
than 510 of our 550 extant volumes have no significant content that cannot
be characterized as medical. Therefore, the main points of book-historical
analysis that can be offered here hinge on content: why certain texts show up
when they do, where they do, and why in particular combinations with
others. Because we can pinpoint where most of these new medical texts were
being composed, we can begin to plot a geography of their dissemination as
physical objects distributed and recreated across Western Europe.

Monte Cassino, Stage 1: Recovery and Renewal

Starting in the mid eleventh century, the recovery of a variety of late
antique Latin texts prompted extensive editing and reorganization.
The research project underlying the present study is the first to identify
the Benedictine monastery of Monte Cassino as something more than the
center at which Constantine the African made his medical translations
from the Arabic. Constantine came into a setting that was already flourish-
ing. Monte Cassino has long been famous among classicists for its pre-
servation of rare texts from antiquity,5 and it should now be recognized for
holding that distinction in the field of medicine as well. Cassinese monks
retrieved centuries-old texts and edited them, both by collating them with
other copies to identify problematic passages and by remaking them –
abbreviating them, fusing them with other works, or putting them into
more rational order.
280 monica h. green
When Augusto Beccaria surveyed what he called the ‘presalernitan’
corpus of medical texts and manuscripts in 1956, he identified 145 extant
volumes and approximately forty medical items in booklists and inven-
tories of the ninth to eleventh centuries. In those volumes, one can
distinguish about six dozen different texts. Few are extant in more than
ten copies across the three centuries, and most are anonymous or attributed
to a handful of ancient medical figures, some real, some apocryphal. For
our period, in contrast, which picks up where Beccaria left off, we can
identify about 200 texts in circulation, some of them of such popularity in
both numbers and breadth of circulation that we can begin to speak of
a common European body of medical knowledge.
Several of the ‘bestsellers’ of the Long Twelfth Century were either
composed or given new life at the monastery of Monte Cassino, or at the
very least, in the southern Italian Beneventan zone (see Figure 15.1).
The top ‘bestseller’, the Passionarius of Gariopontus, a physician of
Salerno, was itself an edited version of an ensemble of texts that had first
come to circulate together in the early Middle Ages and which, by the
mid eleventh century, had been expanded by the addition of a new text, an
excerpt on gout from the sixth-century Greek physician Alexander of
Tralles. Adding together this pre-Gariopontean ‘De podagra Grouping’
and Gariopontus’ own edited text, we come up with sixty extant copies of
this ensemble of late antique works surveying most aspects of medicine,
except for surgery, which hardly registered in the early medieval Latin
corpus at all, and women’s medicine, which would have a different
fortuna.6
A cluster of medical books was written by known Monte Cassino
scribes between the 1060s and the 1080s.7 One of these volumes, now
in Copenhagen, forms part of the general expansion of the abbey’s
buildings and treasure under the abbacy of Desiderius; it was written
by at least two of Monte Cassino’s top scribes, who (as we will see) were
also involved in making other major works of medicine newly available.8
It was probably assembled by retrieving out of Monte Cassino’s own
armaria two texts on women’s medicine from the late antique period.
In 945 Leo, a priest from Larino (Molise), gave to the abbey of Monte
Cassino the church and monastery of San Benedetto along with all its
goods. Those included ‘[libri] medicinales III, Galienum, Aforismum et
Genicia et Asclepius’.9
The texts of both Muscio (a late antique North African writer) and
‘Cleopatra’ that we find in the Copenhagen manuscript would almost
immediately be reworked into new forms. From the ‘Cleopatra’, a new,
 Medical Books 281

Extant Origin (MC = Monte

Title Copies Witnesses1 Cassino; S = Salerno)

*†Gariopontus, Passionarius 52 10 S
*Constantinus Africanus, Pantegni [Arabic] 51 12 MC
*Pseudo-Macer, De viribus herbarum 39 6 Loire Valley?
*Constantinus Africanus, Viaticum [Arabic] 37 8 MC
*†Articella (incl. Hunayn ibn Ishaq, Isagoge 35 23 MC?
[Arabic]; Hippocratic Aphorisms [Greek];
Hippocratic Prognostics [Arabic];
Philaretus, De pulsibus [Greek];
Theophilus, De urinis [Greek]; Galen,
Tegni [Greek]); Hippocratic Regimen
acutorum [Arabic])
*Antidotarium magnum 33 10 MC
*†Constantinus Africanus, De gradibus 29 7 MC
[Arabic] 28 9 MC2
Isaac Israeli, De dietis universalibus et
particularibus [Arabic]
Pseudo-Apuleius Complex 14 83 MC, etc.
*Alexander of Tralles, Practica 14 5 MC, etc.
Bartholomeus, Practica 14 4 S
*Dioscorides alphabeticus 13 6 MC
Pseudo-Cleopatra, Genecia 13 2 MC

Note: For works newly translated in this period, I indicate the source language.
An asterisk (*) identifies texts that at least sometimes constituted a volume unto
themselves. A dagger (†) indicates works that also appear in the library of Johannes
(listed in BAV Vat. lat. 10281, f. 41r).
1. I only list here items witnessed in contemporary catalogues or booklists that cannot be
connected to extant MSS. I also only list items for which the identification is reasonably
certain; thus, ‘liber medicinalis’ is too vague to be interpreted.
2. I have listed the two parts of Isaac’s work on diets together here, though in fact twelve of
these twenty-eight copies have only one or the other text. In Isaac’s Arabic original this
was indeed a single text.
3. I have included in this count all books called herbarius; five of these are specifically
identified as depictus.
Figure 15.1 Medical ‘bestsellers’ of the Long Twelfth Century.

shorter version of the work would be produced, prefaced by a list of thirty-

five chapters. This abbreviated ‘Cleopatra’, of which thirteen copies are
extant, ranks as one of the ‘bestsellers’ of the Long Twelfth Century. From
Muscio’s work, two different texts would be produced, also radically
abbreviated: (1) Non omnes quidem, a work in seventy-eight chapters
made up of selections from Muscio together with other material, mostly
282 monica h. green
focusing on obstetric issues; and (2) De passionibus mulierum B, a work in
twenty chapters crafted from the opening section of a Greek text attributed
to a female figure, Metrodora, the ‘Cleopatra’ text, and selections from
Muscio.
These reworkings of ‘Cleopatra’ and Muscio might seem insignificant,
but for text after text in this period we see these same patterns of retrieval,
editing, and repurposing. All three works of women’s medicine just men-
tioned are consistently prefaced by tables of contents, even in their earliest
copies. In another realm, pharmaceutics, we find alphabetization used as an
organizing factor. This guides the production both of the Antidotarium
magnum, a work collecting up to 1,300 different compound recipes and
found in thirty-three copies from our period, and in the Dioscorides
alphabeticus, a reworking of a late antique translation of the first-century
CE Greek pharmaceutical author. Interestingly, lost in both the produc-
tion of the Dioscorides alphabeticus and the Muscio adaptations were the
illustrations that had accompanied both texts since their origin. Since these
illustrations were found in the very manuscripts at Monte Cassino that had
served as the redactors’ exemplars, the decision to omit the visual elements
was clearly deliberate.
Monte Cassino was not the only house to dust off its old classics and give
them new life in this period. One new branch of the Pseudo-Apuleius
Complex tradition (for which see later in this chapter) seems to begin at St
Gallen and gave rise to a small line of descendants that extends into the
fourteenth century.10 A ninth-century copy of Alexander of Tralles’
Practica, which had probably been translated from the Greek in the sixth
century, gave rise to several new copies in France.11 And, as we shall see,
there is reason to suspect that houses were exchanging copies for compar-
ison and collation. But Monte Cassino is the only place that seems to have
engaged in multiple efforts to assemble – indeed, to reimagine – something
like a full library of medical writing.

Monte Cassino, Stage 2: Translation and Dissemination

Our understanding of Monte Cassino’s role in sponsoring the translation
program of Constantine the African (d. before 1098/9), the first translator
of medicine from Arabic into Latin, has greatly increased in recent years.
More than thirty different texts can now be associated with Constantine.
Whether these brought any codicological features of their Arabic exemplars
over into Latin is as yet unclear, since few comparative studies of both
Arabic and Latin traditions have been done. What we do know is that the
 Medical Books 283
same scribes who were involved in producing the Copenhagen manuscript
also contributed to the three earliest known copies of Constantine’s works,
which were made under his supervision.12
But what happened after Constantine’s works left Monte Cassino?
In fact, we know that by the second quarter of the twelfth century, select
works of Constantine had reached major libraries across Western Europe.
One of the earliest beneficiaries was Fleury, Monte Cassino’s long-time
rival. A number of medical manuscripts survive from Fleury that are either
written in Beneventan script or very early twelfth-century witnesses to texts
coming directly out of southern Italy. This is unlikely to be a coincidence.
In 1106–7, an entourage of monks from Monte Cassino traveled to Fleury
seeking to negotiate the return of the relics of St Benedict, which had been
removed to Fleury in the eighth century after Benedict’s foundation was
sacked by Muslim pirates.13 Nothing in the account given in the Chronicle
of Monte Cassino speaks about gifts exchanged between the two houses,
least of all a specific gift of books.14 Yet the patrimony of Fleury, recon-
structed by modern scholars after having been scattered in the sixteenth
and nineteenth centuries, presents a cluster of extant medical books
unrivaled in number by any other religious house from the period; not
by the cathedral at Durham, whose library has remained intact for
a thousand years, nor even by Monte Cassino itself.
The Fleury collection duplicates works coming out of the first two waves
of medical book production in southern Italy. From the first wave, in
which late antique books were newly edited, we find Galen’s Ad Glauconem
and the pseudo-Galenic Liber tertius, texts that often circulated together in
eleventh-century manuscripts.15 From the second wave, the period of new
translations and the production of the Articella, we find Constantine’s
Viaticum, De gradibus, Pantegni, Theorica, and Johannitius’ Isagoge; the
pharmaceutical reference books, the Dioscorides alphabeticus, and the
Antidotarium magnum; and the teaching compendium, the Articella.16
If further proof of their source were needed, the copies of the Viaticum,
Isagoge, and Dioscorides alphabeticus are all in Beneventan script.
A manuscript not previously associated with Fleury takes us to the next
stage: now that these new (and in some cases, sizeable) texts are available,
what does one do with them? Berlin, Staatsbibl. lat. qu. 198, is one of our
rare dated manuscripts. The colophon ‘in era MCLXX’ appears at the end
of one text, a formula in the Mozarabic calendar (in Gregorian calculation,
1131/2). Written in several hands, the decoration and script point to south-
ern France or northern Spain. Medical texts make up the bulk of the 155-
leaf volume, yet the term ‘medical book’ is not quite appropriate. Excerpts
284 monica h. green
from various texts are inserted, sometimes in places that had been left blank
at the end of main entries. Their character is diverse. While three and a half
lines of De vino were added by a later hand at the end of one text (f. 75r),
a short passage on the Tower of Babel was added in the blank half page
between the end of the prologue of a text and its opening chapter (f. 10r).
Certain interests emerge: a selection from Dioscorides on mandrake
(ff. 1v–2r) and an extended recipe for diacodion, an opiate compound
(ff. 4v–5r), suggest a concern with the alleviation of pain. Several texts on
diets show a desire to maintain health through food choices, while sections
on prognostics focus on determining the likely outcome of a sickness.
A hitherto unnoted excerpt from Vegetius’ Digesta artis mulomedicinalis
and the earliest extant copy of the Physiognomia Loxi, Aristotelis, et
Polemonis expand the volume’s coverage further into the natural
sciences.17 A short text on phlebotomy follows Constantine the African’s
De oblivione (On Forgetfulness), though there are also short texts on
mathematics and the calculus of years since the Creation.
These scattered excerpts (most no more than two to four pages long)
surround the four main texts of the volume: the Euporista of the late
antique North African writer Theodorus Priscianus (ff. 9r–73r);
Constantine the African’s translation of Isaac Israeli’s De dietis particular-
ibus (ff. 76r–111r); Vegetius’ Epitoma rei militaris, here in a standard set of
excerpts characterized by what Reeve called the ‘false preface’ (ff.
116r–128v);18 and the earliest Latin translation of ‘Arīb ibn Sa’d’s Kitāb al-
anwā’, ‘The Calendar of Cordoba for the Year 961’ (ff. 135v–143v), a month-
by-month guide for agriculture, horticulture, and personal regimen,
written following the Christian calendar from the perspective of life in
the Guadalquivir valley.19
The Berlin manuscript is also striking for its illustrations. As indicated
earlier, the use of images was not characteristic of medical manuscripts in
this period, and here we see not innovation, but repetition. On the last
page of the opening bifolium (f. 2v) are three different scenes: a seated
physician holding an apothecary jar; a laborer weighing substances on
a scale; and the physician and assistant laboring together with a mortar
and pestle. The scenes are not unique: either one or two of the three
elements are found in manuscripts produced in Germany and England
later in the twelfth century.20 Since these later manuscripts have no direct
connection with the Berlin volume, one can assume they derive from an
earlier, shared archetype.
Probably also reflecting an earlier archetype is a page showing three
seated physicians, one with no indication of his identity, but the other two
 Medical Books 285
labeled Galen and Hippocrates, the two leading ancient figures of medi-
cine. This falls between the prologue and main text of Theodorus
Priscianus’ Euporista, but is in fact not related to it; Theodorus mentions
neither ancient authority in his prologue. Rather, as with the first quire,
this seems to be a recycled leaf, closing the second quire. The emerging
pattern tells us something. These images do not illustrate the texts in this
particular volume. They constitute a model-book: a collection of sketches
for artists to use in creating new books in the future.21
This particular purpose is seen again towards the end of the volume in
a sequence of images of Plants (in two cases, with human figures). These
come from the Pseudo-Apuleius Complex, a cluster of late antique texts
describing the pharmaceutical properties of various plants and animals,
usually accompanied by images showing the plant or animal in question.
Here in the Berlin manuscript, however, we find only the plant images,
without any text, without even so much as labels to indicate their identity.
These are meant to guide a future artist, not a reader seeking medical
information. Stylistically, their closest parallel is with images in manu-
scripts produced at Monte Cassino.22
The other five full-page illuminations may reflect freehand drawing, and
are of such quality as to suggest we are dealing with a very skilled artist; all
but one of them are framed.23 But why are they here? None can remotely be
called medical or natural philosophical. Again, the volume is functioning
as a model-book. Two images, on the opening page and on f. 146v, have
been largely erased, even though no new images or text are overwritten.
The first, never finished with a defining frame, may have been meant to
depict one of the evangelists.24 Of the three that are clearly legible, one is of
Nimrod carrying his spiked club. The other two are a Crucifixion scene
and a Christ in Majesty. The latter is particularly notable for portraying the
evangelist John as a giant eagle, itself so big that it serves as the throne of
Christ. No comparable image is known.25 Both images would have served
well as models for a sacramentary or lectionary, where these scenes from the
life of Christ would normally be found. Perhaps Nimrod would have
featured in a psalter or biblical history; it is clear that there was some
intention to interpret his story, as he is mentioned in the added passage on
f. 10r telling the story of Babel.
There is not yet sufficient evidence to prove that the Berlin manuscript is
a direct copy of Fleury’s cache of the newest medicine coming out of
Italy.26 But at the very least, the volume must have been made at a center
that was well stocked with new works of that origin. Fleury would have
fitted that definition by 1132. It is also intriguing that the excerpt of Pseudo-
286 monica h. green
Bede De arithmeticis propositionibus adheres in all but one of its readings to
a copy in a tenth-century manuscript associated with Fleury.27 Wherever
its origin, it was not meant to remain at that center. It was made by visitors,
to be taken elsewhere. A general sloppiness, even incomprehension, has
been noted in the scribes’ rendering of the texts.28 Most importantly, this
was not a medical book. The volume’s theme could be described as ‘rules
for ordered living’ for a landowner who also has to function as a military
leader and head of an estate. This unusual volume captures a moment of
transition, when many new medical texts were on offer but what to do with
them was as yet unclear.

Salerno, Stage 3: Teaching the Practice of Medicine

What to do with a medical book, or at least certain medical books, was not
in doubt for Johannes, our medical book owner from the end of the
century. His extant volume was made in the same quarter-century as the
Berlin manuscript. The two books could not be more different in char-
acter, however, both physically and in content. Johannes’ volume has
ample room for glosses, as do many other copies of the teaching collection,
the Articella. Indeed, the wide margins in one of our earliest copies of the
Articella – BnF lat. 7102 from the early twelfth century – suggest that
glossing traditions may have begun with this consummate teaching collec-
tion from the moment it was first assembled.
Salerno had been a center of medical activity in the later eleventh
century, when one of the Lombard princes was developing new compound
medicines. However, the tradition of formal medical instruction and, more
importantly, composition of medical works, with which Salerno’s name is
now most commonly associated, did not start until the early twelfth
century – the only exceptions being the medical editor from the
mid eleventh century, Gariopontus, and the medical enthusiast,
Archbishop Alfanus I. The works coming directly out of Salerno from the
mid twelfth century on are found in multiple versions and with no
distinctive codicological features that might stamp them as ‘Salernitan’.
Whatever books itinerant students commissioned – or made for themselves –
they would have taken home with them.29 A telling example of such
transcontinental production is Bodl. Libr. Digby 79, a tripartite volume of
212 leaves whose first and third sections seem to have been made in
England, but whose middle section bears all signs of having been produced
in Italy. A collection of pharmaceutics, gynecology, cosmetics, general
therapeutics, and surgery, the volume has everything – everything but
 Medical Books 287
elegance or organization. It is a working physician’s book, with little visual
signaling between texts and the many layers of marginal notes to lessen the
inherent difficulty of finding material.
Many of those notes are merely navigational, flagging the topic of
a recipe or a disease description rather than adding new content. Textual
navigation had, of course, driven the first stage of work at Monte Cassino.
The alphabetized texts (the Antidotarium magnum and the Dioscorides
alphabeticus) organized vast amounts of pharmaceutical knowledge, while
the provision of book divisions and tables of contents in the larger books of
practica (Gariopontus’ Passionarius, Constantine’s Pantegni and Viaticum)
made finding content easier. That such paratextual aids were not yet the
norm is indicated by early copies of, for example, the Practica of the early
twelfth-century Salernitan author Copho and his contemporary, the
female author Trota.
A major innovation of the mid-century Salernitan writers was to
recognize that even those early organizational features were insufficient.
The massive Cassinese Antidotarium magnum, for example, comprised
more than 1,300 individual recipes. But such comprehensiveness was
a disadvantage when one had to identify effective remedies for particular
conditions. Already, in the second quarter of the twelfth century,
Nicholas of Salerno had abbreviated the Cassinese compendium.
Retaining the alphabetical organization, he extracted just fifty-five of
the medicinales usuales, the most commonly used recipes; he then added
six new recipes (several of which had been formulated in Salerno itself)
and expanded the original text by adding explanations of the drugs’
properties. The Dioscorides alphabeticus would likewise be replaced by
a shorter Salernitan counterpart, the Circa instans. Both massive
Cassinese compendia would cease being copied by the early thirteenth
century, victims of a transition to new, more concise texts.
As noted earlier, medical books participated in most of the major
codicological changes that became widespread in the later twelfth century:
the regular use of two columns, the addition of running headers, the
development of the initial as a space for decoration and even elaboration
of the text’s agenda. Illuminated initials show physicians, not simply
occupied with the task of healing, but teaching. A good example is that
colossus of Salernitan medicine, the destroyed Codex Salernitanus, a 225-
leaf compendium of about forty different texts composed at the southern
Italian medical center. The codex seems to have been executed in northern
France, and one of its historiated initials showed a master teaching three
students; another showed a seated master, with his scroll, gesturing as if
288 monica h. green
teaching.30 The initials in Bethesda (Maryland), National Library of
Medicine E 78, a copy of the Articella illustrated by an artist known to
have worked ca. 1210–30 in Paris or Oxford, shows the professional
physician reading in isolation, teaching a student, consulting at the
patient’s bedside, and diagnosing urines. In all of these scenes, a book is
present.

A Discipline (and Its Books) Matures

Neither Monte Cassino nor Salerno served as a writing center for the
continued physical production of books. Texts emerged from these centers
but did so haphazardly. The Beneventan script, so distinctive of the region
of origin, can be found in only a small number of volumes, and does not
seem to have left any permanent trace on this corpus. The medical corpus
as a whole shows the same transitions from Caroline to an emerging Gothic
that we see in other areas of book culture.
Perhaps the most vital witness to the success of the new Latin medical
curriculum coming out of southern Italy was its adoption by an anon-
ymous Hebrew translator in southern France, whom we know only by the
demeaning eponym he assigned himself, ‘Doeg the Edomite’.31 Of the
twenty-six texts owned by Johannes, Doeg translated at least fifteen into
Hebrew. In addition, he translated Constantine’s Pantegni, Isaac Israeli’s
books on urines and fevers, three texts on women’s medicine, and more.
No contemporary copies of Doeg’s works survive, so we cannot know how
the earliest Hebrew manuscripts compared in format or layout to the Latin
originals. But the translations themselves are evidence that, at least in
southern France where Doeg lived, the new Latin medical corpus of the
twelfth century had become as much a key to professional success for
Jewish physicians as it had for Christian ones.
The profession of medicine in Europe was a work in progress in the
twelfth century, still defining its major philosophical precepts, its main
authorities, its essential pedagogical methods, and its social identity. What
was fundamentally in flux in the twelfth century would be largely fixed by
the thirteenth. And a measure of that assured status can be found in
medical books. Not all new medical works of the period enjoyed the
success of Constantine or the Salernitans. Both Stephen of Pisa (working
in Antioch) and Gerard of Cremona (working in Toledo), areas recently
retaken from Muslim control by Christians, were, in comparison with the
southern Italian texts, failures: only five copies of Stephen’s retranslation of
al-Majūsī’s great Arabic encyclopedia survive from this period, compared
 Medical Books 289
to fifty of Constantine’s; of Gerard’s twenty-one different medical transla-
tions, most do not appear even once in our corpus. But one would have no
idea of their poor reception from looking at the earliest extant manuscripts
of their works. The earliest copy of Stephen’s Liber regalis is not found
anywhere near the Crusader States but in far-off England, having probably
been copied at the place it now resides, Worcester Cathedral. This is
a large, elegant volume: red, blue, green, and buff initials alternate, some
of which are arabesqued. Even more visually stunning is the earliest known
copy of al-Razi’s Liber ad Almansorem, which employs gold leaf to fill out
the background of all the initials opening the ten books of the text; double
columns, alternating blue and red chapter initials, and numbered lists of
chapters at the head of each book make the volume easily navigable. This,
too, now resides at Worcester, but was probably made in southern
France.32 These are recognizable as ‘university books’ even before there
were university faculties in medicine, and they move us towards the very
different world of high scholasticism of the thirteenth century. While
Stephen of Antioch’s work never attracted attention beyond linguists
trying to better understand Arabic terminology, the Toledan corpus
would emerge from obscurity and become the foundation for university
training in medicine in the mid thirteenth century. Copies of the defining
work of medical learning, Ibn Sina’s Canon, on occasion reached heights of
elegance that have rarely been equaled since; the extraordinary illuminated
Canon from Paris ca. 1260 (now in Besançon), for example, is without
peer.33 And Gerard’s translation of al-Zahrawi’s Chirurgia (Surgery), the
only medical text rendered into Latin with its original illustrations intact,
comes at last out of hiding in the mid thirteenth century and similarly takes
a place of dominance in its field.34
Elegant books for elite physicians are joined by equally elegant books for
the (elite) layperson. Whereas the Norman kings of Sicily in the twelfth
century are known to have patronized other areas of Islamic science at their
new court at Palermo, including astronomy and, most famously, geogra-
phy, there is no evidence of such patronage of medical writing. Under the
Hohenstaufens, however, we see the last revival of the Pseudo-Apuleius
Complex. Two extraordinary manuscripts – ‘twins’, Giulia Orofino calls
them – were prepared, probably for Frederick II and his son Manfred, in
the second and third quarters of the thirteenth century. The text of Pseudo-
Apuleius’ Herbarium, as well as the other works that made up that corpus,
become the scaffolding for full-page illustrations of clinical encounters,
scenes of herb-gathering and a variety of other scenarios related to the
delivery of health care.35 In the decades after mid-century, we see lavishly
290 monica h. green
illustrated regimens of health, like the French Régime du corps, said to have
been made by Aldobrandino of Siena for the countess of Provence, Beatrix of
Savoy, or, replacing the illustrated Pseudo-Apuleius tradition, a new kind of
illustrated herbal developing out of the Salernitan Circa instans.36
Such volumes were not meant for regular use, of course, whether
ponderous study or clinical consultation. For those essential tasks of the
medical art, humbler books – still looking much as they had in the twelfth
century – would continue to have a role. With the introduction of the pecia
system towards the end of the thirteenth century, standardization of
university medical books reached a new level. But again, this innovation
of mass production was a function of the university context as a whole and
in no way unique to medicine. Looked at from the perspective of its
development over the course of the Long Twelfth Century, it could be
ventured that adoption of regularized forms of book production, and
hence the normalizing of medical knowledge, was one of the ways medi-
cine itself became normalized as a learned discipline.

Notes
This chapter draws on materials collected over the past several years by myself and
several colleagues who have styled ourselves informally the Medical Paleography
Team: Winston Black, Florence Eliza Glaze, Erik Kwakkel, Brian Long, Francis
Newton, and Iolanda Ventura. My thanks to them all. The Project has been
unfunded aside from a small grant in 2010 from the National Humanities Center.
Other materials and information have been supplied by Charles Burnett, Klaus-
Dietrich Fischer, Outi Kaltio, Valerie Knight, Outi Merisalo, and Rod Thomson.
Background research was done during a fellowship I held in 2013–4 at the
Institute for Advanced Study, Princeton, NJ, which was supported by funds
from the U.S. National Endowment for the Humanities and the Willis
F. Doney Membership Endowment. Any views, findings, conclusions, or recom-
mendations expressed in this publication do not necessarily reflect those of the
National Endowment for the Humanities.
1. BAV Vat. lat. 10281 is a s. xii2/4 copy of Johannitius, Isagoge. Johannes’ list of
books is added in a s. xii4/4 (Italian?) hand on f. 41r, and opens with the claim
‘Ego Johannes habeo libros istos phisicales’. My thanks to Erik Kwakkel for
these dating estimates.
2. On Salomon, see Barker-Benfield 2008, 1860–1. The extant volume is Bodl. Libr.
Auct. F. 6. 3, s. xiiiin. On Herebertus, my thanks to James Willoughby for sharing
with me the draft of his forthcoming edition of the Durham Priory catalogues.
In the meantime, Herebertus’ booklist is printed in Catalogi Veteres Librorum
Ecclesiae Cathedralis Dunelm, ed. B. Botfield (Surtees Soc. 7, 1838), 7–8.
3. See Acknowledgements preceding n. 1.
 Medical Books 291
4. Kwakkel in Kwakkel and Newton, in press. Of about eighty such books
Kwakkel surveys, two (2.5 per cent) are medical. Several more could be added
to that list, such as BSB Clm 4622, parts VII–VIII (olim part VI) (ratio: 0.54);
even so, they make up only just over 1 per cent of the medical corpus.
5. Reynolds 1983, xxxiii et passim.
6. Manzanero Cano 1996; Knight 2015.
7. Kwakkel and Newton, in press.
8. Copenhagen, Konigl. Bibl., Gamle Kgl. Samling 1653 (1060s or 1070s).
9. Beccaria 1956, 85.
10. Pradel-Baquerre 2013, 98–9; cf. Ferraces Rodríguez 2013.
11. Langslow 2006, 102.
12. Newton 1994; Kwakkel and Newton, in press.
13. Galdi 2014.
14. Hoffmann 1980, 494–5.
15. Leiden, Universiteitsbibl. VLF 85, part 1 (ff. 1–24), s. xiex (Italy?). On the
connection of the MS to Fleury, see Mostert 1989, 95 (BF313).
16. The extant manuscripts I believe should be associated with early twelfth-century
Fleury are: Leiden, Universiteitsbibl. VLF 85, part 1 (ff. 1–24); Orléans, Bibl.
mun. 283 (olim 236), parts 1–2, 285 (olim 239) and 286 (olim 240), part 3, and 301
(olim 254), parts 4–9; and BnF, nouv. acq. lat. 1628. It is possible that Bern,
Burgerbibl. 337, s. xi, should also be added to this list. Of these, only Orléans 283,
part 1, seems to have been copied from an exemplar coming from an older French
tradition; see Langslow 2006, 47–8 and 102.
17. My thanks to Klaus-Dietrich Fischer for identifying the Vegetius text, and to
Vincenzo Ortoleva for information on the Berlin manuscript’s place in the
textual tradition.
18. Reeve 2000, 270.
19. My thanks to Charles Burnett for further information on the version of the
text found here in the Berlin manuscript. The fullest description I have found
of the textual tradition of the Kitāb al-anwā’ is the online Filaha Texts Project.
20. These are Eton Coll. 204, s. xiimed (Germany), f. 1v; BL Harl. 1585, s. xii3/4
(Meuse Valley), f. 7v; and BL Sloane 1975, s. xiiex (northern England?; later
owned by the Cistercian house of Ourscamp, near Noyon), f. 97r.
21. Scheller 1995.
22. D’Aronco 1998.
23. That a single artist was responsible for all the images in the manuscript is
suggested, for example, by the identity of the face of ‘Hippocrates’ on f. 5v and
‘Adam’ arising from his grave at the foot of the Cross (f. 154v).
24. Cahn, Romanesque Manuscripts, 63.
25. My thanks to Karen Reeds for contacting Adelaide Bennett Hagens at the
Index of Christian Art, who confirmed for me that their files record no
comparable image in Western art.
26. My thanks to Christopher Crockett for first suggesting a possible Fleury
connection.
27. Folkerts 1972, 40, ll. 91–107.
292 monica h. green
28. Burnett (in Bos 1994), 225. Schipke calls the entire volume ‘fehlerhaft’.
29. On the rapid movement of Salernitan texts to England, see Green 2008.
30. Wrocław, Stadtbibl. 1302, s. xiiex/xiiiin (N. France?). It was destroyed in
World War II. Reproductions of the initials can be found in Sudhoff 1920.
31. Freudenthal 2013.
32. Worcester Cath. F. 40 (s. xiimed, perhaps produced at Worcester by scribes
trained in France); and Q. 60 (s. xii2, S. France; at Worcester by s. xiiiex). See
Thomson, Worcester, 25–6, 157.
33. Stones 2014.
34. Green 2011.
35. These manuscripts are Florence, Bibl. Laurenz. Plut. 73.16, and ÖNB 93. See
Orofino 2015.
36. Ventura 2009.
 chapter 16

Law Books
Charles M. Radding

By the end of the twelfth century, law books, whether of secular law or
canon law, had arrived at the familiar format of university books, with
(usually) two columns of text, wide margins to accommodate glosses, and
the use of multiple colours of ink to facilitate locating the internal divisions
between laws.1 Yet the eventual convergence of these manuscript formats
should not disguise the very real differences in how they got there. For
secular law, at least in Italy where the most important developments took
place, lay legal professionals were the principal readers and often even the
copyists of legal manuscripts.2 Manuscripts of secular law were also pre-
cocious in making the transition towards page layouts intended for gloss-
ing and study, with a systematic gloss and a format devised to facilitate its
use already in existence for the Lombard law by the third quarter of the
eleventh century. These innovations in format carried over to Roman law
during the early twelfth century, basically as the Justinianic texts acquired
enough glosses to justify the effort. The production and use of canon law
manuscripts, in contrast, was the province of ecclesiastical communities –
monasteries or chapters – and Gratian’s Decretum, the first university
textbook, did not even take its final shape until nearly 1150.
In this chapter I consider Lombard and Roman law on the one hand and
canon law on the other. The inclusion of Roman and canon law requires
no justification, and for Lombard law it should be enough to note that
jurists of the Lombard law were not only the first medieval readers of the
Justinianic codification but also copyists of many of its earliest manu-
scripts. Lombard law is also the only set of Germanic laws to retain
a significant place in manuscript production during the period covered
by this volume.3 Salic law – the obvious candidate among other Germanic
legal traditions – suffered a precipitous decline in circulation, at least to
judge from surviving manuscripts: McKitterick lists only five manuscripts
containing Salic law for the entire period between the late tenth and early
thirteenth centuries, and in one of these (BnF lat. 9656, discussed later in
293
294 charles m. radding
this chapter) the main text, occupying the first 108 of 115 folios, is in fact
a glossed version of the Lombard law.4 Anglo-Saxon law, of course, lost
much of its relevance with the Norman Conquest and the decline of the
Old English in which the laws were written. Nor do we see the develop-
ment of glosses and commentaries around Salic, Anglo-Saxon, or any other
Germanic law except that of the Lombards.
For each of these legal traditions, I look at the process by which the
text was stabilized, the evolution of the format to accommodate an appa-
ratus, and the shift of production towards serving university study. Since
laws are legislative acts rather than free compositions, it would seem that
the content of law codes was settled before redactors or copyists began their
work. Yet with legislative histories measured in centuries, that approach
was not always practical. Laws accumulate over time, sometimes super-
seding or modifying existing legislation and always eventually resulting in
a quantity of legislation too large for easy consultation. Adding to the
complexity of legal compilations is the fact that laws tend to be fairly short
and discontinuous from one to another, so they can be added into existing
compilations, deleted, or simply moved around in a way that could not be
done with texts of Augustine or Aristotle. For Roman law the work of
codification had been accomplished by a commission of trained jurists
appointed by Emperor Justinian; Justinian also used his own legislative
authority to fill gaps in existing legislation collected in the Codex
Justinianus and to endow the Institutes and Digest with the force of law.
Transmission of those texts in our period was far from simple, but the texts
themselves had at least been defined. Lombard and canon law, in contrast,
were still at the beginning of the process of establishing definitive collec-
tions of the laws in force, and that process is part of the story of law books
from the late eleventh to the late twelfth century.
For secular law, the most distinctive aspect of eleventh- and twelfth-
century manuscripts is the important role of laymen – the judges and
notaries sacri palatii of the north Italian kingdom – not only as readers of
Lombard and Roman law manuscripts but also as their redactors and even
copyists. It cannot be entirely surprising to find legal professionals prepar-
ing their own books, for they not only knew how to write but did so
routinely as part of their work; they were, as Petrucci aptly called them,
‘professionals of the pen’.5 The fact that the graphic preparation of such
scribes lay in documentary rather than book culture certainly influenced
the appearance of the books they produced, for their hands are often less
regular and there is much more variation in graphic formation between
different copyists in the same volume than one sees in books produced by
 Law Books 295
an organized scriptorium.6 (Documentary traits in their scripts also con-
tributed to errors in dating manuscripts that had important consequences
for the historiography of medieval law by making certain key manuscripts
seem much older than they are.) But the experience of dealing intellectually
with the texts they were copying, and of working as part of a professional
community, also seems to have freed them – or pushed them – to experi-
ment with the structure and content of their books to maximize the ease
with which they could be used. Manuscripts of Lombard law in particular
were precocious in reserving space on the page for glosses, producing the
manuscripts of the Walcausina (discussed later in this chapter) with stan-
dardized glosses by the 1070s, several generations before Gratian’s
Decretum became the first book of canon law to attract substantial glosses.

Lombard and Roman Law

At the beginning of the eleventh century, the legal inheritance of a north
Italian jurist consisted of two distinct categories of legal texts. The leges
Langobardorum had a fairly well-defined manuscript tradition consisting of
Rothari’s Edictum from 643 and the legislation of four of his successors,
some of which – notably the laws of Liutprand – were subdivided into
blocks according to the year in which the laws were enacted. After the
conquest of the Lombard kingdom by Charlemagne in 774, legislation in
the form of capitularies was issued by the Frankish kings and emperors
controlling northern Italy and eventually by their Ottonian and Salian
successors. Capitularies numbering in the hundreds were issued over the
years, without any official collection of them ever being made; as a result,
the codices preserving them varied widely in their content, with none of
them coming close to containing them all. Some important capitula,
indeed, are found in no collections earlier than the eleventh century, and
must have survived as separate documents in archives, most probably the
one in Pavia. The Lombard leges and post-Lombard capitularies thus
generally had rather distinct manuscript traditions, even though a few
manuscripts do combine both leges and capitularia, for example BnF lat.
4613, a manuscript from around the year 1000.7
In the eleventh century these separate traditions were supplanted by
a new kind of law book that combined Lombard laws and post-Lombard
capitularies into a single compilation. This work, published by the MGH
as the Liber Papiensis in recognition of the role in compiling it of the jurists
based in the old Lombard capital of Pavia,8 simplified the presentation of
the laws by grouping the legislation of each king together, so that
296 charles m. radding
Liutprand’s legislation now formed a single long block of 151 laws. Post-
Lombard capitula similarly were presented as blocks of laws attributed to
successive rulers (Charlemagne, Pippin, Louis the Pious, etc.). Unlike their
approach to the Lombard laws, however, all of which they retained, the
compilers were selective in what they included from the capitularies,
excluding capitularies and even individual capitula within larger capitul-
aries that were without significance for Italian legal practice. The redactors
omitted or took only a small selection of some common capitularies –
fewer than ten of the eighty-two capitula of the Admonitio Generalis made
it into the Liber Papiensis – even as they included other capitula, such as
those of Wido, that are almost unknown except through these
manuscripts.
In practice, the Liber Papiensis was more an idea or model of a text than
a compilation created once and then copied through successive generations
of manuscripts. Even the section devoted to the Lombard Edictum varies so
much from one manuscript to another that scribes must have used what-
ever manuscript of the leges came to hand rather than a common exemplar.
And the variations are still more marked with the capitulare, for the
selection of capitularies was never entirely fixed and no two manuscripts
arrange the capitulary section identically, with the exception of two codices
sharing a common exemplar to be discussed in what follows. Capitula were
moved around, sometimes from one emperor to another, other capitula are
found in only one or two manuscripts, and still others were added only
later, by secondary hands not involved in the original copying. Unlike what
we see in some of the canon law materials, none of these inclusions seems
politically motivated; they may simply reflect the fact that Carolingian and
post-Carolingian legislation was never authoritatively collected.9
This process can be illustrated by the oldest surviving copy of the Liber
Papiensis, Milan, Bibl. Ambrosiana O. 53 /O. 55 sup. (the laws and
capitularies were bound separately though sharing a common format and
some copyists), which is also one of the manuscripts apparently written by
legal professionals. The manuscript itself is datable to ca. 1030 by the
termination of a king list with the reign of Conrad II (1024–38), though
the scripts in which the volumes are written seem much older, from the
ninth or tenth century. (Bischoff’s notes show that he initially dated it to
the tenth century, correcting to the eleventh, one supposes, when he
reached the king list.) The reason for this anomaly is not hard to find,
for the last scribe identified himself at the explicit of the capitularies:
Secundus notarius scripsit oc manus suas.10 Documentary culture had simply
preserved features that book culture had abandoned, and scripts similar to
 Law Books 297
those in this manuscript can be seen elsewhere in the writing of eleventh-
century jurists, notably that of the judge Ambrosius who signed an addi-
tion to a mid-eleventh-century manuscript of the Epitome Juliani.11
The glosses in the Milan manuscript show that its readers were fully
aware of the variations between manuscripts. One gloss notes that
a capitulum of Charlemagne is out of place: Istud cap. [h]ic non debet
esse; the author of the gloss is referring to the fact that most manuscripts
(mistakenly) attribute this capitulum to Lothar, assigning to
Charlemagne a capitulum of Lothar having an identical incipit. Another
gloss notes of a capitulum that it is not found in other manuscripts (non
sunt in ceteris).12 Evidently this manuscript had been collated with others,
a common enough occurrence at the time; copyists of manuscripts often
noted variant readings for certain words – variants presumably already
noted in the exemplar from which they were working. In an ambitious
extension of this practice, eleventh-century glosses occasionally try to
settle debates about which readings are authentic by noting the readings
found in other manuscripts, some of which are described as antiqua or
antiquissima.13
By the third quarter of the eleventh century the proliferation of variants
led to a precocious effort to standardize the text into a kind of critical
edition. This project was attributed to the judge Walcausus or Gualcausus
of Pavia, who was described as making ‘the pages now speak the true sense
of the Edict’ (verum loquitur nunc pagina sensum Edicti). Not content to
systematize the selection of capitularies, Walcausus also intervened in the
language of the laws, even emending some passages where necessary to
arrive at a good legal interpretation. Similar emendations, based on legal or
linguistic logic rather than manuscript authority, are also discussed in
glosses from the period.14 This ‘critical edition’, the Walcausina, survives
in two manuscripts which not only share an exemplar but also have
a copyist in common. And we can identify that copyist: he was the
Iohannes notarius sacri palatii who prepared a document in Pavia in 1070.
The earlier of these two manuscripts, now BnF lat. 9656 (=P)
(Figure 16.1), has seven principal copyists, and while most of them copied
sections corresponding to the gatherings of the manuscript, two began
mid-page and copied only a few folios. Though all wrote Caroline
minuscule, the appearance of the different scripts varies considerably as
do the abbreviations they used. Adding to the visual cacophony, finally, is
the presence of several additional hands copying marginal glosses – an
essential and original part of the recension. Since Iohannes was a notary it
would be reasonable to assume the other copyists are similarly connected to
298 charles m. radding

Figure 16.1 The Walcausina. Paris, Bibliothèque nationale de France, lat. 9656, f. 13r.
 Law Books 299
the legal community; could the more awkward writers have been serving
some kind of apprenticeship?15
Both surviving manuscripts of the Walcausina are somewhat oblong
with essentially identical dimensions: the Paris manuscript is 256 mm ×
167 mm, while the second copy, ÖNB 471 (=V), is 255 mm × 165 mm.
Marginalia were an essential part of the text of the Walcausina and were, for
the most part, copied by the scribes responsible for the main text, often
employing an identical tie-sign for the same reference. Given the impor-
tance of the marginalia – which was used for cross-references, diagram-
matic distinctions, clarifications of the main text, and general comments –
the page layout was planned with them in mind. In V, for example, the
often busy outer margin was slightly more than half the width of the main
text.16 The dimensions seen in other manuscripts from the late eleventh
and early twelfth centuries vary, of course, but all reserve ample space for
glosses and the surviving manuscripts are all glossed.
The last step in the development of the Lombardist textual tradition was
taken in the early twelfth century when the legislation already collected in
the Liber Papiensis was completely restructured according to topic. In fact,
this was done twice, for we have two different versions of the Lombarda,
both dividing the laws into three books subdivided into titles. Nor is this
the first effort to impose a topical order on the Lombardic laws, because
several manuscripts related to the Walcausina tradition contain elaborate
diagrams grouping laws – though not all of them – according to categories
borrowed from Roman law and other abstract criteria such as contractus
(subdivided into consensu, re, verbis, litteris, permutatio), actio furti, and so
forth. It was the Lombarda, however, that eventually won out, and one
does not find copies of the Liber Papiensis copied after the early twelfth
century; indeed, an early twelfth-century hand worked through the entire
Paris copy of the Walcausina inserting cross-references to the organization
of the Lombarda. The transition to the Lombarda did not, however, mean
the abandonment of the analytical work done in the eleventh century, as
the substance of older glosses was often preserved in the glosses to the
Lombarda.17
The involvement of legal professionals also affected the transmis-
sion of Roman law.18 At first glance, the works of Justinian’s codifica-
tion – the Institutes, Code, and Digest – should have been immune
from the kind of interventions that produced the Liber Papiensis and
the Lombarda: the work of codification had, after all, already been
done by Justinian’s commission. Yet the key texts for jurists, the Code
and the Digest, were far longer than the Liber Papiensis. The Code is
300 charles m. radding
about 420,000 words compared to 60,000 for the Lombard law, and
the Digest is nearly a million words. (The Institutes, more manageable
at about 50,000 words, seems to have been copied without strain, but
it was also less useful for jurists interested in the details of
procedures.)19 For scribes working independently of any institutional
context, as they were, for example, in producing the Walcausina,
copying such texts posed challenges for the organization of work
that went beyond anything they faced with even the Walcausina,
complicated as it was.
For the Code the strategy adopted was to take extracts from the entire
work and circulate these separately in a work known to historians as the
Epitome Codicis. The name is something of a misnomer, since the extracts
from the Code were all constitutions taken in their entirety, without the
simplification or abbreviation implied in an epitome. In fact, the earliest
versions of the Epitome Codicis are invaluable for just this reason, since they
contain legislative details – the inscriptions and subscriptions – that scribes
had learned to omit by the time the first complete manuscripts of the Code
were copied. The process of taking constitutions from the Code was not,
however, a one-time event, but must have been repeated (as we shall see)
many times – so often, indeed, that we can be certain that these early
students of Justinian’s Code knew where to find an intact and presumably
ancient exemplar.
The manuscript that tells much of the story is Pistoia, Bibl. Capitulare,
C 106.20 In the nineteenth century, this heavily glossed manuscript was
attributed to the tenth century, creating the impression that the Epitome
Codicis not only existed then, but was the subject of intense study.
An eleventh-century date, most probably in the middle or third quarter
of the century, is confirmed by the use of devices such as the Tironian note
for et, the abbreviation q2 for quia, and a superscript s at the ends of words.
It is, indeed, not hard to see where the error in dating came from, since
some hands preserved features (such as an a open at the top) that one would
not expect to find in eleventh-century books. Yet such archaic forms
persisted in eleventh-century documents even when they were no longer
used when copying books, and it is likely that some of the copyists of this
book came to this work from the world of documents. Also pointing to an
origin outside an institutional scriptorium is the wide variation in the
appearance of the hands, and a rather casual organization of the copying,
with one scribe of eight copying barely more than a page, another only nine
lines, and another only three lines out of 170 leaves in the manuscript as it
exists today.
 Law Books 301
The work of gathering extracts for the Epitome was done not once but
several times. The original text in the Pistoia manuscript represents at least
one and probably several strata of insertions into the original form of the
Epitome, and that process continued after the manuscript was complete: 138
constitutions were added to the manuscript, mainly in the margins but also
on a slip of parchment, by twenty-three different hands, all of them
working before 1100. (Other hands that contributed glosses bring the
total active in the manuscript to forty-three, including the original
eight.) As with the manuscripts of the Liber Papiensis, we catch a glimpse
of a community of scholars for whom collation of manuscripts was
a routine activity. Other manuscripts of the Epitome had, not surprisingly,
somewhat different constitutions than those found in the Pistoia manu-
script, with the exception of BnF lat. 4516, which was copied from the
Pistoia manuscript (and by the scribe of that manuscript’s first leaf).21
By the end of the eleventh century the leap had been made to restoring
all of the Latin-language constitutions to the Code, but this was not done
by simply copying the ancient codex (or codices) available. Instead, and
perhaps because numerous manuscripts of the Epitome were already in
circulation, the missing constitutions were reintegrated into the Epitome,
inadvertently preserving some of the errors of organization that had devel-
oped over the previous decades. (A fragmentary manuscript now Pesaro,
Bibl. Oliveriana 26, suggests how this may have been done, for it consists of
a copy of the Epitome and a supplement containing the missing constitu-
tions both – surprisingly! – in the same hand.)22 One of the earliest
manuscripts of the restored Code, Berlin, Staatsbibl. Lat. fol. 273 from
the end of the eleventh century, seems to have been produced in informal
conditions similar to some of the manuscripts we have already discussed.
A single scribe predominated, but his script is so irregular that it is hard to
imagine that he was trained as a copyist for books. Twelve other hands also
participated, often for very brief passages of less than a folio, even less than
a page. Occasionally constitutions were copied in the margin rather than
the main text, but always by the original scribe: a sign, most likely, that he
was integrating two texts – the Epitome and a supplement – even as he
worked.23
Traditions of instruction eventually shortened Justinian’s Code to nine
books, with books nine to twelve circulating separately and being taught as
the Tres Libri. It is hard to tell when this decision took hold because it was
worked back into pre-existing manuscripts most of which – including all
the earliest manuscripts of the Epitome and Berlin 273 – are truncated at the
end. (A scholar who knew only these manuscripts would conclude that
302 charles m. radding
medieval codices routinely lost their later gatherings!) The manuscript
tradition of Justinian’s Digest was similarly subject to arbitrary divisions,
in this case into three parts – the Vetus, Infortiatum, and Novum – that do
not even break at the end of titles, much less at the end of books. The only
clue to how this happened is that the single copyist responsible for the
oldest manuscript of the Digestum Vetus, BAV Vat. lat. 1406, did his
copying in four distinct blocks, from back to front:24 he must have been
working with loose gatherings rather than a complete book, and it may be
that the fascicle that should have carried through to the end of book
twenty-five just got lost for a while! Lacking in logic as they were, and
despite some variations in the earliest manuscripts, the divisions of the
Digest nonetheless became conventional both for manuscript production
and for university instruction.
A final trend to notice is towards larger manuscripts such as could
comfortably be read only at a desk. Berlin 273, mentioned earlier, is still
comparatively portable at 240 mm × 190 mm, despite adopting a two-
column format. Yet other early Code manuscripts, Berlin, Staatsbibl. Lat.
fol. 272 and Montpellier, Bibl. Universitaire, Section de Médicine H 82,
retained a full-page format but were much larger: 325 mm × 210 mm and
315 mm × 204 mm, respectively. By mid-century the two-column format
and larger size, with ample margins to accommodate glosses, would
become the norm for manuscripts of both the Code and the Digest
(Figure 16.2).
Even then some habits survived from the earlier legal scholars. Some
masters of Roman law seem to have kept their manuscripts unbound for
ease in handling,25 a practice also perhaps witnessed in the Paris manuscript
of the Walcausina. One also sees a continuation of the practice of emending
legal texts by conjecture. The correct readings for certain passages in Code
and Digest were also debated throughout the twelfth century, as those in
the Lombard law had been in the eleventh, leaving a body of conjectures in
glosses collected by Savigny. Interestingly, reported conjectures often take
the form of ‘in the liber of Martinus (or Rogerius or Azo)’,26 a formulation
that gives the impression the books themselves could be consulted.

Canon Law
Shifting our attention from secular to canon law is to enter a different
world. Canon law manuscripts give the impression of having been pro-
duced by well-organized scriptoria, as indeed one would expect for works
used principally by ecclesiastical communities: not only is the parchment
 Law Books 303

Figure 16.2 Justinian’s Code, two-column format with ample margins to

accommodate glosses. Leiden, Universiteitsbibliotheek, d’Ablaing 1, f. 105r.
304 charles m. radding
usually of reasonable quality but the various hands within the volume are
usually fairly homogeneous in appearance. No less striking, though, are the
differences in the quantities that survive. Whereas we have six manuscripts
of the Liber Papiensis and only three of the Epitome Codicis, Burchard of
Worms’ early eleventh-century Decretum, which remained the most widely
used source of canon law into the twelfth century, survives in eighty
complete manuscripts as well as another forty or so fragments; there are,
similarly, about 160 surviving twelfth-century manuscripts of Gratian’s
Decretum – a text that did not even arrive at its final form until around
1150.27 The differences in quantity have perhaps as much to do with where
the manuscripts were used as how many were produced. Manuscripts of
secular law, as we have seen, were predominately copied by and presumably
for lay professionals, most of whom would have had no permanent
institutional base. Manuscripts of the Decretum, in contrast, and of other
books of canon law from both before and after the appearance of uni-
versities, must often have ended up in episcopal or monastic libraries where
they could survive intact for centuries after they had ceased to be of
intellectual or practical interest.
The comparative ease with which books of canon law could be produced
did not, however, accelerate the process of arriving at a stable and definitive
set of texts. One issue, surely, was the absence of a specialized intellectual
community such as worked with Lombard and Roman law. Italian jurists
had to deal with one another’s views on an ongoing basis both by sitting
together in courts and through debates on the proper interpretation of the
law, many of which come down to us as glosses. Knowledge about canon
law, in contrast, though widely diffused in the sense that every diocese
needed some expertise, until the late twelfth century lacked any function-
ing judicial hierarchy to bring together experts from different dioceses or
provinces in a context where they had to confront their differences.
Another factor was the comparative difficulty in even knowing which
canons were in force. For secular law there was general agreement that
only a king or an emperor could make a law, so the body of law was finite
and defined; the Liber Papiensis, accordingly, simply collected all relevant
laws, even those that subsequent legislation had annulled or superseded.
But the situation for canon law was not so straightforward. For some issues,
especially those touching theology, texts excerpted from the Bible or the
writings of the Fathers could be treated as canons. Other questions, such as
those concerning ecclesiastical organization, pastoral care, and penance,
might more easily be settled by the legislation of church councils, but no
complete collection existed of these texts, which included provincial
 Law Books 305
synods as well as more general councils ranging over 700 years of history.
Papal writings were another potential source, though the emphasis was on
the early popes; it was not until the later twelfth century that papal letters
would become a primary source of ecclesiastical law. Completing the
round-up of potential sources of canon law are various forgeries such as
the pseudo-Isidorean materials that had won an exceptionally wide dis-
tribution by the eleventh century.
In practice, compilers did not systematically attempt to gather all these
sources but relied on earlier canon law collections. For example, Burchard’s
Decretum was mainly based on the early tenth-century Libri duo of Regino
of Prüm and the late ninth-century Collectio Anselmo dedicata, with other
canons taken from a few other collections; then, into the twelfth century,
Burchard’s work itself became a source for subsequent compilers. Anselm
of Lucca, author of the most important compilation of the Investiture
Contest, drew on a different set of works to develop his arguments
supporting the importance of the papacy (an issue to which Burchard
had been indifferent), including some from which he took extracts of the
pseudo-Isidorean forgeries. Ivo of Chartres’ Decretum, from around 1100,
included almost all of Burchard’s Decretum as well as materials drawn from
other sources, including Roman law.
One of the reasons for the success of Burchard’s Decretum was that its
organization made it comparatively easy for readers to find the material
they needed. Burchard divided his canons into twenty books according to
subject matter; indeed, topical arrangement was the norm for the most
important canon law collections of the period. Though Burchard’s books
were not subdivided by topic into titles, each book did begin with
a numbered list of the canons it contained, further assisting readers looking
for answers to specific questions. The codices produced at Worms used the
margins to repeat the numbering of the canons and to provide a reference
to their sources, while each canon was also provided with a rubric sum-
marizing its contents. Further simplifying use of the Decretum was the fact
that it gave clear answers to questions because unlike the Lombardist
jurists, who had included every law in their compilations, Burchard simply
excluded canons whose doctrines he disagreed with and rewrote others to
avoid contradicting other canons in his collection.28 Combined with
a selection of topics that emphasized practical and pastoral questions,
this ease of use kept Burchard’s Decretum in wide circulation for more
than a century, in Italy and France as well as in Germany.29
The later eleventh century saw the creation of several other collections
oriented towards the issues debated during the push for Church reform.
306 charles m. radding
These never replaced Burchard’s Decretum, however, perhaps because they
were more oriented towards issues of Church governance and the role of
the papacy than the pastoral questions on which bishops and priests might
require guidance. The Collectio Canonum of Anselm of Lucca, for example,
was divided into thirteen books instead of twenty, and the choice of topics
was very different from Burchard’s, with the role of the papacy made the
subject of book one and penance being pushed back to book eleven. Like
Burchard, Anselm chose only materials that agreed with his view of
ecclesiastical organization, and though he did not rewrite canons as
Burchard sometimes did, his rubrics were written to push the reader
towards the interpretation he favoured. In format though, he stayed closer
to Burchard’s model, providing a numbered list of the canons at the
beginning of each book, with the numbers now being incorporated into
the rubrics rather than being placed in the margins.30
It was not until around 1100 that one sees the first compilation which
embraced rather than disguised the diversity and internal conflicts of the
canon law tradition. This was the collection of Bishop Ivo of Chartres, also
known as the Decretum.31 In addition to Burchard, Ivo drew on other late
eleventh-century collections known as the Tripartita and the Collectio
Brittanica (which is not English in origin but survives in a single manu-
script in the British Library). In contrast to eleventh-century compilers,
Ivo’s practice was to include virtually everything contained in his sources –
virtually all of Burchard, for example – and leave it to his readers to sort out
which was most appropriate for a particular case. For Ivo, these apparent
contradictions were not a bug but a feature, and he advised readers who
read things they think are contradictory to ‘diligently consider what
pertains to rigour or moderation, to justice or mercy’.32 Readers who
wanted clearer guidance than Ivo provided gravitated towards the
Panormia, a compilation that used Ivo’s Decretum extensively while
excluding some of the most contradictory canons and sequencing those
included so that the last of a series of canons on, say, baptism could be seen
as resolving the difficulties posed by the others. As a text the Panormia was
enormously successful, surviving in a large number of twelfth-century
manuscripts (and often attributed, erroneously, to Ivo). Its structure was
still very simple, though, with divisions into eight books, each one prefaced
by a list of topics covered but without actual subdivisions in the text and
without a list of the canons contained in the book.33
It was not until the middle of the twelfth century, therefore, that one
arrives at the book that would define the study of canon law in the early
universities. In structure Gratian’s Concordia discordantium canonum was
 Law Books 307
not a canon law collection like Ivo or Burchard, and certainly not like the
Lombarda or the works of Justinian’s Corpus, because his commentary
(dicta) both structures the canons and provides guidance on how to resolve
apparent contradictions between them. The long pars secunda, indeed, is
organized around hypothetical cases devised to illustrate specific issues,
with canons quoted when needed as part of the analysis. It was, in short,
a text for teaching, closer in spirit to the Sententiae of Peter Lombard than
to Burchard or Ivo, and a text whose evolution over time is witnessed by
the two recensions we have of it.34 Gratian himself seems to have died in
1145,35 and the text of the second recension (whether completed by him
alone or others is uncertain) was already being quoted by 1150.
The manuscripts of the first recension already illustrate this new con-
ception of a book of canon law.36 Whereas earlier collections such as
Burchard’s had used rubrics or marginal notes to highlight the original
source of canons or, occasionally, their content, the rubrics in Gratian’s
book are intended to guide the reader through the steps in his exposition.
Though he usually indicates the original author in his text, references to
specific works are often missing and his analysis often pushes towards
conclusions that may not be clearly stated in any of the canons he quotes.
The reader’s attention has thus been shifted from the canons to Gratian’s
own analysis, and Gratian’s dicta are usually marked with a paragraph sign,
making them easier to find.
As Gratian’s book established itself as a teaching text and began to
acquire glosses of its own, its manuscripts began to incorporate further
innovations in format both to facilitate cross-references and to keep it
current. Paucapalea, some of whose glosses are perhaps as early as 1148,
divided the text of the Pars prima into distinctions; Gratian himself had
already divided the second part into causae and distinctions. (Some early
manuscripts incorporated divisions into titles, but this innovation did not
last.) One also sees a process of adding new material known as paleae to the
text, including excerpts from papal decretals issued after the completion of
the Decretum itself. In some cases, these additions were placed at the
appropriate place of the Decretum; in others, the additions took the form
of an appendix placed at the end.
By the end of the twelfth century decretal collections would take on a life
of their own, with separate manuscript traditions. In form, papal decretals
were letters replying to specific queries from local courts about the rules to
be applied in local cases, resembling Roman imperial rescripts more than
the legislation of medieval kings and emperors. Reduced to their essence by
excerpting, however, and circulated beyond their original recipient, they
308 charles m. radding
became statements of Church law that could be analyzed for underlying
principles, with the additional advantage that they addressed precisely the
questions that were least settled by existing canons. The earliest indepen-
dent decretal collections seem to date from the 1170s, with the first sub-
stantial and systematic collection, the Breviarium extravagantium of
Bernard of Pavia, appearing in the 1190s. Bernard organized his collection
topically, into five books subdivided into titles, including both materials
not used by Gratian from earlier popes and more recent materials, in
particular from Alexander III (1159–81). The success of the Breviarium as
a school text can be gauged from the quantity of glosses it attracted: the
copy at St Gallen, for example, shows three-quarters of the page reserved
for the glosses that surround the text on all sides.37 The early thirteenth
century saw the creation of other decretal collections that tracked the
activities of the papacy – all of these, like Bernard’s Breviarium, essentially
private and thus susceptible to additions and variations – until Gregory IX
took control of the process with the release of a collection of backed by the
authority of the papacy.38
By the early thirteenth century, therefore, books of canon law had
arrived at the point reached by Lombard and Roman law nearly
a century earlier, with texts for teaching and practice established by
authority and convention. In Bologna, still the centre for the study of
both canon and civil law, the books would even resemble each other
physically, being large, written in two columns with ample space reserved
for glosses, copied in the same littera bononiensis, and perhaps even pro-
duced by the same stationers (though this subject remains to be studied).
By the early thirteenth century, one also sees at Bologna the emergence of
the pecia system, which stabilized the division of the main legal texts into
brief sections for renting and for copying, often accompanied by the
appropriate glossa ordinaria.39 Once backed by the authority of the uni-
versity, this would remain the primary system for producing school texts
until the end of the Middle Ages.

Notes
1. Legal manuscripts have usually been studied or described as part of the process
of investigating or editing individual texts; but since many of these editions
date from the nineteenth century, often indeed from before the widespread use
of photography permitted the development of a more scientific palaeography,
errors in dating and description are unfortunately rather common. No modern
overviews of legal manuscripts as a category exists, but the essays collected by
 Law Books 309
Colli 2002, though specialized, will serve to give a sense of the kind of work
being done.
2. See at notes 10–7, 19–21 of this chapter the discussion of the manuscripts of
the Liber Papiensis as well as the Pistoia manuscript of the Epitome Codicis.
3. On the circulation of the Lombard law even outside Italy, see Meyer 2003.
4. McKitterick 1989, Table A (48–55).
5. For notaries and judges as ‘professionals of the pen’, see, especially, the
chapter ‘Scrivere “in iudicio” nel “Regnum Italiae”’, in Petrucci and Romeo
1992, 195–236. Also essential is Petrucci’s concept of ‘scribal inexperience’:
‘Literacy and graphic culture of early medieval scribes’, in Petrucci 1995,
77–102.
6. See, especially, the plates of the different hands of Pistoia, Bibl. Cap. C. 106,
published by Ciaralli 2000.
7. For the contents of this MS, see Mordek 1995.
8. The Liber Papiensis and its glosses were edited by Boretius in MGH Leges IV.
I am in the final stages of editing, also for the MGH, the Recensio Walcausina
of the Liber Papiensis.
9. Radding 2013, 101–6. This volume is a revised and updated version of Radding
1988.
10. Radding and Ciaralli 2006, 78, 85, and pl. 8.
11. See the comments of Antonio Ciaralli in Radding and Ciaralli 2006, 85.
12. See Boretius’ description, MGH Leges IV, p. LIV, at KM 72, extrav. 32.
13. See, for example, the Expositio ad Librum Papiensem at Grim. 2 §1 and Roth.
200 §1, MGH Leges IV, 398, 344.
14. Radding and Ciaralli 2006, 190–2.
15. Radding 2013, 150–3; Radding 1997.
16. For a recent analysis of V, see Gobbitt 2014. This manuscript is available
online at the Digitaler Lesesaal of the ÖNB; search for ‘Lex Langobardorum’.
P can now be seen in handsome new photos at the Gallica section of the BnF’s
website, and the BL manuscript of the Liber Papiensis, also from the eleventh
century and containing an archaic version of the glosses of the Walcausina,
can be seen online at that library’s site.
17. This subject has not been adequately studied, but see, for example, the long
gloss to Roth. 153 preserved in BAV Pal. lat. 772, edited in MGH Leges IV,
319–21.
18. All early (pre-1150) manuscripts of the Institutes, Code, and Digest were
surveyed for Radding and Ciaralli 2006. A somewhat earlier census of manu-
scripts of the Code from before 1200, not always reliable for its datings, is
Dolezalek 1985. For the middle part of the Digest, Infortiatum, see Dolezalek
1984. An exhaustive index, drawn from library catalogues and including some
non-Roman materials, is Dolezalek 1972.
19. Radding and Ciaralli 2006, 140.
20. The essential analysis of this manuscript is to be found in Ciaralli 2000. See
also our comments in Radding and Ciaralli 2006, 143–50.
310 charles m. radding
21. On this manuscript, see Ciaralli 2000, 30–4; Radding and Ciaralli 2006,
148–50.
22. Radding and Ciaralli 2006, 132–3.
23. Ibid., 158–62.
24. Ibid., 199–202.
25. Dolezalek 1989, 211.
26. Von Savigny 1834–51, 3. 467–9 and esp. 468 n. d.
27. Landau 2008, 48
28. Austin 2009, 137–44, 199–222.
29. For the continued use of Burchard’s Decretum in the twelfth century, see
Rolker 2010, 60–3.
30. On this compilation, see Cushing 1998.
31. Rolker 2010, 137–45.
32. Ibid., 302, translating Ivo, Prologus, http://ivo-of-chartres.github.io/decre
tum/ ivodec_1.pdf, p. 1.
33. See Rolker 2010, 248–89, arguing persuasively that Ivo was not the author of
the Panormia.
34. Winroth 2000. For an overview, see Landau 2008.
35. Winroth 2013.
36. See, for example, Barcelona, Archivo de la Corona de Aragón, Ripoll 78 or
BnF nouv. acq. lat. 1761, available online in the Gallica of the BnF; but the
format remains broadly similar in second recension manuscripts such as BnF
lat. 3884, also in the Gallica.
37. St Gallen, Stiftbibl. 715, available in Switzerland’s e-codices site.
38. On decretals, see Pennington 2009.
39. On peciae at Bologna, see Soetermeer 2005, esp. 254–60. For the origin of the
pecia system at Bologna, see Rouse and Rouse 1994.
 chapter 17

Vernacular Manuscripts I: Britain and France

Ian Short

When Dante wrote his De vulgari eloquentia in the opening years of the
fourteenth century, it had long since been accepted that the Romance
vernacular was an entirely appropriate linguistic register in which to
compose literary texts, whether in prose or verse. By vulgare illustre
Dante meant the standard Italian vernacular of his time, both written
and spoken, which he contrasts with the more remote and artificial register
of Latin, accessible only to a few. Of these two varieties of language, affirms
Dante, the vernacular is the more noble, both because the oral predated the
written register and because the vernacular is the mode employed by the
overwhelming majority of the population.1 Dante’s innovative attempts (in
Latin) at establishing the pre-eminence of the vernacular would probably
have struck his Romance-speaking ancestors of the first half of the twelfth
century as implausible. This was perhaps less true in neighbouring Celtic
and Germanic cultures where the vernacular could claim a more deeply
rooted literary tradition.
Whereas for Dante the term vulgare illustre could be applied compre-
hensively to both the spoken and the literary modes, his early twelfth-
century equivalents were unlikely to have so readily conflated the oral and
written forms of language, which for them fulfilled distinctive functions
that were mutually exclusive. In general terms, the prevailing twelfth-
century polarity seems to have been between the language of the educated
élite and that of the illiterate majority, the former the written vehicle of
literacy, the latter the mode of popular oral communication.2
From a modern critical point of view, however, it is more fruitful to see
these two worlds not in opposition but as given points on a constantly
evolving continuum.3 What Haskins termed ‘an age of new creation in
literature and art’, he also saw as ‘the great period of divergence between
Latin and the vernacular’.4 But far from diverging, these two discourses can
in fact be seen as engaged in a process of progressive assimilation or
symbiosis, the first steps along a path that was eventually to lead to the

311
312 ian short
secularisation and vernacularisation of learning. The monopolistic cate-
gories of the early part of the twelfth century were to grow increasingly
permeable: no longer was the clerical litteratus with privileged access to
knowledge through scripta seen in static and antagonistic opposition to the
secular illitteratus, deprived of learning by the restricting orality of his
sermo. Mid-century England, for example, saw Latin and French engaged
in a mutually enriching process, each culture absorbing literary elements
from the other. ‘By 1154’, according to A. G. Rigg, ‘the integration of
Anglo-Latin and Norman-Latin culture was complete . . . In literary terms
Henry [II]’s succession heralded an effusion of Anglo-Angevin culture, in
both French and Latin, that made England an equal partner in the
renaissance of the twelfth century.’5 But the phenomenon was a wider
one also, as Paul Zumthor explains: ‘A partir d’une époque située entre 1150
and 1250, on voit une à une toutes les langues vulgaires de l’Europe accéder
aux prestiges de l’écrit. Langues chaudes de leur vocalité quotidienne, le
francien [sic] du XIIe siècle, le toscan du XIIIe, le haut-allemand, toutes les
autres: les voici bientôt promues au statut de la langue de la Loi.’6
From about the 1160s, a whole new public began to gain access to
literacy. In the French-speaking world the last quarter of the century saw
secular literature develop to a point where the demand for copies of
vernacular texts had to be satisfied by new centres of manuscript produc-
tion operating in parallel to the established monastic scriptoria.7 The era of
the miles litteratus had dawned.8 To Hartmann von Aue’s literate knight
capable of reading books corresponds the teenage daughter portrayed in
Chrétien de Troyes’ Yvain reading out loud to her parents.9
The rôle played by Britain, a melting pot of Celtic, Anglo-Saxon and
Norman societies, in this evolving European secular culture was
a particularly productive one. The most obvious effect of the Norman
Conquest on English intellectual life had been to make it more open to
Continental influences, and this was as true for the twelfth-century book
as it was for other cultural activities and artefacts. Another important
factor influencing the early development of the vernacular book was its
relationship to the traditional Latin book. From its beginnings as very
much a poor relation, it was to grow progressively in status to become, by
the end of the century, a fully fledged member of the extended family of
books. This was the result of an increase in lay literacy and in the number
of enlightened book-buying patrons. Scribes had no scruples, and few
difficulties, in accommodating new requirements. The Church’s will-
ingness to embrace the vernacular as an auxiliary or complementary
medium of communication to Latin must also have been significant in
 Vernacular Manuscripts I: Britain and France 313
facilitating the transition from a virtual monolingual monopoly to a more
inclusive multilingual culture.
The internationalism fostered by the Conquest did not originally extend
equally to all the Celtic-speaking areas of the British Isles, with the result
that, while book production in Wales and Scotland seems to have declined,
the writing of books in Irish, for example, continued to flourish.
The decimation of the English aristocracy, on the other hand, led to the
temporary eclipse of English as a written language of culture and to its
replacement by French. The effect on English book production was
commensurate. Hence, in quantitative terms of book survival from the
twelfth century, while no fewer than fifty books survive in Irish, English
boasts only half that number, while French boasts the lion’s share of 100, of
which two-thirds are of Insular (Anglo-Norman) rather than Continental
provenance.
Admittedly from the domain of fiction, it is recorded that Gilbert fitz
Baderon, an Anglo-Norman marcher lord of Breton descent, owned
a personal library ‘well stocked with books both in Latin and in French’
before 1191.10 More reliably, Denis Piramus, monk of Bury, writing
between 1190 and 1193, states that contemporary members of the nobility
acquired individual written copies of the lais of Marie ‘de France’.11
Similarly, Henry II’s vernacular historiographer Wace, sometime in the
1170s, declared that in aristocratic courts there was money and therefore
literary patronage, and the possibility of finding purchasers for copies of
one’s works.12 Literary activity at such courts is well documented from
other English sources: between 1154 and 1182, for example, at the court of
Roger de Mowbray, earl of Northumberland, four different categories of
jongleur are recorded: gignator, vielator, cantor, joculator.13
A page had definitely been turned also when, between 1189 and 1210,
Gerald of Wales, deploring contemporary standards of Latinity, diminish-
ing patronage and dwindling audiences for Latin literature, could suggest
having his Expugnatio Hibernica translated into French to ensure its wider
appreciation.14 Having gained such recognition as a suitable medium for
serious works of literature, the French vernacular appears to have come of
age in Britain by the late twelfth century. Largely due to a series of
adaptations of religious or didactic works (translatio), French had acquired,
within a generation or two, some of the textual authority that Latin had
previously enjoyed. The emergence of French from the ecclesiastical ambit
led in turn to secular authors validating the use of the vernacular
by selecting French as their preferred language of composition.
The authorisation of the sermo vulgaris had begun.
314 ian short
The cultural predominance of French in Britain should not blind us to
the fact that the Celtic-speaking areas of the country also participated in
twelfth-century vernacular book production. In the Gaelic-speaking areas
of Scotland and Ireland the eleventh and twelfth centuries saw the rise of
a script-type known as Gaelic National Minuscule. About fifty manu-
scripts and fragments in this script survive from the twelfth century,
among which the vernacular is well represented. From Ireland in the first
half of the twelfth century comes the oldest surviving manuscript written
almost entirely in Gaelic, Leabhar na hUidre (Book of the Dun Cow).
Dublin, Royal Irish Academy 1229 (23 E 25) contains the earliest copy of
the Ulster Cycle epic The Cattle Raid of Cooley (Táin bó Cuailgne), as well as
the Voyage of Bran (Immram Brain) and the Birth of Cú Chulainn. Another
extensive collection of Gaelic vernacular texts is contained in the Book of
Glandalough (Leabhar Glinne dá Locha; in Bodl. Libr. Rawl. B. 502, dating
from the 1130s), and another in the Book of Leinster (Leabhar na
Nuachongbhála; Dublin, Trinity Coll. 1339 (H.2.18) from ca. 1150 to
1165). The Annals of Inisfallen (Bodl. Libr. Rawl. B. 503), an important
historical source, has vernacular entries, in a series of hands, covering the
years 1092–1130 and 1159 to the early fourteenth century. From Scotland
comes the Book of Deer (CUL Ii. 6. 32), a small-format gospel book from
the tenth century to which has been added, between ca. 1130 and 1150,
a collection of charters in Gaelic from Aberdeenshire.15
By contrast, the first books written in the Welsh vernacular appear
only towards the mid-thirteenth century. All that survives from the
twelfth century are a few marginalia and glosses: the Book of Llandaff
(Aberystwyth, National Libr. of Wales 17110B), a church register in
Latin from Llanbadarn Fawr, preserves a few names and marginalia in
Welsh; vernacular glosses to a short fragment of Bede’s De natura
rerum appear in National Libr. of Wales Peniarth 540, from the first
half of the century, and a large number of isolated Welsh words in the
genealogies of BL Harley 3859, containing the oldest surviving copies of
‘Nennius’ and the Annales Cambriae.16 Some Cornish words survive in
the marginal commentary to John of Cornwall’s Latin Prophetia
Merlini of 1153–4 in BAV Ottob. lat. 1474, from the last third of the
century.17
The number of surviving manuscripts in Middle High German, includ-
ing a significant proportion from the eleventh century, is more than double
those in French.18 Curiously, neither Italian, Castilian, Catalan nor
Portuguese literatures are represented in the surviving twelfth-century
manuscript corpus.19
 Vernacular Manuscripts I: Britain and France 315
These crude statistics reveal a quite different picture when their con-
stituent elements are subjected to closer analysis. For example, the Middle
English corpus, inevitably curtailed as a result of the Norman Conquest,
appears at first sight to be more or less comparable with the French.
However, sixty of these are administrative documents and glosses, while
a third of the remaining texts are copies of pre-Conquest works. There was,
writes Laing, ‘apparently very little composition in English in the century
after the Conquest apart from a few late additions to the Anglo-Saxon
Chronicle. It was only from the late twelfth century that new writings in
contemporary English began to appear in any quantity’.20
The bulk of twelfth-century literary production in English is situated
firmly within the Old English homiletic prose tradition, and gives no
indication of any significant Continental influence. There is no sign here
of the recession of Anglo-Saxon culture and the ascendancy of French. It is
clear that, in the absence of lay patrons whose native language was English,
secular literature in English had little opportunity to develop and thrive in
the twelfth century, and still less to be written down. Worcester,
Canterbury and Exeter seem to have been the main centres of production.
Apart from the continued use of some Anglo-Saxon letterforms, the scripts
and the (usually single-column) mise-en-page have few recognisable char-
acteristics; indeed some of the hands (mostly English vernacular minus-
cule) have, rightly, been described as idiosyncratic and ‘not very
calligraphic’.21 There is, however, a measure of cultural contact to be
seen in the shape of a dozen or so twelfth-century manuscripts and
fragments in which French and English appear together in the same
book (such as the Eadwine Psalter), or a further fifteen if the chronological
limit is extended to 1220.22
Compared to surviving manuscripts in English, those in French grew
significantly in number and in the range of their functions as the century
progressed. Prior to the last quarter of the century, all surviving manu-
scripts in French were either translations or reworkings of originals written
in Latin (or, exceptionally, Old English in the case of Gaimar). The early
rôle of the French vernacular has to be seen, however, first and foremost as
an adjunct to, rather than as a substitute for, Latin. By the last quarter of
the twelfth century, however, French had become a language of literary
patronage.
Also noteworthy is the contrast between the considerable quantity of
French literature composed in the twelfth century and the meagre amount
that survives in contemporary manuscripts. Striking also is the amount that
survives in fragments rather than whole books or booklets. The vagaries of
316 ian short
conservation must certainly have played a major rôle here, but it is also
possible that only a few copies of individual texts circulated at the time.
This seems to be true, for example, of the widely popular Tristan poems,
which have survived only in fragments. Conservation rates in Britain seem,
for whatever reason, to have been appreciably higher than those on the
Continent.
To judge solely from the number of vernacular manuscripts that have
survived, Insular French gives every indication of having been both more
innovative, more productive and more precocious than its Continental
counterpart. Only a third of the 104 twelfth-century manuscripts contain-
ing French are of Continental rather than Anglo-Norman provenance.23
Before looking in detail at the component parts of this richly varied French
and Occitan corpus, a typology might provide readers unfamiliar with the
field with a general overview of its structure.
Twenty-seven of the 104 are fragments, ranging from a single leaf or
bifolium to two whole quires, scattered across the length of the century and
originating from a variety of geographic areas. In only three cases, all
psalters, are the fragments bilingual. Half of the fragments are of small or
small-to-medium size with minimal decoration, and the majority belongs
to the textual category of ‘recreational literature’. This is the most vulner-
able of all the textual categories: of the dozen or so items belonging to the
romance genre, eight survive only in fragments.
Another category consists of seventeen vernacular additions to Latin
books. This is a well-documented phenomenon, attested from the ninth to
the eleventh centuries, and continuing throughout the twelfth. Only rarely
are the additions out of harmony with the volumes that shelter them.
There is codicological coherence also, for instance, when the additions
were inserted by the scribe of the Latin text which precedes them, or were
made in the same scriptorium, as in the case of the epic accretion to the
Gundulf Bible.24
Fifty-eight manuscripts can be described as books in their own right, of
which twenty-five contain nothing but French material. The remaining
thirty-three have firm links to Latin either because they form part of
bilingual miscellanies or because they are translations of or commentaries
on Latin originals.
The bilingual miscellanies are mostly small-format Insular books with
little decoration, containing texts of a practical nature: scientific, legal or
scholastic. Sometimes the same hand intervenes in both parts: in Bodl.
Libr. Digby 13 the hand of the French Lapidary is the same as the one that
copied the Isidore in the following quire. Their homogeneity is sometimes
 Vernacular Manuscripts I: Britain and France 317
textual: BL Add. 49366 is a collection of legal texts, and BnF nouv. acq. lat.
873 a compendium of didactic material. In rare cases, the French text can
be inserted into a liturgical book, as in the celebrated case of the Chanson
d’Alexis, which found a home in the St Albans Psalter.
The translations and commentaries that present the Latin and the
French texts on the same page tend naturally to come in larger-format
and more sophisticated books. Pride of place here goes to the dozen or so
psalters, all Anglo-Norman, to the translation of Marbod’s Lapidary,
probably made in Sicily (BnF lat. 14470), and the Isidore texts in Epinal,
Bibl. mun. 58 from north-eastern France. In four separate vernacular
commentaries on the Psalms, the French is used to gloss and expound
the Latin.
Books exclusively in French are in a minority. From the first part of the
century come two Insular French books, the celebrated Chanson de Roland
in Bodl. Libr. Digby 23 (but notoriously difficult to date; it could be as late
as the 1160s) and in every sense of the term a unicum. No other comparable
secular book in French survives from this date; it is a somewhat shabby
manuscript and its hand unskilled. The Oxford Psalter (Bodl. Libr. Douce
320), on the other hand, datable to the 1140s and made at the abbey of St
Albans, is written in a professional hand, and was in all likelihood designed
for nuns (Figure 17.1). The second half of the twelfth century sees an
increase in production: two long biblical texts, a Life of Thomas Becket,
a vast historiographic text from Touraine and a group of homiletic books
from eastern France with Cistercian connections. Purely secular books
include a verse epic, a romance and a legendary history. The Codi, a legal
treatise, stands very much in isolation in a remarkably small Occitan
corpus.
More noteworthy are three collections exclusively in French featuring
the works of a single author; the poet Philippe de Thaon merited a book of
his own to house his Comput and his Bestiaire, while the two religious
poems of Frère Angier, another Anglo-Norman and a resident of Oxford,
fill another book. Nantes, Musée Dobrée V is a collection of St Bernard’s
sermons and epistles. In addition, there are three volumes that seem to have
been assembled according to literary genre: an important historiographic
collection from Durham, the vernacular hagiographies in the former
Ashburnham-Libri 112 (now BnF nouv. acq. fr. 4503) and the didactic
miscellany in BL Harley 4388.
The Anglo-Norman survivals from the twelfth century, comprising, as
we have seen, two-thirds of the entire French language corpus, cover an
exceptionally wide spectrum of texts, from the oldest extant copy of the
318 ian short

Figure 17.1 The Oxford Psalter, St Albans, 1140s. Oxford, Bodleian Library, Douce
320, f. 54r.
 Vernacular Manuscripts I: Britain and France 319
French national epic, the Chanson de Roland (Bodl. Libr. Digby 23), to the
earliest law texts (BL Add. 49366 ff. 141–144v: Leis Willelme) and the first
administrative document, so far as we know, to be recorded in the French
vernacular (London-Kew National Archives C 146/10018), a return to the
Inquest of Sheriffs of 1170.25 Bodl. Libr. Douce 320 (the Oxford Psalter,
from St Albans) of ca. 1145 is the earliest book containing exclusively
French text to have survived from post-Conquest Britain.26 In being
monolingual it stands alone within its literary environment of eleven
other vernacular French psalters, all bilingual. With the exception of the
trilingual Eadwine Psalter (Cambridge, Trinity Coll. R. 17. 1 [987] written
at Christ Church, Canterbury in 1155–60) and its copy, BnF lat. 8846, all
the other twelfth-century French psalters reproduce the Oxford text in
varying formats. In the Winchester (or Henry of Blois) Psalter from shortly
before 1161 (BL Cotton Nero C. IV), the French text is in parallel with the
Latin, on facing pages, the same mise-en-page as in the so-called Corbie
Psalter (BnF lat. 768), the Copenhagen palimpsest (Copenhagen,
Universitetsbibl. AM 618 4°), BnF nouv. acq. lat. 1670 (after 1173, probably
from Christ Church, Canterbury) and the St John’s Oxford pastedown
(Oxford, St John’s Coll. HB4/4. a. 4. 21 [I. subt. 1. 47]). On the other hand,
in the Arundel Psalter (BL Arundel 230, probably from Peterborough) and
the fragmentary Orne Psalter (Paris, Archives nationales AB XIX 1734
[dossier Orne n. 1]), the layout is interlinear, whereas the Maidstone
fragment (Maidstone, Kent County Archives Fa Z 1) and the damaged
BL Cotton Vitellius E. IX set out the two texts by alternate verses.27 Of this
corpus only the text of the Orne Psalter shows any significant textual
variation from Douce 320. The writing and illustrations of some of these
psalters place them among the greatest achievements of romanesque art in
Britain. The Oxford Psalter text, copied countless times in succeeding
centuries, was to prove one of Anglo-Norman England’s most popular and
successful literary exports to the Continent. Other biblical texts include
a handsomely decorated translation of the four Books of Kings (Paris, Bibl.
Maz. 54 [70]: Quatre Livres des Rois), and Herman de Valenciennes’ verse
Roman de Dieu et de sa mère (Geneva, Bibl. de Genève Com. Lat. 183).
No complete Bible in French prose survives from earlier than the second
half of the thirteenth century.28
Passing to hagiography: the celebrated St Albans Psalter of ca. 1130
(Hildesheim, Dombibl. St. Godehard 1) contains the French Chanson
d’Alexis and a short prose text of St Gregory on images.29 There are two
more copies of the Alexis in the fragmentary BAV Vat. lat. 5334, and in BnF
nouv. acq. fr. 4503. The latter also contains the Vie de sainte Catherine by
320 ian short
Clemence of Barking, the Anglo-Norman Voyage of St Brendan by
Benedeit and an extract from Herman de Valenciennes’ poem, plus an
added translation of a papal bull dated April 1177. The rhymed Life of
Becket by Guernes de Pont-Ste-Maxence survives in four twelfth-century
copies: BL Harley 270, Bodl. Libr. Rawl. C 641, Wolfenbüttel Cod. Guelf.
34. 6 Aug. 2° and the fragment in London, Society of Antiquaries 716.
A fragment of another French Life of Becket by Beneit of St Albans in tail-
rhyme survives in Cambridge, Gonville and Caius Coll. 123/60. Two
fragments of the Brendan, more a romance quest-type narrative than
a hagiography, are preserved in Bodl. Libr. Rawl. D. 913, and Cologny-
Genève, Bibl. Bodmer 17.
Closer to romance in the generally accepted sense is the Tristan of
Thomas in Bodl. Libr. French d. 16, a two-quire fragment in a skilful
hand and with a historiated initial in gold, perhaps the remains of
a presentation copy. In the composite manuscript BAV Pal. lat. 1971 we
find, in addition to the romance Partonopeus de Blois, fragments of Amadas
et Ydoine and of Floire et Blancheflor. There is a love song El tens d’iver
quant vei palir in Cambridge, Pembroke Coll. 113, and, complete with
musical notation, the crusading poem Chevalier, mult estes guariz, of which
the original dates from 1146, in Erfurt, Universitätsbibl. Dep. Erf. Cod.
Amplon. 8° 32. There are no chansonniers before ca. 1240. Epic literature is
represented not only by the Oxford Roland, but also by Gormont et
Isembart (Brussels, Bibl. royale II 181 frag. 3) and the Chanson
d’Aspremont in Cologny-Genève, Bibl. Bodmer 11 (with another fragment
of eight folios in BAV Pal. lat. 1971).30
Historiography, one of the most notable achievements of twelfth-
century Anglo-Norman culture, is represented by Durham Cath. C. IV.
27, a professionally written volume of historical texts dating from the end
of the century, which ensures that we have complete texts of Wace’s Brut
(ff. 1–94) of 1155, Geffrei Gaimar’s Estoire des Engleis (ff. 94–137) of 1136–7,
the Description of England (ff. 137–138v) of ca. 1139, and Jordan Fantosme’s
chronicle of the revolt of the Young King in 1173–4 (ff. 139–167v), the
whole providing a more or less continuous historical panorama of England
from its legendary foundation by Brutus until the reign of Henry II.
A fragment of Wace’s Brut is in Bodl. Libr. Rawl. D. 913, another Brut,
this time anonymous, in BL Harley 4733, and a hybrid prose genealogy of
the kings of France in BL Cotton App. 56.
Of didactic literature we have six copies of Philippe de Thaon’s rhymed
Comput of 1113–19: complete in Lincoln Cath. 199, in a very formal and
expert hand from the middle of the century, BL Arundel 230, and three
 Vernacular Manuscripts I: Britain and France 321
fragments in CUL Add. 4166 frag. 9, BL Cotton App. 56 ff. 110–111 and
BAV Reg. lat. 1244, ff. 53–58. In BL Cotton Nero A. V the Comput is
accompanied by the only complete text of Philippe’s Bestiaire of 1121–35,
making him the first vernacular author to have more than a single work
surviving in the same book. A verse lapidary in Cambridge, Jesus Coll.
Q. D. 2 (44) is also attributed to Philippe, and there are two surviving
lapidaries from the twelfth century in prose: BnF nouv. acq. lat. 873, and
a single-leaf fragment in Bodl. Libr. Digby 13, f. 21, written in the dis-
tinctive Christ Church ‘prickly’ script and dating therefore from the first
quarter of the century. Elie of Winchester’s Distiques de Caton are pre-
served in the didactic miscellany of BL Harley 4388 together with the
Anglo-Norman translation of Petrus Alfonsi’s Disciplina clericalis,
Guischart de Beaulieu’s Sermon in alexandrines and Sanson de Nanteuil’s
Proverbes de Salomon. The homiletic Roman des Romans has come down to
us in Yale University Beinecke 590, and the Dialogues de Grégoire le Grand
and the Vie de saint Grégoire, both by Angier, Augustinian canon of St
Frideswide in Oxford, appear together in BnF fr. 24766, written in
a skilful, professional hand and explicitly dated to 1212 and 1214,
respectively.31
What is distinctive about the Anglo-Norman corpus from the
codicological point of view is the mise-en-page of the poetic texts,
written in rhyming couplets set in double columns.32 The second line
of each couplet is indented, giving a jagged appearance to the column
but at the same time ensuring the autonomy of the couplet and the
maintenance of metrical regularity by the copyist. Exceptionally, as in
the case of the Chanson d’Alexis, verse can be written continuously as
prose with punctuation marks to separate verse lines. When the text is
laid out as verse, a punctus usually marks the end of each line. In the
case of Amadas et Ydoine, the last letter of each verse is detached,
ranged to the right and justified. In the tail-rhymed life of Becket by
Beneit, the mise-en-texte is adapted to the unusual metrical scheme by
the use of zigzag lines and a secondary column on the right.
A complex system of accented vowels, designed to facilitate reading
aloud but also sometimes functioning as diacritics, is characteristic of
Anglo-Norman vernacular manuscripts. Punctuation is generally mini-
mal, and little use is made of abbreviations. With the exception of the
psalters, decoration is confined to coloured initials, occasionally his-
toriated as in the Mazarine Quatre Livres des Rois, perhaps another
presentation copy. The Alexis in the Hildesheim MS is, needless to
add, sui generis.
322 ian short
No generalisations are possible about the pre-Gothic hands used in the
vernacular French corpus; they range from the expert calligraphy of the
formal libraria, via the more flexible and rapid ‘school’ hand, to clumsy
scrawls.33 Viewed as a whole, about half of the corpus comprises small-
format fragments. Many of these make their appearance as insertions in or
additions to Latin texts. Twenty-five of the complete books comprise
exclusively vernacular texts. Latin, however, is ubiquitous, either visibly
in the fifteen surviving bilingual miscellanies or vicariously as the source
language from which vernacular translations or adaptations have been
made. Only two of the exclusively French books, the Oxford Roland
(very much a unicum) and the Oxford Psalter, can be dated to the first
half of the century. Apart from the Eadwine Psalter, there is a single
trilingual text in which Latin (in black ink), French (in green) and
English (in red) share the same page: formulas for the visitation of the
sick inserted, as f. 156, into BL Cotton Titus D. XXIV from the Cistercian
abbey of Rufford near Nottingham.34 An aid to the localisation of verna-
cular manuscripts, not available for those in Latin, is their dialect char-
acteristics. Anglo-Norman has its own recognisable spelling system, as do
other regional varieties of medieval French, such as western French,
Normano-Picard, north and north-eastern French, Champagne and the
central area of France known as Francien.
Turning from Britain to the Continent, we see that the situation in
France was considerably less heterogeneous, owing primarily to the fact
that fewer than a third of the surviving twelfth-century vernacular manu-
scripts are of Continental, as opposed to Insular, provenance. More than
half of these (some sixteen manuscripts) trace their origins back to north
and north-east France, the rest coming from the north-west regions of
Normandy, Anjou and Touraine. Only two emanate from the Central
(Champagne-Vermandois) area, one a psalter commentary (New York,
Pierpont Morgan Libr., M. 338), the other a reworking of the book of
Genesis in octosyllabic couplets (BnF fr. 900) written for Countess Marie
de Champagne. Conspicuous by his absence from this group is Countess
Marie’s celebrated vernacular protégé, Chrétien de Troyes; the earliest
manuscripts of his romances are from the thirteenth century.35 Homiletic
texts form the majority grouping in books from the north-east, with
a particular concentration on the works of Bernard of Clairvaux (Nantes,
Musée Dobrée V; BnF fr. 24768; Berlin, Staatsbibl. Phillipps 1925) and
Gregory the Great (Bern, Burgerbibl. 79; Laon, Bibl. mun. 224 [455];
London, Lambeth Palace 73) and other texts that can be categorised as
Cistercian (Epinal, Bibl. mun. 58 [209]: Dialogue de l’âme et de la raison;
 Vernacular Manuscripts I: Britain and France 323
Verdun, Bibl. mun. 72: William of St-Thierry’s Lettre aux frères de Mont-
Dieu). Apart from the Munich Brut (BSB Cod. Gall. 29) and a fragment of
the epic Fouque de Candie (BnF nouv. acq. fr. 18217), the other survivals are
also devotional texts, including the fragmentary Vatican Chanson d’Alexis
(BAV Vat. lat. 5334).
A similar distribution is demonstrated by the surviving twelfth-century
manuscripts from west and north-west France: Benoît de St-Maure is
represented by the best text of his Chronique des ducs de Normandie
(Tours, Bibl. mun. 903), by a fragment of his Roman de Troie (Basel
Universitätsbibl. N I 2, 83 + Brussels Bibl. royale II. 139/3) and by
Etienne de Fougères’ didactic and anti-feminist Livre des Manières
(Angers, Bibl. mun. 304 [295]). From Angers comes the octosyllabic Vie
de saint Sylvestre; the remainder of the books are almost all homiletic or
otherwise devotional.36 Special mention should be made of the bilingual
lapidary, from the very end of the century, in BnF lat. 14470 ff. 4v–35v,
probably copied and decorated in Sicily from an Anglo-Norman original.37
The handful of surviving manuscripts of Occitan origin includes the
Justinian law book known as Lo Codi (Paris, Bibl. de la Sorbonne 632),
elegantly written in south-east France at the end of the century. The dating
of the extract of the octosyllabic Roman d’Alexandre by Albéric de Besançon
in Franco-Provençal (Florence, Bibl. Laurenz. Plut. 64. 35) is controversial:
no less a figure than Bernhard Bischoff placed it in the first half of the
eleventh century, whereas it is traditionally attributed to the early years of
the twelfth.38
It is immediately obvious that, in terms of historiography and psalter
translations, Insular production is significantly in advance of that of the
Continent. Even Benoît’s Chronicle was commissioned by Henry II, while
the Munich Brut has an Anglo-Cambrian origin. That the overwhelming
majority of the Continental survivals should prove to be religious texts may
well be a function of manuscript preservation, but could also point to
a dearth of artistic patronage and production. Social conditions in Britain,
its multinational aristocracy, its pluri-culturalism and its pluri-lingualism
must obviously have been determining factors in making Anglo-Norman
vernacular culture so innovative by comparison with its less centralised and
unified Continental equivalent. Enlightened patronage from among the
nobility must also have played a rôle. In literary terms, its direct contact
with Celtic cultures must go a long way to explaining its creativity and
originality, and the transmission, to France and Germany in particular, of
the matière de Bretagne was one of its most significant contributions to
medieval literature. It is only from the 1170s that Continental French
324 ian short
literature can be said to flourish, and indeed to take the ascendancy over its
cross-Channel neighbours, which it does spectacularly.39
The political reintegration of Normandy in 1204 marks the beginning of
the end of Anglo-Norman England’s golden age, and forms a convenient
terminus for our coverage of Insular manuscript production in the twelfth
century. By the 1220s, moreover, the schools were not only giving a new
impetus to book production, but Middle English literature in its written
form was also emerging from the hibernation imposed on it by the
Norman Conquest. In addition, the English episcopal ordinances promul-
gated in the wake of the Fourth Lateran Council of 1215 and prescribing the
teaching of the faithful in their own language (domestico idiomate, in the
words of Richard Poore) were changing the face of Anglo-Norman litera-
ture and the rich body of manuscripts that transmitted it.
The first two decades of the thirteenth century ushered in a new era
in vernacular manuscript production throughout Western Europe, in
which the modest total of twelfth-century survivals can be seen to
mushroom at an unprecedented rate, and from which more than
1,000 books are extant today, in French alone.40 A professional infra-
structure of copyists and artists catered for the needs of a new genera-
tion of book owners, and the vernacularisation of written culture was
becoming an established feature of the social and intellectual land-
scapes of the leisured classes.

Notes
1. Dante, De Vulgari Eloquentia, ed. Botterill I/i. 1–4; cf. Lusignan, 40–7, 67–73.
2. On the status of medieval Latin and its implications, see Richter 1976; Wright
2002; the various contributions to Clanchy 2012, 203–23; Garrison et al. 2013;
Maiden et al. 2013; Stock 1983, 12–87. On Latin vernacular bilingualism, see
Cazal 1998; cf. Smith 1999.
3. Zumthor 1987, 130–7.
4. Haskins, Renaissance, 189–90; cf. Brooke 1971; Benson and Constable,
Renaissance; Southern 1970.
5. Rigg 1992, 63. Rigg speaks (9) of ‘the Normanization of England and the
Anglicization of the Normans being well under way by King Stephen’s death’;
cf. Short 1992 and 1996.
6. Zumthor 1984, 63; cf. the earlier surveys of Richter 1979, 9–20; Stock 1983,
24–6; Zumthor 1984, 55–63, and 1987, 132–7, 196–202.
7. Careri, Ruby and Short 2011, xvi ff. For a more broadly based study of early
Romance manuscripts cf. Frank 1994; Raible 1998.
8. Aurell 2011; Bäuml 1980; Clanchy 2012, 227–34, 246–52; Turner 1978.
 Vernacular Manuscripts I: Britain and France 325
9. ‘sô gelêret was | daz er an den buochen las | swaz er dar an geschriben vant’, ed.
Paul: l. 1–3 [ca. 1195]; ‘une pucele . . . n’ot mie plus de seize anz’: Chrestien de
Troyes, vv. 5356–8 [ca. 1175–80; cf. Laudine reading her psalter at vv. 1418–9].
10. ‘un livre q’il me fist moustrer | dount sis chastels est mult manaunz | e de latyn
e de romaunz’: Hue de Rotelande, Protheselaus, 2. ll. 12706–10.
11. Denis Piramus, ll. 39–45; detail in Short 2007.
12. Wace, Roman de Rou: 110, 3. 163–6; cf. Short 2007, 322. On courtly culture in
general, see Bumke 1991, and on literary patronage Bumke, Mäzene.
13. Greenway 1972, xl, lxii–lxiv.
14. Gerald of Wales, 264–5. Gerald opposes ‘scripta latina paucioribus evidencia’
to ‘verba aperta communi idiomate prolata’; details in Short 2009.
15. Duncan 2008, respectively 39–51, 282–8, 70–120, 289–308, 33–8. I am very
grateful to the author for having kindly provided me with a copy of this
important study. Cf. Edel 2003.
16. Huws 2000, 11–3. Cf. Huws in CHBB II. 390–6; Davies 2003.
17. Curley 1982, 236–40.
18. Our comparative statistics, taken from Careri et al. 2011, xix–xxi, are not
necessarily complete or definitive. The voluminous Old Norse literature has
come down to us almost exclusively in copies later than the thirteenth
century, with the exception of a small number of twelfth-century fragments
and of an incomplete Icelandic version of the Elucidarius; cf. Gunnlaugsson
2008 (Table 1 on 88 lists the twelfth-century survivals.) Cf. McTurk 2008, 184,
246–9. See also Chapter 18.
19. Asperti 2006, 266–76 (268–9 for the Pianto della Vergine and the Tuscan
Ritmo Laurenziano, and 149–53 for the Spanish glosses of San Millán and
Silos). Asperti (274) speaks of ‘il ritardo di un secolo di Penisola Iberica e Italia
rispetto alla Francia’ and ‘la povertà e assoluta sporadicità delle attestazioni
iberiche e italiane’. The Romanian vernacular is not attested before the
sixteenth century.
20. Laing 1993, 3.
21. CHBB II. 24, 127 n.101.
22. Da Rold et al.: Index of Manuscripts containing French [and English]. For
a selection of bilingual Anglo-Norman/Middle English books from the
thirteenth century, see Taylor 2003, 4–15; cf. Emmerson 2007 and Frankis
1986.
23. Careri et al. 2011, xxxiii. For a general survey of Anglo-Norman, see Short
2013. On multilingualism, see Tyler 2011.
24. See n. 30.
25. An attempt at a complete list of Anglo-Norman literary manuscripts is to be
found in Dean and Boulton 1999.
26. Short et al. 2010 and Short 2015. The text was probably translated directly
from the St Albans Psalter.
27. Careri et al. 2011, nos. 13, 81, 60, 34, 76, 19, 88, 67, 68, 31, 49, 36, respectively.
I omit the twelfth-century psalter commentaries in Hereford Cath. O. III. 15
and in Bodl. Libr. Laud. misc. 91, and the rhymed psalter paraphrase of BL
326 ian short
Harl. 4070 (Careri et al., nos. 26, 63 and 39). At least ten psalters in French
held by English monastic houses are listed in Blaess 1973.
28. For other religious texts, see Careri et al. 2011, nos. 15, 35, 55, 59, 61, 77, 85.
On bibles, see Bogaert 1992.
29. Bepler, Kidd and Geddes 2008; Gerry 2009. For the relationship between text
and image in the French corpus, see Careri et al. 2008, lv–lviii.
30. A blank column-end in the Gundulf Bible (San Marino, Huntington Libr.
HM 62 vol. I) has been filled (ca. 1140–50) with the first six verses of an
unknown chanson de geste: Careri et al. 2011, no. 89.
31. For completeness we should mention the proverbs in CUL Ii. 1. 33, an Ælfric
manuscript, and in Bodl. Libr. Digby 53. For medical recipes and glosses, see
Careri et al. 2011, nos. 42–5 and 66.
32. Exceptionally, the Oxford Roland, the Bodmer Brendan and Adam of Ross’
Vision de saint Paul (Bodl. Libr. Douce 381) are laid out in single columns,
which is also the format for texts in alexandrines; cf. Careri et al. 2011, liv.
Single columns are also typical of twelfth-century English language texts.
In Albéric’s Alexandre the verses are written as prose but over two columns.
33. For more codicological and palaeographic detail, see Careri et al. 2011,
xlvii–lviii.
34. Oxford, Jesus Coll. 26 from the 1120s has a six-line trilingual note to a table of
consanguinity on f. 170v.
35. Busby et al. 1993; cf. Busby 2002; Careri et al. 2011, xix, 15–8 for the new three-
column layout.
36. A full list is in Careri et al. 2011: nos. 3, 8, 20, 52, 69, 73, 78, 80, 83, 92; nos. 4,
79 and 90 are documents.
37. Careri et al. 2011, 188–91 no. 84; cf. Avril and Gousset 1984, 175 no. 212.
38. Careri et al. 2011, nos. 24 (Alexandre) and 70 (Codi). Additionally nos. 38, 80
and 82 are Occitan. The Occitan Boeci dates from the end of the eleventh
century.
39. Zink 1981; cf. Zink 1985.
40. Asperti 2006, 271; Careri et al. 2011, xvii, xix.
 chapter 18

Vernacular Manuscripts II: Germany

Nigel F. Palmer

In 1988 Ernst Hellgardt published a list of 276 German vernacular manu-

scripts from the ‘eleventh and twelfth centuries’ (codices, fascicles, inter-
polations and fragments),1 of which 193 had been thought datable to the
twelfth century (forty-seven to the eleventh century and thirty-six to the
boundary zone of the early thirteenth century). None of these manuscripts
are dated, few are datable on the basis of external criteria, and for the rest
Hellgardt, in this pioneering study, gave the dates current in the scholarly
literature. Since then, in the course of the ongoing reassessment of the
significance of the material evidence of manuscripts for literary studies,
there have been further developments. In 2005, in the context of a study
of the punctuation of Middle High German verse texts, I published a list of
109 manuscripts of German verse which I considered datable to the period
1100 to 1250, assigning to each manuscript a date; no account is taken there of
prose.2 Forty-one of the verse manuscripts are assigned to the twelfth century
in this list, and thirty to the period ca. 1200 or to the beginning of the
thirteenth century. Much more important has been the enterprise of collea-
gues in Marburg, who have built up the online ‘Handschriftencensus’
(HSC) as part of the ‘Marburger Repertorien’ database, in which they record
a skeleton description, with extensive bibliography, for every known
medieval manuscript that contains significant German vernacular material.3
There is a huge amount to be done for the later Middle Ages, but the list of
manuscripts for the eleventh/twelfth and early thirteenth centuries is very
complete. The material presented in the HSC is linked to another database,
the ‘Paderborner Repertorium der deutschsprachigen Textüberlieferung des
8. bis 12. Jahrhunderts’, which contains more detailed information.4 Whereas
Hellgardt, whose aim was to document what had been stated in previous
scholarship, came up with nearly 200 German manuscripts and fragments for
the twelfth century, the Paderborn Repertorium lists only 135 German
manuscripts and fragments from the period 1075 to 1225 (largely as
a result of different criteria for identifying a ‘German manuscript’ and the

327
328 nigel f. palmer
taking into account of more recent scholarship, where there has been
a marked trend, for better or worse, to date later). The online databases
suffer from the disadvantage that they focus exclusively on vernacular
material, paying too little attention to what is often a very significant Latin
context.5
A bird’s-eye view, such as these numerical data provide, is full of
imprecisions, as it obscures differences of opinion, especially about the
place of bilingual manuscripts. It obscures the fact that only ten manu-
scripts exhibit Central German dialect features and only seven those of the
Low German language spoken in the north. The surviving material comes
almost entirely from the region east of the Rhine and south of the Main.
Furthermore, as Christa Bertelsmeier-Kierst pointed out in the first of two
overview studies presented at colloquia in 1998 and 2001, the production of
German manuscripts, to judge from what has survived, increases slowly
from very modest proportions (in quantitative terms) in the eleventh
century and the first half of the twelfth, only acquiring dynamic at the
end of the twelfth and in the years around 1200 (in which period she would
place a quarter of all German manuscripts conventionally dated to the
twelfth century), and increasing rapidly in the thirteenth century (some
800 German manuscripts).6 On the most conservative estimates of how
many of the items considered in these surveys truly represent vernacular
book production, and however one tries to account for their survival, it is
clear that the quantity of vernacular books produced in the twelfth century
and in the years around 1200, in the southern and south-east German
lands, was quite considerable and stands out in quantitative terms by
comparison with other European vernaculars.
Many of the German vernacular manuscripts from this period are
monastic in origin, and conservative forms of monastic literature, in
particular bilingual texts written in a mixture of Latin and German,
dominate at least the earlier part of the period under consideration. But
how does this manuscript-based perspective sit alongside what we know of
the new developments in the literary world during a period that saw
a blossoming of new literary forms?7 The second half of the twelfth century
and the early years of the thirteenth had witnessed the rise of German court
love poetry (the ‘Minnesang’) as well as the didactic and religious lyric, the
first courtly romances and epic poems, such as Pfaffe Lambrecht’s
Alexanderlied (ca. 1150/60),8 the literary composition of a major example
of heroic epic (the Nibelungenlied and Klage, ca. 1180–1210)9 and court
romances culminating in Gottfried von Strassburg’s Tristan (ca. 1210),10
Wolfram von Eschenbach’s Parzival (before 1210) and the same author’s
 Vernacular Manuscripts II: Germany 329
remarkable epic poem in the manner of the ‘chansons de geste’, his
Willehalm (before 1217).11
In the monastic domain major changes resulted from the reform move-
ments, important examples of which are the Hirsau reforms from the later
eleventh and early twelfth centuries, which encompassed not only great
Benedictine abbeys in the south-west such as Hirsau itself, St. Blasien and
Zwiefalten, but also extended northwards to Reinhardsbrunn in Thuringia
and most importantly eastwards as far as Admont and Millstatt in Austria,12
and the new reformed orders that followed the Benedictine Rule, notably the
Cistercians (whose first foundations in the German lands date from 1123
onwards) and the Premonstratensians (1122 onwards). The Carthusians did
not yet have foundations in this region. In the German context the
Benedictine reforms had a considerable impact on libraries and manuscript
production, preceding the coming of the mendicant orders in the thirteenth
century and the efflorescence of new forms of female monasticism. Where
provenances can be attributed to German monastic manuscripts, as with
a number of copies of Williram’s Commentary on the Song of Songs
(Figure 18.1) the overall picture is that of transmission throughout the net-
work of Benedictine abbeys, such as Ebersberg, Einsiedeln, Lambach and
Kremsmünster, and the Augustinian canons of Indersdorf and Vorau. At the
end of the century, however, we encounter a cluster of texts and manuscripts
that can be associated with the double house of Admont, which had
become a stronghold of the Hirsau Reform in the eastern territories of the
Empire. A good case has been made for the nuns of Admont as the intended
recipients of the St Trudperter Hohes Lied (ca. 1160, with an Austro-Bavarian
fragment datable ca. 1170–90 and a complete South-West German codex
from ca. 1230) and for Admont as the abbey where the Karlsruhe-Cracow
manuscript of the Millstatt Sermons (early thirteenth century) was copied,
perhaps for the lay brothers and quite likely from an exemplar deriving from
the Hirsau reform monastery of St. Georgen in the Black Forest. Admont is
also the ‘Schreibheimat’ of the Millstatt Psalter and Hymnal, which contains
German interlinear translations (datable ca. 1200).13 Some of the Benedictine
abbeys were double monasteries, with men and women, but scholarship has
in general preferred to think in terms of lay brothers (conversi) rather than
nuns as the likely audience for the German manuscripts. An important
exception, the only twelfth-century German manuscript definitely written
for a woman, is the minuscule bilingual Prayer Book of Muri from
the second half of the century (possibly before 1175), which is probably
monastic, despite elements most likely written with laywomen in mind
(Figure 18.2).14
330 nigel f. palmer

Figure 18.1 Williram von Ebersberg, Commentary on the Song of Songs. Written area:
198–203 × 146–54 mm. Vienna, Österreichische Nationalbibliothek, Cod. 2686,
f. 17v.
 Vernacular Manuscripts II: Germany 331

Figure 18.2 Prayer Book from Muri. Written area: 70 × 50 mm. Sarnen, Bibliothek
des Benediktinerkollegiums, Cod. membr. 69, ff. 1v–2r.

A second factor is lay patronage for German literature, as articulated in

the prologue to the Rolandslied (Henry the Lion, duke of Bavaria, after his
marriage to Matilda in 1168, datable ca. 1170 or a little later),15 or in the
epilogue of Heinrich von Veldeke’s Servatius (Countess Agnes of Loon,
1170s or 1180s).16 The epilogue to Heinrich von Veldeke’s Eneas states that
the manuscript of his German adaptation of the Roman d’Énéas, which had
been entrusted to a lady-in-waiting, was stolen by a certain Count
Heinrich on the occasion of the marriage of the Countess of Kleve and
Landgrave Ludwig III of Thuringia. Nine years later it was returned to the
poet by Landgrave Hermann of Thuringia, Ludwig’s brother, for whom
Veldeke completed the work.17 Other texts, all from the second half of the
century, can be associated with specific noble individuals or families on the
basis of their naming in the texts, as for example with ‘von Horbvrc her
Walther’, attested in Alsace fl. 1130–56/62, in Reinhart Fuchs.18 In the
bridal-quest epic König Rother an association with noble families in
Bavaria is established by the characters Wolfrat von Tengelingen, the
duke of Meran and the duke of Diessen, names that suggest an audience
332 nigel f. palmer
among the Bavarian lay nobility.19 The epilogue to Ulrich von
Zatzikhofen’s Lanzelet, from the end of the century, describes how an
English nobleman Hugh of Morville, who had been brought to the
German lands as a hostage in return for the freeing of Richard Lionheart
from captivity, was carrying the manuscript of the French source with him,
which the German poet, at the behest of a group of ‘friends’ (evidently
German nobles), rendered into German.20 The references cited here
almost all make specific mention of manuscript copies, while at the same
time implying a close association with lay patrons who would surely have
wanted to receive presentation copies of some kind, but there is no
evidence of surviving books that would appear to be directly associated
with the patrons. Two illuminated manuscripts of the Rolandslied (one
destroyed in 1870) provide an idea of what more prestigious copies of epic
poetry with historical subject matter might look like, but both these
manuscripts date long after the time of composition, and that in
Heidelberg (Figure 18.3), datable ca. 1200, has a Bavarian text, whereas
the court of the patron, Henry the Lion, was in Braunschweig (Lower
Saxony).21
A third factor in the development of German literary production in the late
twelfth century, the so-called ‘Blütezeit’, is the adaptation of French courtly
literature into German, not in this period close renderings, but rather free
adaptations. The earliest such work is Pfaffe Lambrecht’s Alexanderlied (ca.
1150/60), based on an Occitan source.22 It was followed by Pfaffe Konrad’s
Rolandslied (1170s), Heinrich von Veldeke’s Eneas (mid-1170s, completed
after ca. 1184), Hartmann von Aue’s Erec, Iwein and Gregorius
(ca. 1180–1205), Eilhart von Oberge’s Tristrant (1190s?), Herbort von
Fritslar’s Liet von Troye (ca. 1195–1215), Reinhart Fuchs by a certain Heinrich
(1190s), Ulrich von Zatzikhofen’s Lanzelet (author attested in 1214), Otte’s
Eraclius (1190–1230?), anonymous texts such as Graf Rudolf (1170–90?) and
Athis und Prophilias (ca. 1210?) and then in the early thirteenth century
Gottfried’s Tristan and the works of Wolfram von Eschenbach: Parzival,
Titurel and Willehalm. This same period also saw a flourishing of the courtly
love lyric in Middle High German in which the direct influence of northern
French and Occitan poetry combined with indigenous traditions.23
Yet a fourth factor is the development of the German four-beat couplet
verse with end-rhyme, which was used for vernacular texts from the late
eleventh century onwards, and during the period of direct French influence
came to be modified, abandoning the older aesthetic of assonance in favour
of greater regularity and pure end-rhyme. This development was
accompanied by the regularisation of the four-beat verse form with
 Vernacular Manuscripts II: Germany 333

Figure 18.3 Pfaffe Konrad, Rolandslied. Written area: 274–80 × 115–25 mm.
Heidelberg, Universitätsbibliothek, Cod. Pal. germ. 112, f. 41v.
334 nigel f. palmer
distinctive rhythmical patterns according to the kind of cadence employed,
and a more sophisticated entwinement of the couplet, which in the earlier
texts was the basic unit of meaning, with more complex syntactic struc-
tures. These refinements of German couplet verse achieved their most
distinctively elegant regular form in the language of Gottfried von
Strassburg’s Tristan, and in the thirteenth century a number of older
poems, such as the Kaiserchronik, Herzog Ernst, König Rother and
Eilhart’s Tristrant, were redacted in order to adapt the earlier assonance
and less regular metrical forms to current taste.24 These modifications of
the metrical form may well have had implications for the layout of verse in
the manuscripts, as is discussed later in this chapter. Of the narrative texts
mentioned here, the Alexanderlied, Veldeke’s Eneas, König Rother, Eilhart’s
Tristrant, Reinhart Fuchs and Graf Rudolf are all attested in manuscripts or
(mostly) fragments datable to ca. 1200 or the beginning of the thirteenth
century.
Finally, there is the emergence of a new type of author profile which
must have given rise to different expectations on the part of the authors as
to what they might expect from their scribes. Whereas the authors of texts
that circulated in the tenth and eleventh centuries were monks (notably
Notker the German of St. Gallen, Williram of Ebersberg and Otloh of St.
Emmeram) and the early biblical epics from the later eleventh and early
twelfth centuries and the mid-century Kaiserchronik remain anonymous,
a number of poetical texts from the mid-century onwards are presented as
the work of clerks (‘clericus’) or priests (‘plebanus’), whose learning and
evident knowledge of Latin (requiring access to manuscripts) can hardly
have been acquired in the context of an aristocratic court and without
access to monastic libraries. In some significant, but (as Bumke has
suggested) most likely exceptional cases, they present themselves as learned
knights (‘miles’, ‘ministerialis’ ‒ ‘ritter’/‘rîter’, ‘dienstman’).25 The author
of the Rolandslied (1170s) presents himself as a priest, ‘der pfaffe Chunrat’
(v. 9079), who translated the French text into Latin (implying he wrote out
the complete narrative in a Latin manuscript), and then from Latin into
German (vv. 9077–85). He claims to have used as his source a (French)
manuscript written in ‘Karlingen’ provided by Duke Heinrich (Henry the
Lion) at the behest of ‘the noble duchess, daughter of a mighty king’
(Matilda, daughter of King Henry II of England), thereby presenting
himself as a clerk and man of learning in the entourage of the duke.26
Similar authorial status, but in the plural, is implied in the A prologue of
the German prose encyclopaedia, the Lucidarius (late twelfth/early thir-
teenth century), where it is stated that inspired by God, the duke (‘herzoge
 Vernacular Manuscripts II: Germany 335
Heinrich’, Henry the Lion) instructed his chaplains (‘cappellane’) at the
court in Braunschweig to collect the necessary source materials from Latin
manuscripts and to write them up in German prose.27 When Herbort von
Fritslar, in his epilogue to the Liet von Troye, calls himself a ‘gelarter
schulere’ (‘learned clerk’, v. 18451), who composed his work at the behest
of Landgrave Hermann of Thuringia, basing himself on a French manu-
script exemplar provided by the Count of Leiningen (vv. 91–8), he is
presenting himself as a literate cleric in the entourage of the landgrave at
the Thuringian court.28 Hartmann von Aue’s often-quoted prologues to
Der arme Heinrich and Iwein strike out in a different direction. In Der arme
Heinrich Hartmann presents himself as a layman and learned knight (‘Ein
ritter sô gelêret was . . . ’ – ‘A knight was so learned . . . ’, v. 1) and
a ‘ministerialis’ of Aue (‘dienstman was er zOuwe’ – ‘he was a ministerial
at Aue’), who read widely in books in search of material that might grant
him relief in times of trouble (the ‘otium’ topos) as well as serving God’s
glory, seeking prayers for intercession from his readers (vv. 1–28).29
In Iwein he presents himself as Hartmann, of the family of Ouwe,
a learned knight who wrote poetry as a leisure-time activity, composing
works such as this Arthurian romance which he hopes his audience will
enjoy listening to (vv. 21–30).30 Gottfried von Strassburg refrains from
presenting himself as a socially defined individual, except in an acrostic
which entwines the names ‘Gote(vrit)’ and an unidentifiable ‘Dieterich’
(presumably a patron or dedicatee).31 When later poets call him ‘master’, as
in Konrad von Würzburg’s designation ‘Von Strâzburc meister Gotfrit’ in
his preface to the Herzmaere, a title which Heinrich von Veldeke had also
claimed for himself (Eneas v. 13465), this chimes with the self-presentation
of the authorial voice as a ‘poeta doctus’ who claims to have searched for
the right version of the Tristan story in ‘books of both kinds, French and
Latin’ (vv. 168–9), until he found what he needed with the work of Thomas
of Brittany. Furthermore, in his literary excursus (vv. 4555–5011), Gottfried
presents himself as an authority who is in a position to evaluate the most
sophisticated narrative poetry and song, and in doing so associates his own
work with a corpus of modern book literature which is entirely in the
German vernacular, certainly indebted in its rhetorical ambitions to Latin,
but nonetheless a cultural entity in itself: for Gottfried, around 1210,
German literature had emancipated itself. At about the same time
Wolfram von Eschenbach, in his Parzival, asserts the position of the
exception that proves the rule among the learned poets by styling himself
as an illiterate knight and minstrel, a man whose ‘office’ is that of the
shield, not the written word, and who narrates the story of Parzival to
336 nigel f. palmer
a female audience: his work, he says, can proceed without writing, without
book learning (115, 21–116, 4).32
A teleological reading of these twelfth-century developments might
suggest a development from monastic to secular literature, the transition
to a new autonomous literary aesthetic, and the extension of authorship
and reading practices from the monastic sphere to the world of secular
priests with noble patrons, leading in turn to a renaissance of lay literacy,
including female literacy, that includes both literary composition and
private reading of ‘German literature’. This would be a gross simplifica-
tion, and the contribution of codicological studies to providing the neces-
sary nuances needs to be accorded greater weight. Moving forward from
a cut-off date of ‘ca. 1200’ to ‘ca. 1225’, we are confronted with a number of
German manuscripts which mark a major move forward, particularly with
regard to the transmission of German verse. Throughout the eleventh and
twelfth centuries and up to about 1225, all of the numerous surviving
manuscripts and fragments of German verse texts present the text written
continuously rather than being set out in lines. This changes from the
period 1225 to 1250 onwards, when, apart from strophic verse and some of
the more conservative religious epics, the modern style of layout, according
to which the verses are set out in lines on the page (or as couplets with
the second verse indented) became the norm. This development repre-
sented a fundamental change in the attitude of scribes and readers to the
book as a physical object, or to how the book was seen to function in its
mediation between literary composition and reception (whether by private
reading or oral delivery in the form of reading to an audience).33 When the
text was set out in lines, the page layout became a material, visual mani-
festation of the poetic form employed, and when such a codex was used for
private reading, the visual presentation communicated something of what
in other circumstances might have been communicated by performance.
We do not know if any of the poets of the later twelfth or early thirteenth
centuries themselves were involved in this radical change in the presenta-
tion of German verse, but we certainly cannot exclude the possibility.
The question is of some significance because the new layout corresponded
rather closely to the standard practice for the vernacular in France and
Anglo-Norman England, and if (as we might surmise) the new presenta-
tion carried with it the resonances of French literary culture, the modern
layout, which the poets will have known from their exemplars, would be in
line with the emphasis they placed on the fact that the German composi-
tions were based on French codices. Nonetheless, most of our earlier
narrative texts have been preserved in one or more copies written in the
 Vernacular Manuscripts II: Germany 337
traditional German style, the only major exception being the Tristan of
Gottfried von Strassburg. The material dependency on manuscript copies
of the French poems is made explicit by some of the poets, notably in Pfaffe
Konrad’s Rolandslied, Veldeke’s Eneas and Ulrich’s Lanzelet. We should,
however, guard against the obvious temptation to see the history of verse
layout entirely in terms of dependency on French, and as I have argued
elsewhere, using the German Lanzelet as an example, the surviving manu-
scripts of German courtly romance show only sporadic dependency in their
page layout on French models.34
A manuscript that represents what could be achieved as a result of the
new developments in German literature is the Berlin copy of Heinrich von
Veldeke’s Eneas, executed ca. 1220–30.35 The complete poem consists of
13,528 lines of couplet verse (now wanting 11,492–13,528 through loss of the
last quire), presented here in a large-format manuscript containing just this
one text, and combining the text of the German poem with an extensive
picture cycle of exceptional quality, amounting to seventy-four leaves in all
(ca. 250 mm × 170 mm, written area ca. 230 mm × 155 mm). The text is in
three columns36 and occupies thirty-five leaves (quaternions), and the cycle
of 136 illustrations, executed as bifolia to be inserted into the text quires as
interleaving, is arranged, with one exception, in pairs, one above the other
on each page. The result is that, except at the middle of the quire, each text
page faces a page with two images. The choice of the three-column layout
no doubt served to compress the extent of the text and make possible the
extraordinarily lavish programme of pictures. Apart from some preliminary
experimentation on the opening pages, the text is set out in lines of verse,
the first verse of each couplet beginning with a majuscule and drawn out
into the margin, and each concluded with a rhyme point. This layout
emphasises the integrity of the metrical couplet, as distinct from the single
four-beat line, and came to be common from the middle of the thirteenth
century onwards. As in all other copies of the poem, the text is divided into
paragraphs by plain two-line red initials with no further decoration.
The calligraphy is relatively modest, hardly in line with the quite excep-
tional ambitions of the illustrative programme. It is all the work of one
scribe, which, given the extent of the poem, might speak against produc-
tion in a workshop where the copying of German manuscripts was routine
(notwithstanding the combination of work by craftsmen of two different
trades). Whereas Veldeke’s own language was the Low Franconian dialect
of the Maasland, and his work was completed for the Thuringian court in
the East Central German region, this manuscript, while displaying some
Central German features taken over from an exemplar, is written in the
338 nigel f. palmer

Figure 18.4 Hartmann von Aue, Iwein. Written area: 126 × ca. 84 mm. Gießen,
Universitätsbibliothek, Hs. 89, ff. 1v–2r.

Bavarian dialect.37 The manuscript thus documents the way German

literary texts were being transported and copied outside the region in
which they had originally been composed. Veldeke’s second German
literary composition in couplet verse, the Servatius, was composed at the
behest of Agnes, countess of Loon (vv. 6177–82), probably the wife of
Count Lodewijk I (d. 1171). It has plausibly been suggested that the
Wittelsbach connections to the Maasland and the Thuringian court
might provide a context in which the copying of the Eneas for a noble
and wealthy family in Bavaria might be situated.38
A second manuscript codex which, like the Berlin Eneas, illustrates the
new literary situation in the 1220s is the celebrated Gießen codex of
Hartmann’s Iwein (Figure 18.4), a volume of exceptionally small format
(126 mm × ca. 84 mm, written area ca. 92 mm × 55 mm), in which an
Arthurian romance of 8,166 lines of verse is presented as a single volume of
160 leaves copied in one column with twenty-six lines to the page.39
The verses are set out in lines, each beginning with an offset minuscule
(occasionally a majuscule) and concluded with a rhyme point, a common
 Vernacular Manuscripts II: Germany 339
layout which is also found in French manuscripts and which treats the line
of verse rather than the couplet as the basic unit. The poem is introduced
on f. 1r with an eight-line multicoloured initial and paragraphs are marked
by minor three-line initials, alternating red and blue, with fine decoration,
but only in exactly the first half of the manuscript, ff. 1r–79v (presumably
so planned for reasons of economy). The script is a Northern Textualis,
employing quite a number of abbreviations and with frequent accent
marks, perhaps as an aid to performance. The language is Eastern Upper
German (Eastern Swabian/Bavarian). There are no clues to suggest the
status of the patron for whom the manuscript was made, other than the
high quality of execution. The overall design of the book, containing just
a single text, and its small format, smaller than a modern German ‘Reclam’
edition, suggests that it was intended for private reading. That this type of
manuscript was thought appropriate for Arthurian romance in the early
thirteenth century is evident from the Linz fragment of Iwein, which is
only slightly larger than Gießen (164 mm × 97 mm, written area 112 mm ×
60 mm) and has broader margins, but makes use of the same single-column
layout with offset minuscules at the beginning of each line.40 Other early
Iwein manuscripts, such as the two-column fragments in Kremsmünster
(most likely [225 mm × 150 mm], written area 175 × 120 mm) and the single-
column fragment in Munich (BSB Cgm 191: 152 mm × 115 mm, written
area 120 mm × 80 mm), although adhering to the older style in that the
verses are written out continuously, provide further evidence for small-
format manuscripts for the romance, probably containing just a single
text.41 Such manuscripts demanded a different type of reading practice
from the larger, heavier codices such as the Berlin Eneas.
If we now turn back to the twelfth century proper, we are faced with
a very large body of manuscript production that can only with difficulty
be coordinated with the innovations in German literary culture that are
evident from the content, literary form and social history of the texts.
Several factors need to be borne in mind in sifting the evidence. First, as
we have noted repeatedly, the vernacular material is mostly presented in
a Latin context.42 Secondly, most of the evidence consists of fragments.
Thirdly, whereas the innovations that can be plotted against a social
history of German literary production tend to be associated with the
secular clergy at princely courts, the provenance of the surviving manu-
scripts is almost entirely monastic. The monasteries were the real home
of the foundational literary culture of the period, and it is in their
libraries that manuscripts survived, whether as codices or as scrap
parchment used in later bookbindings. And finally we should bear in
340 nigel f. palmer
mind that although German vernacular manuscripts can be assigned
a fairly accurate regional provenance on the basis of their language,
palaeographical datings have to be treated with considerable caution,
wherever possible allowing leeway of plus or minus fifteen to twenty
years.43
In all, perhaps fewer than twenty ‘German’ or bilingual Latin and
German codices have survived intact from the twelfth century, the rest of
the manuscripts from this period being interpolations in a Latin context or
fragments. The following list aims to be complete.44
• Vienna Notker (Psalms 1–50, 101–50) and Old High German sermons
(fragments), ca. 1100 (‘late eleventh/early twelfth century’);
• Leiden Williram, from Egmond (North Holland), ca. 1100;
• Trier Williram, from St. Matthias, Trier, eleventh/twelfth century;
• Kremsmünster Williram, first half twelfth century;
• London British Library Williram, first half twelfth century;
• Stuttgart Williram, first half twelfth century;
• Vienna Williram, possibly from Ebersberg, second quarter twelfth
century;
• Einsiedeln Williram, second quarter twelfth century;
• St. Gallen manuscript of Notker’s Boethius, from Einsiedeln, second
quarter twelfth century;
• Munich Williram, BSB Cgm 77, from Indersdorf, second half twelfth
century;
• Prayerbook of Muri, northern Switzerland, early in the last quarter,
perhaps ca. 1175;
• Windberg Psalter with interlinear translation, BSB Cgm 17, last quarter
twelfth century (1174?);
• Berlin Williram, from Lambach, ca. 1180;
• The Vorau Manuscript (an extensive collection, including the
Kaiserchronik), last quarter twelfth century;
• Vienna Genesis and Exodus, ÖNB 2721, last quarter twelfth century;
• Speculum ecclesiae sermon collection, late twelfth century;
• Zurich, Zentralbibl. cod. C 58 (Latin-German ‘Sammelhandschrift’),
late twelfth century/ca. 1200;
• Millstatt Manuscript (with Genesis, Physiologus and Exodus), ca. 1200;
• Heidelberg Rolandslied, ca. 1200;
• Interlinear version of the Psalter in Trier, ca. 1200;
• ‘Sammelband’ with the Strassburger Alexander (destroyed 1870), early
thirteenth century;
 Vernacular Manuscripts II: Germany 341
• König Rother in Heidelberg, early thirteenth century;
• Millstatt Psalter and Hymns from Admont (interlinear versions), early
thirteenth century;
• Priester Wernher’s Maria in Cracow (formerly in Berlin), early thir-
teenth century.
This list needs to be augmented, if we are to aim at an accurate picture
of the state of German vernacular manuscript production in the twelfth
century, by some of the early fragments, such as those of three different
manuscripts of the Mittelfränkische Reimbibel (Central Franconian
Rhymed Bible), two of which could be from the first half of the century,
the fragments in Freiburg i. Br. and Klagenfurt of the Kaiserchronik (the
first verse chronicle of the world in a Western European vernacular,
composed ca. 1150) from the third quarter of the century, and fragments
of what must have been two extensive collections of verse in Colmar
(with Crescentia, Schoph von dem lone and the Cantilena de conversione
Sancti Pauli, datable to the late twelfth century) and Trier (with the
verse legends of Saints Giles and Silvester, and the love romance of
Floyris und Blanscheflor, ca. 1200/beginning of the thirteenth century).45
The latter manuscript, combining religious and secular verse narrative,
is exceptional in that it seems to point outside the monastic sphere.
Finally, special mention should be made of several examples of religious
poetry, two of them texts intended to be sung, that were added on blank
pages of Latin manuscripts, most notably the Strasbourg copies of the
Ezzolied (Song of Ezzo) and Memento mori added to a manuscript of
Gregory’s Moralia in Job (first half of the twelfth century), the poem
Vom himelrîche added to a copy of the Moralia in Job from Oberaltaich
(datable 1174), the Arnsteiner Marienlied (a hymn to the Virgin) added
to a psalter from the Premonstratensian abbey of Arnstein on the Lahn
(last third of the twelfth century) and the Oxford manuscript of the
Mariensequenz aus Muri, with neumes (ca. 1200) and based on the Latin
sequence Ave praeclara maris stella, copied at the end of the calendar
preceding a manuscript of the Hirsau Liber ordinarius and a collection
of miracles of the Virgin made for the abbey of Moggio in Friuli.46 It is
in such manuscripts that the dominant twelfth-century phenomenon of
a bilingual literary culture becomes most immediately visible, although,
as Christine Putzo elaborated in a recent study, it in fact underlies all
written texts in the vernacular from the German lands for the greater
part of the Middle Ages: they constitute what Hugo Kuhn in 1967
christened a ‘volksschriftsprachliche Zwischenkultur’.47
342 nigel f. palmer
Notes
1. Hellgardt 1988. Karin Schneider’s foundational monograph, dealing with
Gothic script from ca. 1175 onwards, appeared a year earlier. Jürgen Wolf’s
Buch und Text, which treats our subject from a different perspective, came
twenty years later.
2. Palmer 2005a.
3. Available online from the ‘Marburger Repertorien’ (www.handschriftencen
sus.de); last accessed 25 November 2015.
4. Available online from the ‘Marburger Repertorien’ (www.paderborner-
repertorium.de); last accessed 25 November 2015.
5. For Latin-German text ensembles in this period, see Hellgardt 1992.
6. Bertelsmeier-Kierst 2000, 2003a.
7. For an important and forward-looking discussion of the relationship between
the scriptographic medium (manuscripts) and twelfth-century literature
understood in ‘anthropological’ terms, see Müller 2005.
8. First attested in the Vorau Manuscript (last quarter of the twelfth century);
see n. 44.
9. For the position of these two poems in the broader codicological context of
epic and romance, see Bumke 1996.
10. Gottfried von Strassburg: Tristan und Isold. For a codicological study, see
Wetzel 1992.
11. Wolfram von Eschenbach: Parzival, Willehalm. Cf. Heinzle 2011, esp. 1.
308–65 (Schirok), 1. 591–652 (Gerhardt) and the list of manuscripts by Klein
(2. 941–1002).
12. See the essays in the collected writings of Heinzer 2008; for contributions in
English, see Beach, Manuscripts and Monastic Culture.
13. Das St. Trudperter Hohelied; Die Millstätter Predigten, ed. Schiewer, with
a discussion of the German texts and manuscripts associated with Admont at
xxii–xxviii.
14. See Palmer 2017.
15. Das Rolandslied des Pfaffen Konrad. Cf. Bumke, Mäzene, 85–91; Bumke 1986,
2. 638–77 (English transl. Bumke 1991, 458–88); Kartschoke 1989.
16. Heinric van Veldeken: Sente Servas. Cf. Goossens 1991; Scheepsma 2013,
310–5.
17. Heinrich von Veldeke: Eneasroman. For the circumstances of composition of
Veldeke’s work, see Bumke, Mäzene, 113–8; Bumke 1986, 2. 681–2 (English
transl. Bumke 1991, 491–2).
18. Heinrich der Glîchezâre [sic]: Reinhart Fuchs. Cf. Bumke, Mäzene,
100–5.
19. König Rother. Cf. Bumke, Mäzene, 91–6.
20. Ulrich von Zatzikhoven: Lanzelet. Cf. Bumke, Mäzene, 153–4; McLelland
2000, 3–27.
21. Schneider 1987, 1. 79–81. Online from Heidelberg University Library.
 Vernacular Manuscripts II: Germany 343
22. For this, and all the other adaptations from French, see the Verfasserlexikon;
Bumke, Mäzene; and the list of ‘Twelfth-century secular narrative’ texts in
Palmer 2005, 43–5.
23. For the (much later) transmission of the Middle High German lyric, see
Holznagel 1995.
24. For modernisation in the manuscripts of König Rother, see Palmer 2011.
25. Bumke, Mäzene, 68–72, suggesting that the non-clerical authors were very
much exceptions to the rule (71).
26. Kartschoke 1989.
27. Lucidarius, 102*–3*. For the controversy about the status of the prologues,
dating and patronage, see Steer 1990; Bumke 1995; Bertelsmeier-Kierst
2003.
28. Herbort von Fritslar, Liet von Troye.
29. Hartmann von Aue, Der arme Heinrich.
30. Hartmann von Aue, Iwein.
31. Gottfried von Strassburg, Tristan und Isold, vv. 1–40. For author and patron,
see the editors’ comments, 2. 207–11.
32. Wolfram von Eschenbach, Parzival.
33. Schneider 1987, 1. 91–2 and 99–100; Palmer 1993, 2005a, 86–91; Putzo 2009,
317–8 (with a comparison of the layout of verse in French and German
manuscripts in this period); Palmer 2010, 66–9.
34. Palmer 2010.
35. Facsimile ed. Fingernagel and Henkel 1992. Cf. Schneider 1987, 1. 96–9; Wolf
2008, 91, 129.
36. For Middle High German manuscripts in three columns, see Klein 2000.
37. Klein 1988, 136–9. Klein’s article contains an important discussion of the
topography of manuscript distribution observable for German poetic texts in
the High Middle Ages.
38. Bumke Mäzene, 116–8, who also considers the possibility that this was Agnes,
the wife of Count Otto V of Scheyern, from whom the counts of Wittelsbach
were descended and whose daughter Sophie was married to Landgrave
Hermann I of Thuringia. See note 17.
39. Facsimile ed. Heinrichs 1964. Cf. Schneider 1987, 1. 147–9; Klein 1988, 148–50
(with an analysis of the language); Palmer 1993, 16; Wolf 2008, 128, 301–2.
40. For the Linz fragment, see Okken 1974, 17–8.
41. For the Iwein fragments in Munich and Kremsmünster, see Petzet and
Glauning, part 3, 1912, pl. 31B, and Wiesinger 1978, plate after 194;
Wiesinger 1984.
42. For different types of bilingual collective manuscript, see Hellgardt 1992,
29–31; Wolf, 56–65 (with illustrations), who distinguishes four different
types, suggesting that, as in France and England, during the eleventh and
the first half of the twelfth centuries some 90 per cent of all manuscripts with
vernacular texts were bilingual, in the second half of the twelfth century
75 per cent.
43. Palmer 2005a, 92.
344 nigel f. palmer
44. For all these texts, see the Verfasserlexikon and in many cases the detailed
discussion in Schneider 1987. Furthermore, for the Vienna Notker, see
Hellgardt 2014. For the Williram manuscripts, see Gärtner 1988. For the
Vorau Manuscript, a classic bilingual text ensemble presenting a collection
of more than twenty extensive German poems and Otto of Freising’s Gesta
imperii side by side, in a large format (450 mm × 325 mm, written area 340 mm
× 230–5 mm), see Schneider 1987, 1. 37–41; Grubmüller 2000; Grote 2012. For
the Vienna Genesis, see Gutfleisch-Ziche 1997.
45. Such large collections of German verse texts are well documented for the later
twelfth century, although mostly by fragments; see Grote 2012.
46. Heinzer 2008, 104–5.
47. Kuhn 1968, 5; Putzo 2011, 5.
 Bibliography

Introduction
Abulafia, A. S. 2006, ‘Intellectual and cultural creativity’, in The Central Middle
Ages, ed. D. Power (Oxford), 149–77.
Benson, R. L. and G. Constable with C. D. Lanham. 1982, The Renaissance of the
Twelfth Century (Oxford).
Cahn, W. 1996, Romanesque Manuscripts: The Twelfth Century (a survey of
manuscripts illuminated in France: 2 vols., London).
Damian-Grint, P. 1999, The New Historians of the Twelfth-Century Renaissance:
Inventing Vernacular Authority (Woodbridge).
De Hamel, C. 1984, Glossed Books of the Bible and the Origins of the Paris Booktrade
(Woodbridge).
Derolez, A. 2003, The Palaeography of Gothic Manuscript Books: From the Twelfth
to the Early Sixteenth Century (2003).
Donovan, C. 1993, The Winchester Bible (Toronto).
Ganz, D. 1995, ‘Book production in the Carolingian Empire and the spread of
Caroline minuscule’, in The New Cambridge Medieval History, 2: c. 700–
c. 900, ed. R. McKitterick (Cambridge), 786–808.
Gullick, M. 1990, ‘The scribe of the Carilef Bible: A new look at some late-
eleventh-century Durham Cathedral manuscripts’, in Medieval Book
Production, 61–83.
Hall, S. G. 2004, ‘In the beginning was the codex: The early Church and its
revolutionary books’, in The Church and the Book, ed. R. N. Swanson
(Woodbridge), 1–10.
Haskins C. H. 1927, The Renaissance of the Twelfth Century (Cambridge, MA).
Jaeger, C. S. 1994, The Envy of Angels: Cathedral Schools and Social Ideals in
Medieval Europe (Philadelphia, PA).
Kauffmann, C. M. 1975, Romanesque Manuscripts 1066–1190 (A Survey of
Manuscripts Illuminated in the British Isles 3: London).
Ker N. R. 1960, English Manuscripts in the Century after the Norman Conquest
(Oxford).
Knowles, D. 1962, The Evolution of Medieval Thought (London).

345
346 Bibliography
Luscombe, D. 2004, ‘Thought and learning’, in The New Cambridge Medieval
History, 4: c. 1024–c. 1198, ed. D. Luscombe and J. Riley-Smith (Cambridge),
461–98.
Palmer, N. F. 1998, Zisterzienser und Ihre Bücher: Die mittelalterliche
Bibliotheksgeschichte von Kloster Eberbach im Rheingau (Regensburg).
Roberts, C. H. and T. C. Skeat 1983, The Birth of the Codex (London).
Sheppard, J. M. 1990, ‘Some twelfth-century monastic bindings and the question
of localization’, in Medieval Book Production, 181–98.
Swanson, R. N. 1999, The Twelfth-Century Renaissance (Manchester).
Thomson R. M. 1985, Manuscripts from Saint Abbey 1066 –1235 (2nd edn., 2 vols.,
Woodbridge).
1998, England and the Twelfth-Century Renaissance (Aldershot).
2006, Books and Learning in Twelfth-Century England: The Ending of ‘Alter
Orbis’ (The Lyell Lectures in Bibliography 2000–1: Walkern).
Verger, J. 1999, ‘The universities and scholasticism’, in The New Cambridge
Medieval History, 5: c. 1198–c. 1300, ed. D. Abulafia (Cambridge), 256–76.

1 Codicology

Dimensions and Layout

Ayres, L. 1994, ‘The Italian Giant Bibles: Aspects of their Touronian ancestry and
early history’, in The Early Medieval Bible: Its Production, Decoration and Use,
ed. R. Gameson (Cambridge), 125–54.
Bozzolo, C. and E. Ornato 1983, Pour une histoire du livre manuscrit au Moyen Âge:
Trois essais de codicologie quantitative (Paris).
De Hamel, C. 2001, The Book: A History of the Bible (London).
Gumbert, J. P. 1999, ‘The layout of the Bible gloss in manuscript and early print’,
in The Bible as Book: The First Printed Editions, ed. P. Saenger and K. van
Kampen (London), 7–13.
Huglo, M. 2001, ‘The cantatorium: From Charlemagne to the fourteenth cen-
tury’, in The Study of Medieval Chant: Paths and Bridges, East and West. In
Honor of Kenneth Levy, ed. P. Jeffery (Rochester, NY), 89–104.
Kwakkel, E. 2012, ‘Biting, kissing and the treatment of feet: The transitional script
of the long twelfth century’, in Turning Over a New Leaf, 78–126, 206–8.
2012a, ‘Dit boek heeft niet de vereiste breedte: Afwijkende bladdimensies in de
elfde en twaalfde eeuw’, Jaarboek Nederlandse Boekgeschiedenis 19, 33–49.
2015, ‘Decoding the material book: Cultural residue in medieval manuscripts’,
in The Medieval Manuscript Book: Cultural Approaches, ed. M. Van Dussen
and M. Johnson (Cambridge), 60–76.
Palmer, N. F. 2010, ‘Simul cantemus, simul pausemus: Zur mittelalterlichen
Zisterzienserinterpunktion’, in Lesevorgänge: Prozesse des Erkennens in mitte-
lalterlichen Texten, Bildern und Handschriften, ed. E. C. Lutz et al. (Zurich),
483–569.
 Bibliography 347
Parkes, M. B. 1991, ‘The influence of the concepts of ordinatio and compilatio on
the development of the book’, in M. B. Parkes, Scribes, Scripts and Readers:
Studies in the Communication, Presentation and Dissemination of Medieval
Texts (London), 35–70.
2008, ‘Layout and presentation of the text’, in CHBB II, 55–74.
Reilly, D. J. 2012, ‘Art’, in The Cambridge Companion to the Cistercian Order, ed.
M. Bruun (Cambridge), 125–39.
Reynolds, S. 1996, Medieval Reading: Grammar, Rhetoric and the Classical Text
(Cambridge).
1996a, ‘Glossing Horace: Using the classics in the medieval classroom’, in
Chavannes-Mazel and Smith, Latin Classics, 103–17.
Robinson, P. R. 2008, ‘The format of books: Books, booklets and rolls’, in CHBB
II, 41–54.
Rouse, M. A. and R. H. Rouse 1982, ‘Statim invenire: Schools, preachers and new
attitudes to the page’, in Benson and Constable, Renaissance, 201–25; repr. in
their Authentic Witnesses, 191–219.
2000, Manuscripts and Their Makers: Commercial Book Producers in Medieval
Paris 1200–1500 (2 vols., London).
Smith, L. 2009, The Glossa Ordinaria: The Making of a Medieval Bible
Commentary (Leiden).

Parchment
Brown, T. J. 1993, ‘The distribution and significance of membrane prepared in the
Insular manner’, in La paléographie hébraïque médiévale (Colloques
Internationaux du CNRS, 547: Paris, 1974), 127–35; repr. in A
Palaeographer’s View: The Selected Writings of Julian Brown, ed. J. Bateley,
M. P. Brown and J. Roberts (London), 1. 125–39.
Clemens, R. and T. Graham 2007, Introduction to Manuscript Studies (Ithaca, NY).
De Hamel, C. 1992, Scribes and Illuminators (London).
Gullick, M. 1991, ‘From parchmenter to scribe: Some observations on the man-
ufacture and preparation of medieval parchment based upon a review of the
literary evidence’, in Pergament: Geschichte, Struktur, Restaurierung,
Herstellung heute, ed. P. Rück (Sigmaringen), 145–57.
Hunter, D. 1947, Papermaking: The History and Technique of an Ancient Craft, 2nd
edn. (New York).
Kwakkel, E. 2003, ‘A new type of book for a new type of reader: The emergence of
paper in vernacular book production’, The Library 4, 7th ser., 219–48.
Thompson, D. V. 1936, The Materials and Techniques of Medieval Painting
(London; repr. New York, 1957).
Thomson, R. M. 2008, ‘Technology of production of the manuscript book I:
Parchment and paper, ruling and ink’, in CHBB II, 75–8.
348 Bibliography
Bindings
Borrie, M. F. 1968, ‘The binding of the Sherborne Cartulary’, British Museum
Quarterly 32, 96–8.
Clarkson, C. 2013, ‘English monastic bookbinding in the twelfth century’, in Der
Albani-Psalter, ed. J. Bepler and C. Heitzmann (Hildesheim), 177–99.
De Hamel, C. 1984, ‘Romanesque bindings and glossed books of the Bible’, in his
Glossed Books, 64–86.
Doyle, I. A. 1972, ‘Further observations on Durham Cathedral Ms. A. IV. 34’, in
Varia Codicologica: Essays Presented to G. I. Lieftinck, 1, ed. P. Gumbert and
M. de Haan (Litterae Textuales: Amsterdam), 35–47.
Fingernagel, A. (ed.) 2007, Geschichte der Buchkultur, 4: Romanik (2 vols.,
Graz).
Ganz, D. 2014, Buch-Gewänder: Prachteinbände im Mittelalter (Berlin).
Gullick, M. 1996, ‘From scribe to binder: Quire tackets in twelfth-century
European manuscripts’, in Roger Powell: The Compleat Binder, ed.
J. Sharpe (Turnhout), 240–59.
2000, ‘A Romanesque blind-stamped binding at the Queen’s College, Oxford’,
in ‘For the Love of Binding’: Studies in Bookbinding History Presented to
Mirjam Foot (London and Newcastle).
2012, ‘Bookbindings’, in CHBB I, 294–309.
Gullick, M. and N. Hadgraft 2008, ‘Bookbindings’, in CHBB II, 95–109.
Gumbert, J. P. 2011, ‘The tacketed quire: An exercise in comparative codicology’,
Scriptorium 64, 299–320 and plates 50–4.
Hobson, G. D. 1929, English Binding before 1500 (Cambridge).
1988, Studies in the History of Bookbinding (London).
Huws, D. 1987, ‘The making of the Liber Landavensis’, National Library of Wales
Journal 25, 133–60.
James, M. R. 1907–14, A Descriptive Catalogue of the Manuscripts in the Library of
Gonville and Caius College (3 vols., Cambridge).
Le Neve, J. 1968, Fasti Ecclesiae Anglicanae, 1066–1300, I: St. Paul’s London, ed.
D. Greenway (London).
Loubier, H. 1926, Der Bucheinband: Von seinen Anfängen bis zum Ende des 18.
Jahrhunderts, 2nd edn. (Leipzig).
Mynors, R. A. B. and R. M. Thomson. 1993, Catalogue of the Manuscripts in
Hereford Cathedral Library (Cambridge).
Pollard, G. 1975, ‘Some Anglo-Saxon bookbindings’, The Book Collector 24,
130–59.
Powell, R. and P. Waters 1969, ‘Technical Description of the Binding’, in The
Stonyhurst Gospel, ed. T. J. Brown, (Oxford), 45–55.
Schmidt-Künsemüller, F. A. 1985, Die abendländischen romanischen
Blindstempeleinbände (Stuttgart).
Scholla, A. 2002, ‘Libri sine asseribus: Zur Einbandtechnik, Form und Inhalt
mitteleuropä ischer Koperte des 8. bis 14. Jahrhunderts’, Unpublished doc-
toral dissertation, Leiden.
 Bibliography 349
Szirmai, J. 1999, The Archaeology of Medieval Bookbinding (Aldershot).
Thomson, R. M. 2001, A Descriptive Catalogue of the Medieval Manuscripts in
Worcester Cathedral Library (Cambridge).
Wormald, F. and C. E. Wright 1958, The English Library before 1700 (London).

2 Book Script in the Twelfth Century

Bischoff, B. 1953, ‘La nomenclature des écritures livresques du IXe au XIIIe siècle’,
in Nomenclature des écritures livresques du IXe au XVIe siècle (Paris), 7–14.
1979, Paläographie des römischen Altertums und des abendländischen Mittelalters
(Berlin) (English transl. Bischoff, Latin Palaeography).
Brown, M. P. 2002, A Guide to Western Historical Scripts from Antiquity to 1600;
repr. 1993 (London).
Brown, T. J. 1993, A Palaeographer’s View: The Selected Writings of Julian Brown,
ed. J. Bately, M. P. Brown and J. Roberts (London).
Cohen-Mushlin, A. 2010, ‘A school for scribes’, in Teaching Writing, Learning to
Write: Proceedings of the XVIth Colloquium of the Comité international de
paléographie latine, ed. P. R. Robinson (London), 61–87.
Kwakkel, E. 2012, ‘Kissing, biting and the treatment of feet: The transitional script
of the long twelfth century’, in Turning Over a New Leaf, 78–126, 206–8.
2013, ‘Hidden in plain sight: Continental scribes in Rochester Cathedral Priory,
1075–1150’, in Writing in Context: Insular Manuscript Culture, 500–1200, ed.
E. Kwakkel (Leiden), 231–61.
‘The digital eye of the palaeographer’, in Digital Palaeography, ed. S. Brookes,
M. Rehbein and P. Stokes (Routledge, in press).
Lowe [Loew], E. A. 1999, The Beneventan Script: A History of the South Italian
Minuscule. Special edition for Sandpiper Books (Oxford).
Parkes, M. B. 2008, ‘Handwriting in English books’, in CHBB II, 110–35.
2008a, Their Hands Before Our Eyes: A Closer Look at Scribes (Aldershot).
Roberts, J. 2005, Guide to Scripts Used in English Writings up to 1500 (London).
Schneider, K. 1987, Gotische Schriften in deutscher Sprache, I: Vom späten 12.
Jahrhundert bis um 1300 (Wiesbaden).
1999, Paläographie und Handschriftenkunde für Germanisten: Eine Einführung
(Tübingen).
Steinmann, M. 2010, ‘Lesen und Schreiben in den Klöstern des frühen
Mittelalters’, in Teaching Writing, Learning to Write: Proceedings of the
XVIth Colloquium of the Comité international de paléographie latine, ed.
P. R. Robinson (London), 25–35.
Waller, K. M. 1984, ‘Rochester Cathedral Library: An English book collection
based on Norman models’, in Les mutations socio-culturelles au tournant des
XIe–XIIe siècles, ed. R. Foreville (Paris), 237–51.
Webber, T. 1995, ‘Script and manuscript production at Christ Church,
Canterbury, after the Norman Conquest’, in Canterbury and the Norman
Conquest: Churches, Saints and Scholars, ed. R. Eales and R. W. Sharpe
(London), 145–58.
350 Bibliography
2011, ‘The Norman Conquest and handwriting in England to 1100’, in CHBB I,
211–24.

3 Decoration and Illustration

Abou-el-Haj, B. 1994, The Medieval Cult of Saints: Formations and Transformations
(Cambridge).
Alexander, J. J. G. 1978, The Decorated Letter (London).
1978a, ‘Scribes as artists: The arabesque initial in twelfth-century English
manuscripts’, in Ker, Essays, 87–116.
1992, Medieval Illuminators and Their Methods of Work (New Haven, CT).
1980, Norman Illumination at Mont St Michel 966–1100 (Oxford).
Barral i Altet, X. (ed.) 1986–90, Artistes, artisans et production artistique au Moyen
Âge (3 vols., Paris).
Belting, H. 1994, Likeness and Presence: A History of the Image before the Era of Art,
trans. E. Jephcott (Chicago, IL).
Bepler, J., P. Kidd and J. Geddes 2008, The St Albans Psalter (Albani Psalter):
Facsimile and Commentary (2 vols., Simbach am Inn).
Boeckler, A. 1924, Die Regensburg-Prüfeninger Buchmalerei des XII. und XIII.
Jahrhunderts (Munich).
Bovey, A. and J. Lowden 2007, Under the Influence: The Concept of Influence and
the Study of Illuminated Manuscripts (Turnhout).
Buchthal, H. 1979, The ‘Musterbuch’ of Wolfenbüttel and Its Position in the Art of the
Thirteenth Century (Vienna).
Butz, A. and S. von Borries-Schulten 1987, Die Romanischen Handschriften der
Württembergischen Landesbibliothek Stuttgart (2 vols., Stuttgart).
Cahn, W. 1982, Romanesque Bible Illumination (Ithaca, NY).
Romanesque Manuscripts.
Camille, M. 1985, ‘Seeing and reading: Some visual implications of medieval
literacy and illiteracy’, Art History 8, 26–49.
Carruthers, M. 2008, The Book of Memory: A Study of Memory in Medieval Culture,
2nd edn. (Cambridge).
Cassidy, B. 1993, Iconography at the Crossroads (Princeton, NJ).
Cavallo, G. 1973, Rotoli di Exultet dell’Italia meridionale (Bari).
Caviness, M. 1983, ‘Images of divine order and the third mode of seeing’, Gesta 22,
99–120.
Clanchy, M. 2012, From Memory to Written Record: England, 1066–1307, 3rd edn.
(Chichester).
Clark, W. B. 2006, A Medieval Book of Beasts: The Second-Family Bestiary
(Woodbridge).
Clarke, M. 2001, The Art of All Colours: Mediaeval Recipe Books for Painters and
Illuminators (London).
Collins, M. 2000, Medieval Herbals: The Illustrative Traditions (London and
Toronto).
De Hamel, C. 1992, Medieval Craftsmen: Scribes and Illuminators (London).
Demus, O. 1970, Byzantine Art and the West (New York).
 Bibliography 351
Derolez, A. 1968, Lamberti S. Audomari canonici Liber Floridus: Codex autographus
Bibliothecae Universitatis Gandavensis (Ghent).
Dodwell, C. R. 1954, The Canterbury School of Illumination, 1066–1200
(Cambridge).
1993, The Pictorial Arts of the West, 800–1200 (New Haven, CT).
Evans, M. W. 1969, Medieval Drawings (London).
Fingernagel, A. (ed.) 2007, Romanik (Geschichte der Buchkultur, 4: 2 vols., Graz).
Gameson, R. 1999, ‘Manuscrits normands à Exeter aux XIe et XIIe siècles’, in
Manuscrits et enluminures dans le monde normande (XIe–XVe siècle), ed.
P. Bouet and M. Dosdat (Caen), 107–27.
Glorieux-De Gand, T. 1990, Manuscrits cisterciens de la Bibliothèque royale de
Belgique (Brussels).
Grabar, A. and C. Nordenfalk 1958, Romanesque Painting from the Eleventh to the
Thirteenth Century, trans. S. Gilbert (Lausanne).
Green, R., M. W. Evans, C. Bischoff and M. Curschmann 1979, Herrad of
Hohenbourg, Hortus Deliciarum (2 vols., London and Leiden).
Gullick, M. 2006, ‘Self-referential portraits of artists and scribes in Romanesque
manuscripts’, in M. Gullick (ed.), Pen in Hand: Medieval Scribal Portraits,
Colophons and Tools (Walkern), 97–114.
Hahn, C. 2001, Portrayed on the Heart: Narrative Effect in Pictorial Lives of Saints
from the Tenth through the Thirteenth Century (Berkeley and Los Angeles, CA,
London).
Hamburger, J. F. and A.-M. Bouché (eds.) 2006, The Mind’s Eye: Art and
Theological Argument in the Middle Ages (Princeton, NJ).
Hamburger, J. F. 2014, Script as Image (Louvain).
Heinzer, F. (ed.) 1992, Der Landgrafenpsalter (2 vols., Graz).
Heslop, T. A. 1986, ‘“Brief in words but heavy in the weight of its mysteries”’, Art
History 9, 1–11.
1990, ‘Romanesque painting and social distinction: The Magi and the
Shepherds’, in England in the Twelfth Century, ed. D. Williams
(Woodbridge), 137–52.
Hoffmann, K. (ed.) 1970. The Year 1200: A Centennial Exhibition at the
Metropolitan Museum of Art (2 vols., New York).
Holcomb, M. (ed.) 2009, Pen and Parchment: Drawing in the Middle Ages (New York).
Hourihane, C. (ed.) 2008, Romanesque Art and Thought in the Twelfth Century:
Essays in Honor of Walter Cahn (Princeton, NJ).
James, M. R. 1929, Marvels of the East (Roxburghe Club 177: Oxford).
Jones, L. W. and C. R. Morey 1931, The Miniatures of the Manuscripts of Terence
prior to the Thirteenth Century (2 vols., Princeton, NJ).
Katzenellenbogen, A. 1939, Allegories of the Virtues and Vices in Medieval Art from
Early Christian Times to the Thirteenth Century (London).
Kauffmann, C. M. 1975, Romanesque Manuscripts, 1066–1190 (A Survey of
Manuscripts Illuminated in the British Isles, 3: London).
2003, Biblical Imagery in Medieval England, 700–1550 (London and Turnhout).
352 Bibliography
Klemm, E. 1980, 1988, Die Romanischen Handschriften der Bayerischen
Staatsbibliothek (4 vols., Wiesbaden).
Kölzer, T. and M. Stähli (eds.) 1994, Petrus de Ebulo: Liber ad honorem Augusti sive
de rebus Siculis. Codex 120 II der Burgerbibliothek Bern: Eine Bilderchronik der
Stauferzeit (Sigmaringen).
Kötzsche, D. 1989, Das Evangeliar Heinrichs des Löwen (2 vols., Frankfurt).
Lawrence, A. 1995, ‘Cistercian decoration: Twelfth-century legislation on illumi-
nation and its interpretation in England’, Reading Medieval Studies 21, 31–52.
Lawrence-Mathers, A. 1996, Manuscripts in Northumbria in the Eleventh and
Twelfth Centuries (Woodbridge).
Legner, A. 1985, Ornamenta Ecclesiae: Kunst und Künstler der Romanik, (3 vols.,
Cologne).
MacKinney, L. 1965, Medical Illustrations in Medieval Manuscripts (London).
Melnikas, A. 1975, The Corpus of the Miniatures in the Manuscripts of Decretum
Gratiani (3 vols., Rome).
Michon, S. 1990, Le Grand Passionnaire enluminé de Weissenau et son scriptorium
autour de 1200 (Geneva).
Murdoch, J. E. 1984, Album of Science: Antiquity and the Middle Ages (New York).
Mütherich, F. and K. Dachs 1987, Regensburger Buchmalerei: Von
frühkarolingischer Zeit bis zum Ausgang des Mittelalters (Munich).
Norman, J. 1988, Metamorphoses of an Allegory: The Iconography of the
Psychomachia in Medieval Art (New York).
Oakeshott, W. 1981, The Two Winchester Bibles (Oxford).
Pächt, O. 1956, ‘The illustration of St. Anselm’s Prayers and Meditations’, JWCI
19, 68–83.
Pächt, O., C. R. Dodwell and F. Wormald 1960, The St. Albans Psalter (London).
Petzold A. 1999, ‘“Of the significance of colours”: The iconography of colour in
Romanesque and early Gothic book illumination’, in Image and Belief:
Studies in Celebration of the Eightieth Anniversary of the Index of Christian
Art, ed. C. Hourihane (Princeton, NJ), 125–34.
Putatoro Donati Murano, A. and A. Perriccioli Saggese 2005, La Miniatura in
Italia, I: Dal tardoantico al Trecento con riferimenti al Medio Oriente e
all’Occidente europeo (Naples).
Reinecke, H., K. Reinecke and D. Tivig 1998, Buchmalerei der Zisterzienser:
Kulturelle Schätze aus Sechs Jahrhunderten (Stuttgart).
Riedmaier, J. 1994, Die ‘Lambeth Bibel’: Struktur und Bildaussage einer englischen
Bibelhandschrift des 12. Jahrhunderts (Frankfurt).
Rudolph, C. 1990, The ‘Things of Greater Importance’: Bernard of Clairvaux’s
Apologia and the Medieval Attitude toward Art (Philadelphia, PA).
Saxl, F. 1957, ‘Macrocosm and microcosm in mediaeval pictures’, in Lectures
(2 vols., London), 1. 58–72.
Scheller, R. W. 1995, Exemplum: Model-Book Drawings and the Practice of Artistic
Transmission in the Middle Ages (ca. 900–ca. 1470), trans. M. Hoyle
(Amsterdam).
 Bibliography 353
Shepard D. M. 2007, Introducing the Lambeth Bible: A Study of Texts and Imagery
(Turnhout).
Seeberg, S. 2002, Die Illustrationen im Admonter Nonnenbrevier von 1180.
Marienkrönung und Nonnenfrömmigkeit: Die Rolle der Brevierillustration in
der Entwicklung von Bildthemen im 12. Jahrhundert (Wiesbaden).
Stettiner, R. 1895, 1905, Die illustrierten Prudentiushandschriften (2 vols., Berlin).
Teviotdale, E. C. 2001, The Stammheim Missal (Los Angeles, CA).
Van der Horst, K., W. Noel and W. C. M. Wüstefeld 1996, The Utrecht Psalter in
Medieval Art: Picturing the Psalms of David (Westrenen).
Weinryb, I. 2013, 'Living matter: Materiality, maker and ornament in the Middle
Ages” Gesta 52/2, 113-132.
Williams, J. 2002–3, The Illustrated Beatus. A Corpus of the Illustrations of the
Commentary on the Apocalypse, 4: The Eleventh and Twelfth Centuries, 5: The
Twelfth and Thirteenth Centuries (London and Turnhout).
Wirth, K.-A. 2006, Pictor in Carmine: Ein Handbuch der Typologie aus der Zeit um
1200 (Berlin).

4 Scribes and Scriptoria

Alexander, J. J. G. 1978, ‘Scribes as artists: The arabesque initial in twelfth-century
English manuscripts’, in Ker, Essays, 87–116.
Beach, A. 2002, ‘Voices from a distant land: Fragments of a twelfth-century nuns’
letter collection’, Speculum 77, 34–54.
Bernards, M. 1981, ‘Speculum Virginum’: Geistigkeit und Seelenleben der Frau in
Hochmittelalter, 2nd edn. (Cologne).
Cahn, W. 1996, Romanesque Manuscripts: The Twelfth Century (a survey of manu-
scripts illuminated in France: 2 vols., London).
Cohen-Mushlin, A. 1983, The Making of a Manuscript: The Worms Bible of 1148
(Wiesbaden).
1990, A Medieval Scriptorium: Sancta Maria Magdalena de Frankendal (2 vols.,
Wiesbaden). A summary version is her ‘The Twelfth-Century Scriptorium at
Frankenthal’, in Medieval Book Production, 85–101.
2004, Scriptoria in Medieval Saxony: St. Pancras in Hamersleben
(Wiesbaden).
Delisle, L. 1909, Le rouleau mortuaire de bienheureux Vital, abbé de Savigny (Paris).
Dodwell, C. R. 1959, The Great Lambeth Bible (London).
Egbert, V. W. 1967, The Medieval Artist at Work (Princeton, NJ).
Gameson, R. 1999, ‘Manuscrits normands à Exeter aux XIe et XIIe siècles’, in
Manuscrits et enluminures dans le monde normande (Xe–XVe siècles), ed.
P. Bouet and M. Dosdat (Caen), 107–27.
Gullick, M. 1990, ‘The scribe of the Carilef Bible: A new look at some late-
eleventh-century Durham Cathedral manuscripts’, in Medieval Book
Production, 61–83.
354 Bibliography
1995, ‘How fast did scribes write?’, in Making the Medieval Book: Techniques of
Production, ed. L. Brownrigg, (Los Altos Hills, CA), 39–58.
1998, ‘The hand of Symeon of Durham: Further observations on the Durham
Martyrology Scribe’, in Symeon of Durham: Historian of Durham and the
North, ed. D. Rollason (Stamford), 14–31.
1998a, ‘The scribal work of Eadmer of Canterbury to 1109’, Archaeologia
Cantiana 118, 173–89.
2006, 'Self-referential artist and scribe portraits in Romanesque manuscripts',
in Pen in Hand: Medieval Scribal Portraits, Colophons and Tools, ed.
M. Gullick (Walkern), 97–114.
Härtel, H. 2006, Geschrieben und Gemalt. Gelehrte Bücher aus Frauenhand:
Eine Klosterbibliothek sächsischer Benediktinerinnen des 12. Jahrhunderts
(Wiesbaden).
Hotchin, J. 2007, ‘Women’s reading and monastic reform in twelfth-century
Germany: The library of the nuns of Lippoldsberg’, in Beach, Manuscripts
and Monastic Culture, 140–89.
Ker, N. R. 1980, ‘The correction of mistakes in twelfth-century manuscripts,
illustrated from Winchcombe books’, in Manuscripts at Oxford: R. W.
Hunt Memorial Exhibition, ed. B. Barker-Benfield and A. de la Mare
(Oxford), 30–2.
1985, ‘The beginnings of Salisbury Cathedral library’, in Hunt Essays, 23–49;
repr. in Ker, BCL, 143–73.
Liber Ordinis Sancti Victoris Parisiensis, ed. La. Jocqué and L. Milis (CCCM 61:
Turnhout, 1984).
Oakeshott, W. 1945, The Artists of the Winchester Bible (London).
Pächt, O. 1950, ‘Hugo Pictor’, Bodleian Library Record 3, 96–103.
Paris, M. 1639, Gesta Abbatum Sancti Albani, ed. J. Wats (Paris).
Rouse, R. H. and M. A. Rouse 1997, ‘Wandering scribes and traveling artists:
Raulinus of Fremington and his Bolognese Bible’, in A Distinct Voice:
Medieval Studies in Honor of Leonard E. Boyle, O.P., ed. J. Brown and
W. P. Stoneman (Notre Dame, IN), 32–67.
Sears, E. 2006, 'The afterlife of scribes: Swicher's prayer in the Prüfening Isidore',
in Pen in Hand: Medieval Scribal Portraits, Colophons and Tools, ed.
M. Gullick (Walkern), 75-96.
Stammberger, R. 2003, Scriptor und Scriptorium (Graz).
Stirnemann P. 1984, ‘Quelques bibliothèques princières et la production hors
scriptorium au XIIe siècle’, Bulletin archéologique du Comité des travaux
historiques et scientifiques 17–18, 7–38.
1989, ‘Les bibliothèques princiéres et privées aux XIIe et XIIIe siècles’, in HBF,
173–91.
Thomson, R. M. 1978, ‘The “scriptorium” of William of Malmesbury’, in Ker,
Essays, 117–42; revised version as ch. 4 in Thomson, William of Malmesbury.
1989, Catalogue of the Manuscripts of Lincoln Cathedral Chapter Library
(Cambridge).
 Bibliography 355
1997, ‘Books and learning at Gloucester Abbey’, in Books and Collectors,
1200–1700: Essays Presented to Andrew Watson, ed. J. Carley and C. Tite
(London), 3–26.
2002, ‘Minor manuscript decoration from the west of England in the twelfth
century’, in Reading Texts and Images: Essays on Medieval and Renaissance Art
and Patronage in Honour of Margaret M. Manion, ed. B. Muir (Exeter),
19–34.
2003, William of Malmesbury, 2nd edn. (Woodbridge).
2006, Books and Learning in Twelfth-Century England: The Ending of ‘Alter
Orbis’ (Walkern).
2007, ‘The place of Germany in the Twelfth-Century Renaissance’, in Beach,
Manuscripts and Monastic Culture, 19–42.
2012, ‘The place of Germany in the Twelfth-Century Renaissance: Books,
scriptoria and libraries’, in Kwakkel, Turning Over a New Leaf, 127–40.
2015, ‘William of Malmesbury and the Latin classics: New research’, in
Kwakkel, Latin Classics, 169–85.
Venarde, B. 1997, Women’s Monasticism and Medieval Society: Nunneries in France
and England, 890–1215 (Ithaca, NY).
Walter-von dem Knesebeck, H. 1995, ‘Lamspringe, ein unbekanntes Scriptorium
des Hamersleben-Halberstädter Reformkreises zur Zeit Heinriches des
Löwen’, in Heinrich der Löwe und sein Zeit, ed. J. Luckhardt and
F. Niehoff (3 vols., Munich), 2. 477–88.
Watson, A. G. 1984, Catalogue of Dated and Datable Manuscripts, c. 435–1600 in
Oxford Libraries (2 vols., Oxford).
Webber, T. 1992, Scribes and Scholars at Salisbury Cathedral, c. 1075–c. 1125
(Oxford).
1995, ‘Script and manuscript production at Christ Church, Canterbury, after
the Norman Conquest’, in Canterbury and the Norman Conquest: Churches,
Saints and Scholars, 1066–1109, ed. R. Eales and R. Sharpe (London), 145–58.
1996, ‘The diffusion of Augustine’s Confessions in England during the eleventh
and twelfth centuries’, in The Cloister and the World: Essays in Medieval
History in Honour of Barbara Harvey, ed. J. Blair and B. Golding (Oxford),
29–45.
Załuska, Y. 1989, L’enluminure et le scriptorium de Cîteaux au XIIe siècle (Brecht).

5 Scholars and Their Books

Primary Sources
Anthologia latina, ed. A. Riese, 1. 1 (Leipzig, 1869; repr. 1894).
Aristotle, Analytica Priora, ed. L. Minio-Paluello (Bruges and Paris, 1962).
Baldricus Burgulianus, Carmina, ed. J.-Y. Tilliette (2 vols., Paris, 1998).
Bernard of Clairvaux, Opera, ed. J. Leclercq et al. (8 vols., Rome, 1957–77).
356 Bibliography
Bernard of Utrecht, Commentum in Theodolum, in Accessus ad Auctores, ed.
R. B. C. Huygens (Leiden, 1970), 55–69.
Cartulaire de Notre-Dame de Chartres, ed. E. de Lépinois and L. Merlet (3 vols.,
Chartres, 1862–5).
Carmina Leodiensia, ed. W. Bulst (Heidelberg, 1975).
La Chronique de Morigny, ed. L. Mirot (Paris, 1909).
Conrad of Hirsau, Dialogus super auctores, in Accessus ad Auctores, 71–31.
Epistolae duorum amantium, ed. E. Könsgen (Leiden, 1974).
Garlandus Compotista, Dialectica, ed. L. M. De Rijk (Assen, 1959).
Garland, Candela, Preface, in Thesaurus novus anecdororum, ed. E. Martène and
U. Durand (6 vols., Paris, 1717), 1. 372–3.
Ivo of Chartres, Decretum, PL 161. 47B–1022C.
John of Salisbury, Historia Pontificalis, ed. M. Chibnall (Oxford, 1986).
Metalogicon, ed. J. B. Hall, (CCCM 98: 1991).
Obituaires de la province de Sens, ed. A. Molinier (2 vols., Paris, 1906).
Otto of Freising and Rahewin, Gesta Friderici I, ed. B. von Simson (MGH srg 46,
1912).
Papias, Ars grammatica, ed. R. Cervani (Bologna, 1998).
Peter Abelard, Opera Theologica 2, ed. E. M. Buytaert, CCCM 12 (1969).
Opera Theologica 3, ed. E. M. Buytaert and C. J. Mews CCCM 13 (1987).
Peter Lombard, Sententiae in IV Libris Distinctiae, ed. I. Brady (3 vols.,
Grottaferrata, 1971–81).
Robert of Melun, Sententiae, ed. R. J. Martin, Oeuvres de Robert de Melun, 3. 1–2
(Spicilegium sacrum Lovaniense 13 and 18: Louvain, 1947 and 1952).
William of Malmesbury, Gesta pontificum Anglorum, ed. and trans.
M. Winterbottom (Oxford, 2007).
Ysagoge in theologiam, ed. A. Landgraf, Ecrits théologiques de l’école d’Abélard
(Spicilegium sacrum Lovaniense 14: Louvain, 1934), 63–289.

Secondary Sources
Agus, I. A. 1968, Urban Civilization in Pre-Crusade Europe (2 vols., Leiden).
Álvarez López, F. J. 2012, ‘Monastic learning in twelfth-century England:
Marginalia, provenance and use in London, British Library, Cotton MS.
Faustina A. X, Part B’, Electronic British Library Journal, Article 11, 1–8
www.bl.uk/eblj/2012articles/pdf/ebljarticle112012.pdf.
Beach, A. I. 2004, Women as Scribes: Book Production and Monastic Reform in
Twelfth-Century Bavaria (Cambridge).
Bertolacci, A. 2011, ‘A community of translators: The Latin medieval versions
of Avicenna’s Book of the Cure’, in Mews and Crossley, Communities of
Learning, 37–54.
Burman, T. E. 2007, Reading the Qurʾān in Latin Christendom, 1140–1560
(Philadelphia, PA).
Burnett, C. 1997, The Introduction of Arabic Learning into England (London).
 Bibliography 357
2001, ‘The coherence of the Arabic-Latin translation programme in Toledo in
the twelfth century’, Science in Context 14, 249–88.
Burnett, C. and D. E. Luscombe 2001, ‘Communities of learning in twelfth-
century Toledo’, in Mews and Crossley, Communities of Learning, 9–18.
2005, ‘A new student for Peter Abelard: The marginalia in British Library MS
Cotton Faustina A. X’, in Itinéraires de la raison: Études de philosophie
médiévale offertes à Maria Cândida Pacheco, ed. J. F. Meirinhos (Louvain-
la-Neuve), 163–86.
Delisle, L. 1886, ‘Notice sur un manuscrit de l’abbaye de Luxeuil copié en 625’, in
Notices et extraits des manuscrits de la Bibliothèque nationale 31. 2 (Paris),
149–64.
Dronke, P. 1979, ‘A note on Pamphilus’, JWCI 42, 225–30.
Gasper, G. E. M. and F. Wallis 2004, ‘Anselm and the Articella’, Traditio 59,
129–74.
Genevois, A.-M. J.-F. Genest and A. Chalandon 1987, Bibliothèques de manuscrits
médiévaux en France: Relevé des inventaires du VIIIe au XVIIIe siècle (Paris).
Giacone, R. 1974, ‘Masters, books and library at Chartres according to the
cartularies of Notre-Dame and Saint-Père’, Vivarium 12/1, 30–51.
Gibson, M. T. 1979, ‘The early scholastic “Glosule” to Priscian, “Institutiones
grammaticae”: The text and its influence’, Studi Medievali 3a, ser. 20, 235–54;
repr. M. T. Gibson, ‘Artes’ and Bible in the Medieval West (Aldershot, 1993).
Giraud, C. 2010, Per verba magistri: Anselme de Laon et son école au XIIe siècle
(Turnhout).
James, M. R. 1934, ‘The Salomites’, Journal of Theological Studies, 35, 287–97.
Jeauneau, ‘Prologue’.
Lejbowicz, M. 2003, ‘Le premier témoin scolaire des Élements arabo-latins
d’Euclide: Thierry de Chartres et l’Heptateuchon’, Revue d’histoire des
sciences, 56, 347–68.
Lieftinck, G. 1955, ‘The Psalterium Hebraicum from St Augustine’s Canterbury
rediscovered in the Scaliger Bequest at Leyden’, Transactions of the Cambridge
Bibliographical Society 2/2, 97–104.
Luscombe, D. E. 1968, ‘The authorship of the Ysagoge in theologiam’, AHDLMA
35, 7–16.
Luscombe, D. E. and G. R. Evans 1996, Anselm: Aosta, Bec and Canterbury.
Proceedings in Commemoration of the Nine-Hundredth Anniversary of
Anselm’s Enthronement as Archbishop, 25 September 1093 (Sheffield).
Malacek, W. 1981, ‘Das Kardinalskollegium unter Innocenz II und Anaklet II’,
Annuarium historiae Pontificiae 19, 27–78.
Mews, C. J. 1996, ‘St Anselm, Roscelin and the See of Beauvais’, in Anselm: Aosta,
Bec and Canterbury, ed. D. Luscombe and G. Evans 106–19; repr. in C. J.
Mews, Reason and Belief in the Age of Roscelin and Abelard (Aldershot, 2002).
2001, ‘Hugh Metel, Heloise and Peter Abelard: The letters of an Augustinian
canon and the challenge of innovation in twelfth-century Lorraine’, Viator
32, 59–91.
358 Bibliography
Mews, C. J. and John N. Crossley (eds.) 2001, Communities of Learning: Networks
and the Shaping of Intellectual Identity in Europe, 1100–1500 (Turnhout).
2007, ‘Scholastic theology in a monastic milieu in the twelfth century: The case
of Admont’, in Beach, Manuscripts and Monastic Culture, 217–39.
2014, ‘Abelard, Heloise, and discussion of love in the twelfth-century schools’,
in Rethinking Peter Abelard: A Collection of Critical Essays, ed. B. S. Hellemans
(Leiden), 11–36.
Mews, C. J. and C. Giraud 2014, ‘John of Salisbury and the schools of the twelfth
century’, in The Brill Companion to John of Salisbury, ed. C. Grellard
(Leiden), 31–62.
2016, ‘Three classicizing poems in a manuscript of Pistoia (C. 101) from the early
twelfth century’, in La rigueur et la passion: Mélanges en l’honneur de Pascale
Bourgain, ed. C. Giraud and D. Poirel (Turnhout), 217–31.
Moore, R. 1998, Jews and Christians in the Life and Thought of Hugh of St. Victor
(Atlanta, GA).
Munk Olsen, B. 1989, ‘Les bibliothèques bénédictines et les bibliothèques de
cathédrales: Les mutations des XIe et XIIe siècles’, in HBF, 31–43.
Murano, G., G. Savino and S. Zamponi 1998, I manoscritti medievali della
provincia di Pistoia (Manoscritti medievali della Toscana, 1: Florence).
Nebbiai-Dalla Guarda, D. 1992, I documenti per la storia della biblioteche medievali:
(secoli IX–XV) (Rome).
Nebbiai-Dalla Guarda, D. and J.-F. Genest 1998, Du copiste au collectionneur:
Mélanges d’histoire des textes et des bibliothèques en l’honneur d‘André Vernet
(Turnhout).
Nortier, G. 1971, Les Bibliothèques médiévales des abbayes bénédictines de Normandie
(Paris).
Poirel, D. 2011, ‘Tene fontem et totum habes: L’unité du Didascalicon
d’Hugues de Saint-Victor’, in Universitas scolarium. Mélanges offerts à
Jacques Verger par ses anciens étudiants, ed. C. Giraud and M. Morard
(Geneva), 293–328.
Riché, P. 1988, ‘La bibliothèque de Gerbert d’Aurillac’, in Mélanges de la
bibliothèque de la Sorbonne offerts à A. Tuillier (Paris), 94–103.
Robert, U. 1873, ‘De Gerlandi vita et operibus disserit’, Analecta juris pontificii, ser.
12 (Rome), cols. 596–614.
Savino, G. 1987, ‘La libreria della Cattedrale di San Zenone nel suo più antico
inventario’, Bullettino storico Pistoiese 89, 25–39, available at www.archivio
capitolaredipistoia.it/bibliografia.php.
Smith, L. 2013, ‘Robert Amiclas and the Glossed Bible’, in From Knowledge to
Beatitude: St. Victor, Twelfth-Century Scholars, and Beyond. Essays in Honor
of Grover A. Zinn, Jr., ed. E. A. Matter and L. Smith (Notre Dame, IN),
131–52.
Stirnemann, P. D. 1994, ‘Où ont été fabriqués les livres de la glose ordinaire dans la
première moitié du XIIe siècle?’, in Le XIIe siècle: Mutations et renouveau et
France dans la première moitié du XIIe siècle, ed. F. Gasparri (Paris), 257–301.
 Bibliography 359
1998, ‘Histoire tripartite: Un inventaire des livres de Pierre Lombard, un
exemplaire de ses Sentences et le destinataire du Psautier de Copenhague’,
in Nebbiai-Dalla Guarda and Genest, Du copiste au collectionneur, 301–17.
Stohlmann, J. 1973, ‘Deidamia Achilli: Eine Ovid-imitation aus der 11.
Jahrhundert’, in Literatur und Sprache im Europaischen Mittelalter:
Festschrift für Kurt Langosch zum 70 Geburtstag (Darmstadt), 195–231.
Thomson, R. M. 1995, ‘Robert Amiclas: A twelfth-century Parisian master and his
books’, Scriptorium 49, 238–43; repr. in Thomson, England and the Twelfth-
Century Renaissance (Aldershot, 1998), art. III.
Tilliette, J.-Y. 1998, ‘La place d’Ovide dans la bibliothèque idéale de Conrad
d’Hirsau’, in Nebbiai-Dalla Guarda and Genest, Du copiste au collectionneur,
137–52.
Turcan-Verkerk, A.-M. 2007, ‘Ouvrages de dames? A propos d’un catalogue du
XIe siècle jadis attribué à Notre-Dame de Paris’, Scriptorium 61/2, 286–353.
Webb, C. C. J. 1941, ‘Note on books bequeathed by John of Salisbury to the
cathedral library of Chartres’, Mediaeval and Renaissance Studies 1, 128–29.
Wilmart, A. 1923, ‘Les livres légués par Célestin II à Città-di-Castello’, RB 35,
98–102.

6 The Libraries of Religious Houses

Primary Sources
Constitutions canonicorum regularium ordinis Arroasiensis, ed. L. Milis and
J. Becquet CCCM 20 (1970).
Consuetudines Floriacenses Antiquiores, ed. A. Davril and L. Donnat (Corpus
Consuetudinum Monasticarum 7/3: Siegburg, 1984).
Le coutumier de l’abbaye d’Oigny en Bourgogne au XIIe siècle, ed. P. F. Lefèvre and
A. H. Thomas (Louvain, 1976).
The Customary of the Benedictine Abbey of Eynsham in Oxfordshire, ed. A. Gransden
(Corpus Consuetudinum Monasticarum 2: Siegburg, 1963).
‘De obedientariis abbatiae Abbendonensis’, in Chronicon monasterii de Abingdon,
ed. J. Stevenson (2 vols., RS, 1858), 2. 335–417.
Gervase of Canterbury, The Historical Works, ed. W. Stubbs (2 vols., RS, 1879–80).
Lanfranc, Monastic Constitutions, ed. and trans. D. Knowles, 2nd edn., rev. C. N.
L. Brooke (Oxford, 2002).
Liber Ordinis Sancti Victoris, ed. L. Jocqué and L. Milis, CCCM 61 (1984).
Matthew Paris, Gesta Abbatum Sancti Albani, ed. H. T. Riley in Thomas
Walsingham, Gesta Abbatum Monasterii Sancti Albani, 1 (RS, 1867).
Thomas of Marlborough, History of the Abbey of Evesham, ed. and trans. J. Sayers
and L. Watkiss (Oxford, 2003).
Ulrich of Zell, Consuetudines Antiquiores Cluniacenses, PL 149. 643–779.
William of Hirsau, Constitutiones Hirsaugienses, ed. P. Engelbert (2 vols., Corpus
Consuetudinum Monasticarum 15: Siegburg, 2010).
360 Bibliography
Secondary Sources
Aubert, M. 1947, L’architecture cistercienne en France, 2nd edn., with Geneviève
Aliette de Rohan-Chabot Maillé (2 vols., Paris).
Becquet, J. 1989, ‘Les bibliothèques de chanoines réguliers (Prémontrés, Victorins,
etc.)’, in HBF, 83–91.
Bondéelle, A. 1989, ‘Trésor des moines: Les Chartreux, les Cisterciens et leurs
livres’, in HBF, 65–81.
Brunius, J. (ed.) 2005, Medieval Book Fragments in Sweden: An International
Seminar in Stockholm, 13–16 November 2003 (Stockholm).
Casson, L. 2001, Libraries in the Ancient World (New Haven, CT).
Choisselet, D. and P. Vernet 1989, Les ‘Ecclesiastica officia’ cisterciens du XIIème siècle
(Reiningue).
Clark, J. W. 1901, The Care of Books (Cambridge; repr. London, 1975).
Coates, A. 1999, English Medieval Books: The Reading Abbey Collections from
Foundation to Dispersal (Oxford).
Constable, G. 1996, The Reformation of the Twelfth Century (Cambridge).
Coppack, G. 2009, Fountains Abbey: The Cistercians in Northern England (Stroud).
Corpus Catalogorum Belgii, ed. A. Derolez et al. (7 vols., Brussels, 1996–2009).
De Jong, M. 2000, ‘Internal cloisters: The case of Ekkehard’s Casus sancti Galli’, in
Grenze und Differenz im frühen Mittelalter, ed. W. Pohl and H. Reimitz
(Vienna), 209–29.
De Mérindol, C. 1976, La production des livres peints à l'abbaye de Corbie au XIIe
siécle (3 vols., Lille).
Delisle, L. 1868–81, Le cabinet des manuscrits de la Bibliothèque nationale (4 vols.,
Paris; repr. New York, 1973).
Falmagne, T. 2000, ‘Le réseau des bibliothèques cisterciennes aux XIIe et XIIIe
siècles: Perspectives des recherche’, in Unanimité et diversité cisterciennes:
Filiations, réseaux, relectures du XIIe au XVIIe siècle, Actes du Quatrième
colloque international du CERCOR, Dijon 23–25 septembre 1998 (Saint-
Étienne), 195–222.
Fassler, M. E. 1985, ‘The office of the cantor in early western monastic rules and
customaries: A preliminary investigation’, Early Music History 6, 29–51.
Fergusson, P. and S. Harrison 1999, Rievaulx Abbey: Community, Architecture,
Memory (New Haven, CT).
Fiesoli, G., et al. 2009–, Repertorio di inventari e cataloghi di biblioteche medievali,
1– (Florence).
Gameson, R. 2006, ‘The medieval library (to c. 1450)’, in CHL, 13–50.
Genest, J.-F. 1989, ‘Le mobilier des bibliothèques d’après les inventaires
médiévaux’, in Vocabulaire du livre et de l’écriture au Moyen Âge: Actes de la
table ronde, Paris 24–26 septembre 1987, ed. O. Weijers (Turnhout), 136–54.
Golding, B. 1995, Gilbert of Sempringham and the Gilbertine Order, c. 1130–c. 1300
(Oxford).
 Bibliography 361
González, A. S. 2015, ‘Cistercian scriptoria in the twelfth and thirteenth centuries:
A starting-point’, in Culture and Society in Medieval Galicia: A Cultural
Crossroads at the Edge of Europe, ed. and trans. J. d’Emilio (Leiden), 765–811.
Gullick, M. 1998, ‘Professional Scribes in Eleventh and Twelfth-Century
England’, English Manuscript Studies 1100–1700 7, 1–24.
Hirst, S. M., D. A. Walsh and S. M. Wright 1983, Bordesley Abbey II: Second Report
on Excavations at Bordesley Abbey, Redditch, Hereford–Worcestershire (British
Archaeological Reports, British Series, 111, Oxford).
Horn, W. 1973, ‘On the origins of the medieval cloister’, Gesta 12, 13–52.
Karlsen, E. (ed.) 2013, Latin Manuscripts of Medieval Norway: Studies in Memory of
Lilli Gjerløw (Oslo).
Ker, N. R. 1964, Medieval Libraries of Great Britain, 2nd edn. (London: RHS), and
Supplement (with A. G. Watson) (London: RHS, 1987).
Kinder, T. N. 2002, Cistercian Europe: Architecture of Contemplation (Grand
Rapids and Kalamazoo, MI).
Klein, P. K. 2004, Der mittelalterliche Kreuzgang: Architektur, Funktion und
Programm (Regensburg).
Kottje, R. 1982, ‘Claustra sine armario? Zum unterschied von Kloster und Stift im
Mittelalter’, in Consuetudines monasticae: Eine Festgabe für Kassius Hallinger
aus Anlass seines 70. Geburtstages, ed. J. F. Angere and J. Lenzenweger, Studia
Anselmiana 85, 125–44.
Krämer, S. 1989–90, Handschriftenerbe des deutschen Mittelalters (3 vols., Munich).
Lapidge, M. 2006, The Anglo-Saxon Library (Oxford).
Lefèvre, P. F. 1972, ‘Á propos de la “lectio divina” dans la vie monastique et
canoniale’, Revue d’histoire ecclésiastique 67, 800–9.
Lehmann, E. 1957, Die Bibliotheksräume der deutschen Klöster im Mittelalter
(Berlin).
McKitterick, R. 1989, The Carolingians and the Written Word (Cambridge).
Masson, A. 1972, Le décor des bibliothèques du Moyen Âge à la Révolution (Geneva).
Mews, C. J. 2002, ‘Manuscripts in Polish libraries copied before 1200 and the
expansion of Latin Christendom in the eleventh and twelfth centuries’,
Scriptorium 56, 80–118.
Meyvaert, P. 1973, ‘The medieval monastic claustrum’, Gesta 12, 53–9.
Mittelalterliche Bibliothekskataloge Österreichs (5 vols. and supplement, Vienna,
1915–71).
Munk Olsen, B. 1989, ‘Les bibliothèques bénédictines et les bibliothèques de
cathedrals: Les mutations des XIe and XIIe siècles’, in HBF, 31–43.
Nebbiai-Dalla Guarda, D. 1989, ‘Classifications et classements’, in HBF, 373–93.
1994, ‘Livres et bibliothèques dans les monastères français au XIIe siècle’, in Le
XIIe siècle: Mutations et renouveau en France dans la première moitié du XIIe
siècle, ed. F. Gasparri (Paris).
1996, ‘La bibliothèque commune des institutions religieuses’, Scriptorium 50,
254–68.
362 Bibliography
1997, ‘Normes médiévales régissant l’accès aux bibliothèques’, in Usages des
bibliothèques: Lieux d’histoire et état des lieux (Actes de la Table-Ronde organisée
par Histoire au Présent et l’Institut Historique Allemand de Paris) (Paris), 31–44.
2013, Le discours des livres: Bibliothèques et manuscrits en Europe IXe–XVe siècle
(Rennes).
Oakeshott, W. 1954, ‘Winchester College library before 1750’, The Library 5, ser. 9,
1–16.
Ó Néill, P. 2006, ‘Celtic Britain and Ireland in the early Middle Ages’, in CHL,
69–90.
Pfaff, R. W. 1981, ‘The “Abbreviatio Amalarii” of William of Malmesbury’, RTAM
48, 128–71.
Prache, A. 1989, ‘Bâtiments et décor’, in HBF, 351–63.
Pressouyre, L. 1973, ‘St Bernard to St Francis: Monastic ideals and iconographic
programs in the cloister’, Gesta 12, 71–92.
Ramsay, N. 1995, ‘The cathedral archives and library’, in A History of Canterbury
Cathedral, ed. P. Collinson, N. Ramsay and M. Sparks (Oxford), 341–407.
Robinson, D. M. 2006, The Cistercians in Wales: Architecture and Archaeology,
1130–1540 (London).
Silvestre, H. 1964, ‘À propos du dicton “Claustrum sine armario, quasi castrum
sine armamentario”’, Mediaeval Studies 26, 351–3.
Smith, L. 1992, ‘Lending books: The growth of a medieval question from Langton
to Bonaventure’, in Intellectual Life in the Middle Ages: Essays Presented to
Margaret Gibson, ed. L. Smith and B. Ward (London), 265–79.
1996, ‘Scriba, femina: Medieval depictions of women writing’, in Women and the
Book: Assessing the Visual Evidence, ed. L. Smith and J. Taylor (London and
Toronto, 21-44
Swartling, I. 1969, Alvastra Abbey: The First Cistercian Settlement in Sweden
(Stockholm).
Thomson, R. M. 2012, ‘The place of Germany in the twelfth-century renaissance:
Books, scriptoria and libraries’, in Turning Over a New Leaf, 127–44.
Van Waefelghem, R. 1913, ‘Les premiers statuts de l’ordre de Prémontré: Le Clm
17, 174 (XIIe siècle)’, Analectes de l’ordre de Prémontré 9, 1–74.
Vernet, A. 1989, ‘Du “chartophilax” au “librarian”’, in Vocabulaire du livre et de
l’écriture au Moyen Âge: Actes de la table ronde, Paris 24–26 Septembre 1987, ed.
O. Weijers (Turnhout), 155–67.
Vezin, J. 1989, ‘Le mobilier des bibliothèques’, in HBF, 365–71.
Webber, T. 2006, ‘Monastic and cathedral book collections in the late eleventh
and twelfth centuries’, in CHL, 109–25.
2013, ‘Monastic space and the use of books in the Anglo-Norman period’, ANS
36, 221–40.
2017, ‘Cantor, prior or sacrist: The provision of books in Anglo-Norman
England’, in Medieval Cantors and Their Craft: Music, Liturgy, and the
Shaping of History (800–1500), ed. M. Fassler, K. Bugyis and A. Kraebel
(Cambridge), 172–89.
 Bibliography 363
7 Modes of Reading

Primary Sources
Ambrose, Expositio evangelii secundum Lucam. Fragmenta in Esaiam (Ambrosius
Mediolanensis), ed. M. Adriaen, CCSL 14. 1–6 (1957).
Anselm, Orationes and Meditationes, in S. Anselmi cantuariensis archiepiscopi opera
omni, ed. F. S. Schmitt (rev. edn., 6 vols. in 2, Stuttgart, 1968), 2. 2–91.
The Prayers and Meditations of Saint Anselm with the Proslogion, trans. B. Ward
(Harmondsworth, 1973).
Augustine, Confessions, Vol. III, Commentary Books 8–13, ed. J. J. O’Donnell
(Oxford, 1992).
De Doctrina Christiana, ed. and trans. R. P. H. Green (Oxford, 1996).
St Benedict, Rule, ed. and trans. B. L. Venarde (Dumbarton Oaks Medieval
Library 6: Cambridge, MA, 2011).
Cassian, Conferences, ed. and trans. B. Ramsey (New York, 1997).
Guigo II, The Ladder of Monks, A Letter on the Contemplative Life and Twelve
Meditations, trans. E. Colledge and J. Walsh (Kalamazoo, MI, 1979).
The Holy Bible: Containing the Old and New Testaments with the Apocryphal/
Deuterocanonical Books: New Revised Standard Version, trans. B. M.
Metzger (Oxford, 1989).
Hugh of St-Victor, Didascalicon, trans. J. Taylor (New York, 1961).
Jerome, In Isaiam, Prologue to Book XVIII, PL 24. 17–22.
Translatio Regulae Sancti Pachomii, PL 23. 61–86.
Liber Comicus: sive Lectionarius Missae quo Toletana Ecclesia ante annos mille et
ducentos utebatur, ed. G. D. Morin (Maredsous, 1893).
The Ordinal of the Abbey of the Holy Trinity Fécamp, ed. D. Chadd (2 vols.,
London, 2000).
Origen, The Song of Songs: Commentary and Homilies, trans. and annotated by
R. P. Lawson (London, 1957).
Peter Abelard, Sic et Non, ed. B. Boyer and R. McKeon (Chicago, IL, 1977).
William of St-Thierry, The Golden Epistle: A Letter to the Brethren at Mont Dieu,
trans. T. Berkeley, intro. J. M. Déchanet (Kalamazoo, MI, 1971).

Secondary Sources
Bhattacharji, S. (ed.) 2014, Prayer and Thought in Monastic Tradition: Essays in
Honour of Benedicta Ward (New York).
Boynton, S. 2011, ‘The Bible and the liturgy’, in The Practice of the Bible in the
Middle Ages: Production, Reception, and Performance in Western Christianity,
ed. S. Boynton and D. J. Reilly (New York), 10–33.
Bynum, C. W. 1982, Jesus as Mother: Studies in the Spirituality of the High Middle
Ages (Berkeley, CA).
Casey, M. 1995, Sacred Reading: The Ancient Art of Lectio Divina (Liguori, MS).
364 Bibliography
Collamore, L. 2000, ‘Prelude: Charting the Divine Office’, in The Divine Office in
the Latin Middle Ages: Methodology and Source Studies, Regional Developments,
Hagiography, ed. M. E. Fassler and R. A. Baltzer (Oxford), 3–11.
Crichton, J. D. 1992, ‘The Office in the West: The early Middle Ages’, in The
Study of Liturgy, rev. edn, ed. C. Jones et al. (Oxford), 420–9.
Davies, B. and B. Leftow (eds.) 2004, The Cambridge Companion to Anselm
(Cambridge).
Dronke, P. 1970, Poetic Individuality in the Middle Ages (Oxford).
Edsall, M. A. 2000, Reading Like a Monk: Lectio Divina, Religious Literature, and
Lay Devotion, PhD Dissertation (Columbia University).
Elders, L. J. 2003, ‘Scholastiche Methode’, in Lexikon des Mittelalters VII
(Munich), 1526–8.
Evans, G. R. 1974, ‘Mens Devota: The literary community of the devotional works
of John of Fécamp and St. Anselm’, Medium Aevum 43, 105–15.
Exalto, J. 2012, ‘Orating from the pulpit: The Dutch Augustine and the reformed
godly until 1700’, in Augustine beyond the Book: Intermediality, Transmediality
and Reception, ed. K. Pollman and M. J. Gill (Leiden), 195–216.
Ferruolo, S. 1985, The Origins of the University: The Schools of Paris and Their
Critics, 1100–1215 (Stanford, CA).
Grabmann, M. 1909, Geschichte der scholastischen Methode (2 vols., Freiburg).
Grémont, D. B. 1971, ‘Lectiones ad prandium à l’abbaye de Fécamp au XIIIe
siècle’, Cahiers Léopold Delisle 20/3–4, 3–41.
Grodecki, L. 1975, ‘Abélard et Suger’, in Pierre Abélard et Pierre le Vénérable: Les
courants philosophiques, littéraires et artistiques en occident au milieu du XIIe
siècle – Abbaye de Cluny, 2 au 9 juillet 1972, ed. R. Louis, J. Jolivet and
J. Châtillon (Actes et mémoires des colloques internationaux du CNRS 546:
Paris), 279–86.
Gumbert, J. P. 2009, ‘Points and signposts: Whom do they help?’, Scriptorium 63,
231–7.
Hamesse, J. 1999, ‘The scholastic model of reading’, in Cavallo and Chartier,
Reading, 103–19.
Hankins, J. 1990, Plato in the Italian Renaissance (Leiden and New York).
Harper, J. 1991, The Forms and Orders of Western Liturgy from the Tenth to the
Eighteenth Century: A Historical Introduction and Guide for Students and
Musicians (Oxford).
Heintz, M. 2003, ‘Prologue of Ambrose of Milan’s Homilies on Luke’, Antiphon
8/2, 26–31.
Holzherr, G. 1994, The Rule of Benedict: A Guide to Christian Living, trans.
P. Murray (Dublin), 232–4.
Illich, I. 1993, ‘Lectio Divina’, in Schriftlichkeit im frühen Mittelalter, ed.
U. Schaefer (Scripta Oralia 53: Tübingen), 19–35.
Irvine, R. 2010, ‘How to read: Lectio divina in an English Benedictine monastery’,
Culture and Religion 11, 395–411.
Kardong, T. 1996, Benedict’s Rule: A Translation and Commentary
(Collegeville, MN).
 Bibliography 365
Knowles, D. 1962, The Evolution of Medieval Thought (London).
Köpf, U. 2000, ‘The institutional framework of Christian exegesis in the Middle
Ages’, in Hebrew Bible/Old Testament: The History of Its Interpretation, 2
(Göttingen), 148–79.
Lampe, G. W. H., et al. 1969, ‘The exposition and exegesis of Scripture’, in The
Cambridge History of the Bible II, The West from the Fathers to the Reformation,
ed. G. W. H. Lampe (Cambridge), 155–279.
Lawless, G. (ed.) 1987, Augustine of Hippo and His Monastic Rule (Oxford).
Lawn, B. 1993, The Rise and Decline of the Scholastic ‘Quaestio Disputa’ (Leiden).
Leclercq, H. 1922, ‘Épitres’ and ‘Évangiles’, in F. Cabrol and H. Leclercq,
Dictionnaire d’archéologie chrétienne et de liturgie (Paris), 5, cols. 244–344,
852–923.
Leclercq J. 1947, L’amour des lettres et le désir de Dieu (Paris); trans. C. Misrahi as
The Love of Learning and the Desire for God: A Study of Monastic Culture
(Fordham, NY, 1961).
1986, ‘Monastic and scholastic theology in the reformers of the fourteenth to
sixteenth century’, in From Cloister to Classroom: Monastic and Scholastic
Approaches to Truth, ed. E. R. Elder (Kalamazoo, MI), 178–201.
2012, ‘Monastic commentary on biblical and ecclesiastical literature from late
antiquity to the twelfth century’, The Medieval Journal 1/2, 27–53.
Leclercq J. and J.-P. Bonnes 1946, Un Maître de la vie spirituelle au XIe siècle, Jean
de Fécamp (Paris).
Leclercq J., F. Vandenbroucke and L. Bouyer 1968, The Spirituality of the Middle
Ages (London).
Lefler, N. 2014, Theologizing Friendship: How Amicitia in the Thought of Aelred and
Aquinas Inscribes the Scholastic Turn (Eugene, OR).
Marenbon, J. 2004, ‘Life, milieu, and intellectual contexts’, in The Cambridge
Companion to Abelard, ed. J. E. Brower and K. Guilfoy (Cambridge), 13–44.
McGinn, B. 1994, The Presence of God: A History of Western Christian Mysticism, 2:
The Growth of Mysticism (New York).
McLaughlin, M. 2010, Sex, Gender, and Episcopal Authority in the Age of Reform,
1000–1122 (Cambridge).
McNamer, S. 2010, Affective Meditation and the Invention of Medieval Compassion
(Philadelphia, PA).
Mews, C. J. 2000, ‘Monastic educational culture revisited: The witness of
Zwiefalten and the Hirsau reform’, in Medieval Monastic Education, ed.
G. Ferzoco and C. Muessig (London and New York), 182–97.
2007, ‘Scholastic theology in a monastic milieu in the twelfth century: The case
of Admont’, in Beach, Manuscripts and Monastic Culture, 217–39.
Newman, M. G. 1996, The Boundaries of Charity: Cistercian Culture and
Ecclesiastical Reform, 1098–1180 (Stanford, CA).
Palazzo, E. 1998, A History of Liturgical Books from the Beginning to the Thirteenth
Century, trans. M. Beaumont (Collegeville, MN).
Parkes, M. B. 1976, ‘The influence of the concepts of ordinatio and compilatio on
the development of the book’, in Hunt Essays, 115–41.
366 Bibliography
1999, ‘Reading, copying and interpreting a text in the early Middle Ages’, in
Cavallo and Chartier, Reading, 90–102.
2008, ‘Layout and presentation of the text’, in CHBB II, 55–74.
Pennington, B. 1998, Lectio Divina: Renewing the Ancient Practice of Praying the
Scriptures (New York).
Rahner, K. (ed.) 1975, Encyclopedia of Theology: A Concise Sacramentum Mundi
(New York).
Rees, D. 1978, Consider Your Call: A Theology of Monastic Life Today (London).
Robertson, D. 1996, Lectio Divina: The Medieval Experience of Reading
(Collegeville, MN).
Rouse, M. A. and R. H. Rouse 1979, Preachers, Florilegia and Sermons: Studies on
the ‘Manipulus Florum’ of Thomas of Ireland (Toronto).
1982, ‘Statim invenire: Schools, preachers, and new attitudes to the page’, in
Benson and Constable, Renaissance, 201–25; repr. in their Authentic Witnesses,
191–219.
Salmon, P. 1965, ‘Monastic asceticism and the origins of Citeaux’, trans. Monk of
Gethsemani Abbey, Monastic Studies 3, 119–38.
Sandor, M. 1989, ‘Lectio Divina and the monastic spirituality of reading’, The
American Benedictine Review 40, 82–114.
Seel, N. M. (ed.) 2012, Encyclopedia of the Sciences of Learning (New York).
Sheerin, D. 1996, ‘The liturgy’, in F. A. C. Mantello and A. G. Rigg, Medieval
Latin: An Introduction and Bibliographical Guide (Washington, DC), 157–82.
Silvas, A. M. (trans.) 2013, The Rule of St Basil in Latin and English: A Revised
Critical Edition (Collegeville, MN).
Sodeika, T. 2005, ‘The mysticism of Meister Eckhart and the phenomenology of
Edmund Husserl’, in Contemporary Philosophical Discourse in Lithuania, ed.
J. Baranova (Washington, DC), 21–40.
Southern, R. W. 1953, The Making of the Middle Ages (New Haven, CT).
1979, Platonism, Scholastic Method, and the School of Chartres (Reading).
1982, ‘The Schools of Paris and the School of Chartres’, in Benson and
Constable, Renaissance, 113–37.
1990, Saint Anselm: A Portrait in a Landscape (Cambridge).
1995, Scholastic Humanism and the Unification of Europe (2 vols., Oxford), 1.
Stock, B. 2001, After Augustine: The Meditative Reader and the Text
(Philadelphia, PA).
Studzinski, R. 2009, Reading to Live: The Evolving Practice of Lectio Divina
(Trappist, KY).
Taylor, A. 2002, Textual Situations: Three Medieval Manuscripts and Their Readers
(Philadelphia, PA).
Ullmann, W. 1966, The Individual and Society in the Middle Ages
(Baltimore, MD).
Vandenbroucke, F. 1966, ‘La lectio divina du XIe au XIVe siècle’, Studia Monastica
8, 267–93.
Verger, J. 1999, ‘The universities and scholasticism’, in The New Cambridge
Medieval History 5, c. 1198–c. 1300, ed. D. Abulafia (Cambridge), 256–76.
 Bibliography 367
Vogel, C. 1986, Medieval Liturgy: An Introduction to the Sources, rev. and trans.
W. G. Storey and N. K. Rasmussen (Washington, DC).
Webber, T. 2013, ‘Reading in the refectory: Monastic practice in England,
c. 1000–c. 1300’, London University Annual John Coffin Memorial
Palaeography Lecture 2010, revised 2013, 1–49, available at www.ies.sas.ac
.uk/publications/trust-fund-lectures/john-coffin-memorial-lectures-and-li
terary-readings.
2014, ‘Monastic space and the use of books in the Anglo-Norman period’, ANS
36, 221–40.
2015, ‘Bede’s Historica Ecclesiastica as a source of lections in pre- and post-
Conquest England’, in The Long Twelfth-Century View of the Anglo-Saxon
Past, ed. M. Brett and D. A. Woodman (Farnham), 47–74.
Weston, J. 2015, The Spirit of the Page: Books and Readers at the Abbey of
Fécamp, c. 1000–1200. Unpublished PhD Dissertation (Leiden University).

8 Practices of Appropriation
Blair, A. M. 2010, Too Much to Know. Managing Scholarly Information before the
Modern Age (New Haven, CT, and London).
Caillet, J.-P. 2009, ‘Caractères et statut du livre d’apparat carolingien’, in Les
manuscrits carolingiens: Actes du colloquies de Paris, Bibliothèque nationale de
France, le 4 mai 2007, ed. J.-P. Callet and M.-P. Lafitte (Bibliologia 27:
Turnhout), 1–43.
Contreni, J. J. 2014, ‘Learning for God: Education in the Carolingian age’, JML 24
(2014), 89–129.
De Meyier, K. A. 1973, Codices Vossiani Latini, Pars I: Codices in Folio (Leiden),
84–6.
Eastwood, B. S. 2011, ‘The power of diagrams: The place of the anonymous
commentary in the development of Carolingian astronomy and cosmology’,
in Carolingian Scholarship and Martianus Capella: Ninth-Century
Commentaries on De nuptiis in Context, ed. M. Teeuwen and S. O’Sullivan
(Turnhout), 193–220.
Ganz, D. 1995, ‘Book production in the Carolingian empire and the spread of
Caroline minuscule’, in The New Cambridge Medieval History 2, c. 700–c. 900,
ed. R. McKitterick (Cambridge), 786–808.
Glaze, F. E. 2008, ‘Gariopontus and the Salernitans: Textual traditions in the
eleventh and twelfth centuries’, in La Collectio Salernitana di Salvatore De
Renzi, ed. D. Jaquart and A. P. Bagliani (Florence), 149–90.
Grafton, A. 1999, ‘The humanist as reader’, in Cavallo and Chartier, Reading, 179–212.
Gumbert, P. 2009, Illustrated Inventory of Medieval Manuscripts, 2: Leiden,
Universiteitsbibliotheek BPL (Hilversum).
Hamesse, J. 1999, ‘The scholastic model of reading’, in Cavallo and Chartier,
Reading, 103–19.
Hofman R. 1996, The Sankt Gall Priscian Commentary: Part 1, (2 vols., Münster).
368 Bibliography
Holtz, L. 2009, ‘L’émergence de l’oeuvre grammaticale de Priscien et la chron-
ologie de sa diffusion’, in M. Baratin, B. Colombat and L. Holtz, Priscien.
Transmission et refondation de la grammaire: De l’Antiquité aux Modernes
(Turnhout), 37–55.
Jardine, L. and A. Grafton 1990, ‘“Studied for action”: How Gabriel Harvey read
his Livy’, Past and Present 129, 30–78.
Kwakkel, E. 2012, ‘“Dit boek heeft niet de vereiste breedte”: Afwijkende bladdi-
mensies in de elfde en twaalfde eeuw’, Jaarboek voor Nederlandse
Boekgeschiedenis 19, 33–49.
2012a, ‘Biting, kissing and the treatment of feet: The transitional script of the
Long Twelfth Century’, in Turning Over a New Leaf, 78–126, 206–8.
Kwakkel, E. and F. Newton, Medicine at Monte Cassino: Constantine the
African and the Oldest Manuscript of His Pantegni (Turnhout: Brepols, in
press).
Lowe, E. A. 1925, ‘Some facts about our eldest Latin manuscripts’, Classical
Quarterly 19, 197–208; repr. in his Palaeographical Papers 1907–1965, ed.
L. Bieler (2 vols., Oxford, 1972), 1. 187–202.
1928, ‘More facts about our oldest Latin manuscripts’, Classical Quarterly 22,
43–62; repr. in his Palaeographical Papers, 1. 250–74.
McKitterick, R. 2012, ‘Glossaries and other innovations in Carolingian book
production’, in Turning Over a New Leaf, 21–76.
Munk Olsen, B. 1982–9, L’étude des auteurs classiques latins aux XIe et XIIe siècles
(3 vols., Paris).
1996, ‘The production of classics in the eleventh and twelfth centuries’, in
Chavannes-Mazel and Smith, Latin Classics, 1–17.
Reynolds, S. 1996, ‘Glossing Horace: Using the classics in the medieval classroom’,
in Chavannes-Mazel and Smith, Latin Classics, 103–17.
Rouse, M. A. and R. H. Rouse 1982, ‘Statim invenire: Schools, preachers, and new
attitudes to the page’, in Benson and Constable, Renaissance, 201–25; repr. in
their Authentic Witnesses, 191–219.
1991, ‘Backgrounds to print: Aspects of the manuscript book in northern
Europe of the fifteenth century’, in their Authentic Witnesses, 449–66.
1991a, ‘The development of research tools in the thirteenth century’, in their
Authentic Witnesses, 221–55.
Saenger, P. 1999, ‘Reading in the later Middle Ages’, in Cavallo and Chartier,
Reading, 120–48.
Steinová, E. 2016, ‘Notam superponere studui’: The use of technical signs in the
early Middle Ages. Unpublished dissertation, Utrecht University.
Teeuwen, M. 2011, ‘Marginal scholarship: Rethinking the function of Latin glosses
in early medieval manuscripts’, in Rethinking and Recontextualizing Glosses:
New Perspectives in the Study of Late Anglo-Saxon Glossography, ed.
P. Lendinara, L. Lazzari and C. Di Sciacca (Porto), 19–37.
2015, ‘Carolingian scholarship on classical authors: Practices of reading and
writing’, in Kwakkel, Latin Classics, 23–50.
 Bibliography 369
2016, ‘Three annotated letter-manuscripts: Scholarly practices of religious
Franks in the margin unveiled’, in Religious Franks: Religion and Power in
the Frankish Kingdoms: Studies in Honour of Mayke de Jong, ed. R. Meens et al.
(Manchester), 221–39.
Tura, A. 2005, ‘Essai sur les marginalia en tant que pratique et documents’, in
Scientia in margine: Études sur les marginalia dans les manuscrits scientifiques du
Moyen Âge à la Renaissance, ed. D. Jacquart and C. Burnett (Geneva), 261–387.
Van Renswoude, I. and E. Steinová 2017, ‘The annotated Gottschalk: Symbolic
annotation and control of heterodoxy in the Carolingian age’, in Collections
des Études Augustiniennes.
Zetzel, J. E. G. 2005, Marginal Scholarship & Textual Deviance. The ‘Commentum
Cornuti’ & the Early Scholia on Persius (Bulletin of the Institute of Classical
Studies Supplement 84: London).

9 Hebrew Books
Attia, E. 2014, ‘Targum layouts in Ashkenazi manuscripts. Preliminary methodo-
logical observations’, in A Jewish Targum in a Christian World, ed.
A. Houtman, E. van Staalduine-Sulman and H.-M. Kirn (Leiden), 99–122.
Beit-Arié, M. 1985, The Only Dated Medieval Hebrew Manuscript Written in
England (1189) (London).
Beit-Arié, M., C. Sirat and M. Glatzer 2006, Codices hebraicis litteris exarati
(Monumenta Palaeographica Medii Aevi: Turnhout).
De Lange, N. and J. Olszowy-Schlanger 2014, Manuscrits hébreux et arabes:
Mélanges en l’honneur de Colette Sirat (Bibliologia 38: Turnhout).
De Visscher, E. 2014, Reading the Rabbis: Christian Hebraism in the Works of
Herbert of Bosham (Leiden).
Engel, E. 2014, ‘Between France and Germany: Gothic characteristics in
Ashkenazi script’, in De Lange and Olszowy-Schlanger, Manuscrits hébreux
et arabes, 197–219.
Entin-Rokéah, Z. 1985, ‘A Jewish payment memorandum’, in Beit-Arié 1985, 33–56.
˙ Selected Manuscripts and Printed Editions from the Treasures of the
Glatzer, M. 1985,
Jewish National and University Library (in Hebrew) (Jerusalem).
Goldin, S. 1995, ‘The synagogue in medieval Jewish community as an integral
institution’, Journal of Ritual Studies 9/1, 15–39.
Goldschmidt, E. D. 1966, ‘Le texte de prière du manuscrit Reggio du Mahzor
Vitry’, Revue des études juives 125, 63–75. ˙
Grossman, A. 1975, ‘The migration of the Kalonymos family from Italy to
Germany: The origins of the Jewish settlement in Germany in the Middle
Ages’ (in Hebrew), Zion 40, 154–85.
2001, The Early Sages of France: Their Lives, Leadership and Works (in Hebrew)
(Jerusalem).
Haran, M. 1985, ‘Bible scrolls in eastern and western Jewish communities from
Qumran to the High Middle Ages’, Hebrew Union College Annual 56, 21–62.
370 Bibliography
Isserles, J. 2012, Mahzor Vitry: Étude d’un corpus de manuscrits hébreux ashkénazes
˙
de type liturgico-légal du XIIe au XIVe siècle, PhD Thesis, École pratique des
hautes études and University of Geneva (Paris).
2014, ‘Les parallèles esthétiques des manuscrits hébreux ashkenazes de type
liturgico-légal et des manuscrits latins et vernaculaires médiévaux’, in De
Lange and Olszowy-Schlanger, 77–113.
Kanarfogel, E. 1992, Jewish Education and Society in the High Middle Ages
(Detroit, MI).
Kogel, J. 2014, ‘Les fragments du Talmud de Babylone conservés à la Bibliothèque
municipal de Colmar’, in De Lange and Olszowy-Schlanger, 115–26.
Kwakkel, E. 2012, ‘Biting, kissing and the treatment of feet: The transitional script
of the Long Twelfth Century’, in Turning Over a New Leaf, 78–126, 206–8.
Landgraf, A. 1934, Ecrits théologiques de l’école d’Abélard (Spicilegium sacrum
Lovaniense 14).
Lévy, W. D. 2008, Le livre dans la société juive médiévale de la France du Nord
(Paris).
Lieftinck, G. 1955, ‘The Psalterium Hebraicum from St. Augustine’s Canterbury
rediscovered in the Scaliger bequest at Leiden’, Transactions of the Cambridge
Bibliographical Society 2, 97–107.
Margoliouth, G. 1899, Catalogue of the Hebrew and Samaritan Manuscripts in the
British Museum, I (London).
Montfaucon, B., de 1702, Diarium italicum sive Monumentorum veterum, bib-
liothecarum, musaeorum, etc. notitiae singulares in Itinerario Italico collectae
(Paris).
Mortara Ottolenghi, L. 1985, ‘La Bibbia di La Rochelle’, in Les Juifs au regard de
l’histoire. Mélanges en l’honneur de Bernard Blumenkranz, ed. G. Dahan
(Paris), 149–56.
Mundill, R. R. 1998, England’s Jewish Solution: Experiment and Expulsion,
1262–1290 (Cambridge).
Nahon, G. 1966, ‘Alfonse de Poitiers et les Juifs’, Revue des études juives 125,
167–211.
Olszowy-Schlanger, J. 2003, Les manuscrits hébreux dans l’Angleterre médiévale:
Étude historique et paléographique (Louvain).
2011, ‘The money language: Latin and Hebrew in Jewish legal contracts from
medieval England’, in Studies in the History of Culture and Science. A Tribute
to Gad Freudenthal, ed. R. Fontaine, R. Glasner, R. Leicht and G. Veltri
(Leiden), 233–50.
2012, ‘The Hebrew Bible’, in The New Cambridge History of the Bible II: From
600 to 1450, ed. R. Marsden and E. A. Matter (Cambridge).
Olszowy-Schlanger, J. and P. Stirnemann 2008, ‘The twelfth-century trilingual
Psalter in Leiden’, Scripta 1, 103–12.
Perani, M. and G. Corazzol 2013, ‘Nuovo catalogo dei manoscritti ebraici della
Biblioteca Universitaria di Bologna’, in In BUB: Ricerche e cataloghi sui fondo
della Biblioteca Universitaria di Bologna, ed. B. Antinino and P. Moscatelli
(Bologna), 13–191.
 Bibliography 371
Peretz, Y. 2008, ‘Twice the Bible and once the Targum’ (in Hebrew), Tallelei Orot
14, 53–62.
Richler, B., et al. 2008, Hebrew Manuscripts in the Vatican Library: Catalogue
Compiled by the Staff of the Institute of Microfilmed Hebrew Manuscripts,
Jewish National and University Library, Jerusalem (Vatican City).
Rosenthal, D. 1972, Babylonian Talmud, Codex Florence, Florence National Library
II 7–9. The Earliest Dated Talmud Manuscript (Jerusalem).
Sassoon, D. 1932, Ohel David: Descriptive Catalogue of the Hebrew and Samaritan
Manuscripts in the Sassoon Library (Oxford).
Sirat, C. 1999, ‘Note sur la circulation des livres entre juifs et chrétiens au Moyen
Âge’, in Du copiste au collectionneur: Mélanges d’histoire des textes et des
bibliothèques en l’honneur d’André Vernet, ed. D. Nebbiai-Dalla Guarda and
J.-F. Genest (Bibliologia 18: Turnhout), 383–403.
2000, ‘En vision globale: les juifs médiévaux et les livres latins’, in La tradition
vive: Mélanges d’histoire des textes en l’honneur de Louis Holtz, ed. P. Lardet
(Bibliologia 20: Turnhout), 15–20.
Sirat, C. et al. 1996, La conception du livre chez les piétistes ashkenazes (Geneva).
2006, La conception du livre chez les piétistes ashkenazes (Geneva).
Sirat, C. and M. Dukan 1976, Écriture et civilization (Paris).
Smithuis, R. 2006, ‘Abraham ibn Ezra’s astrological works in Hebrew and Latin:
New discoveries and exhaustive listing’, Aleph 6, 239–338.
Stern, D. 2012, ‘The Hebrew Bible in Europe in the Middle Ages: A preliminary
typology’, Jewish Studies, an Internet Journal 11, 290–301.
Stern, S. and J. Isserles 2015, ‘The astrological and calendar section of the earliest
Mahzor Vitry manuscript (MS ex-Sassoon 535)’, Aleph 15/2, 199–318.
Ta-Shma, ˙ I. 2004, The Early Ashkenazic Prayer: Literary and Historical Aspects (in
Hebrew) (Jerusalem).
Toch, M. 2013, The Economic History of European Jews: Late Antiquity and Early
Middle Ages (Leiden).
Urbach, E. E. 1963, R. Abraham ben Azriel, Arugat ha-Bosem (in Hebrew)
(Jerusalem).
Vollandt, R. 2009, ‘Two fragments (T-S AS 72.79 and T-S Ar.1a.38) of Saadiah’s
tafsīr by Samuel ben Jacob’, Cambridge University Library Fragment of the
Month (November), www.lib.cam.ac.uk/Taylor-Schechter/fotm/november
-2009/.
Von Mutius, H.-G. 2006, Die Hebräischen Bibelzitate beim englischen Scholastiker
Odo (Frankfurt).
Weil-Guény, A.-M. 1991, ‘Les manuscrits bibliques de la Bibliothèque
Universitaire de Bologne’, Henoch 13/3, 287–317.

10 Liturgical Books
Chadd, D. 1986, ‘Liturgy and liturgical music: The limits of uniformity’, in
Cistercian Art and Architecture in the British Isles, ed. C. Norton and
D. Park (Cambridge), 299–314.
372 Bibliography
Colette, M.-N., et al., MANNO: Manuscrits notés en neumes en Occident, available
at saprat.ephe.sorbonne.fr.
Fassler, M. 1993, Gothic Song: Victorine Sequences and Augustinian Reform in
Twelfth-Century Paris (Cambridge).
Grotefend, H. 1891–8, Zeitrechnung des deutschen Mittelalters und der Neuzeit
(2 vols., Hanover).
Haines, J. 2008, ‘The origins of the musical staff’, Musical Quarterly 91, 327–78.
(ed.) 2011, The Calligraphy of Medieval Music (Musicalia medii aevi 1:
Turnhout).
Harper, J. 1991, The Forms and Orders of Western Liturgy from the Tenth to the
Eighteenth Century: A Historical Introduction and Guide for Students and
Musicians (Oxford).
Hartzell, K. D. 2006, Catalogue of Manuscripts Written or Owned in England up to
1200 Containing Music (Woodbridge).
Heinzer, F. 2008, Klosterreform und mittelalterliche Buchkultur im deutschen
Südwesten (Leiden).
[Huglo, M.] 1957, Le graduel romain: Édition critique par les moines de Solesmes, 2:
Les Sources (Solesmes).
1988, Les livres de chant liturgique, (Typologie des sources du Moyen Âge
occidental 52, Turnhout).
1999, 2004, Les Manuscrits du processional (2 vols., Répertoire international des
sources musicales B XIV: Munich).
2001, ‘The cantatorium: From Charlemagne to the fourteenth century’, in The
Study of Medieval Chant: Paths and Bridges, East and West, ed. P. Jeffery
(Woodbridge), 89–103.
Husmann, H. 1964, Tropen- und Sequenzenhandschriften (Répertoire international
des sources musicales B V1: Munich).
Iversen, G. 2010, Laus angelica: Poetry in the Medieval Mass, ed. J. Flynn, trans.
W. Flynn (Turnhout).
Jacobsson, R. M. and A. Haug 2001, ‘Versified office’, in The New Grove
Dictionary of Music and Musicians, 2nd edn. (London), 26. 493–9.
Jonsson, R. 1968, Historia: Études sur la genèse des offices versifiés (Studia Latina
Stockholmiensia 15: Stockholm).
Jungmann, J. 1951, The Mass of the Roman Rite: Its Origins and Development
(2 vols., New York; repr. Notre Dame, IN, 2012).
Kay, R. 2007, Pontificalia: A Repertory of Latin Manuscript Pontificals and
Benedictionals (Lawrence, KS), available at kuscholarworks.ku.edu.
King, A. A. 1955, Liturgies of the Religious Orders (London).
Klugseder, R., et al. 2014, Katalog der mittelalterlichen Musikhandschriften der
Österreichischen Nationalbibliothek Wien (Purkersdorf).
Leroquais, V. 1934, Les Bréviaires manuscrits des bibliothèques publiques de France
(6 vols., Paris).
1940–1, Les Psautiers manuscrits latins des bibliothèques publiques de France
(3 vols., Maçon).
Marosszéki, S. 1952, Les Origines du chant cistercien: Recherches sur les réformes du
plain-chant cistercien au XIIe siècle (Vatican City).
 Bibliography 373
McKitterick, R. 1993, Carolingian Culture: Emulation and Innovation
(Cambridge).
Meyer, C., et al. 2006–, Catalogue des manuscrits notés du Moyen Âge conservés dans
les bibliothèques publiques de France (Turnhout).
Morand, M. 2008, ‘Quand liturgie épousa prédication’, in Prédication et liturgie au
Moyen Âge, ed. N. Bériou and F. Morenzoni (Turnhout), 79–126.
Palazzo, E. 1998, A History of Liturgical Books from the Beginning to the Thirteenth
Century, trans. M. Beaumont (Collegeville, MN).
Parkes, Pause and Effect.
Pfaff, R. W. 2009, The Liturgy in Medieval England: A History (Cambridge).
Rasmussen, N. K. 1998, Les Pontificaux du haut Moyen Âge: Gènese du livre de
l’évêque, ed. M. Haverals (Spicilegium sacrum Lovaniense 49: Louvain).
Salmon, P. 1967, L’Office divin au Moyen Âge: Histoire de la formation du bréviaire
du IXe au XVIe siècle (Paris).
Tolhurst, J. B. L. (ed.) 1932–42, The Monastic Breviary of Hyde Abbey, Winchester
(6 vols., Henry Bradshaw Society 69–71, 76, 78, 80: London).
Van Dijk, S. J. P. and J. Hazelden Walker 1960, The Origins of the Modern Roman
Liturgy: The Liturgy of the Papal Court and the Franciscan Order in the
Thirteenth Century (Westminster, MD).
Vogel, C. 1986, Medieval Liturgy: An Introduction to the Sources
(Washington, DC).
Waddell, C. 2007, The Primitive Cistercian Breviary (Spicilegium Friburgense 44:
Fribourg).
Zieman, K. 2008, Singing the New Song: Literacy and Liturgy in Late Medieval
England (Philadelphia, PA).

11 Books of Theology and Bible Study

Primary Sources
Alexander of Hales, Glossa in IV Libros Sententiarum Petri Lombardi (4 vols.,
Quaracchi, 1951–7).
Andrée, A. 2005, Gilbertus Universalis: Glossa Ordinaria in Lamentationes Ieremie
Prophete. Prothemata et Liber I. A Critical Edition with an Introduction and
Translation (Stockholm).
Biblia cum glossa ordinaria (Strassburg: A. Rusch, 1480–1); Biblia latina cum glossa
ordinaria: Facsimile reprint of the editio princeps Adolph Rusch of Strassburg
1480/81, ed. K. Froehlich and M. T. Gibson (4 vols., Turnhout, 1992).
Dove, M. 1997, Glossa Ordinaria, Pars 22. In Canticum Canticorum, CCCM 170.
Hugh of St-Cher, Postilla in totam bibliam (Nuremberg, 1498–1502).
Hugh of St-Victor, De sacramentis christianae fidei, ed. R. Berndt (Münster, 2008);
ed. and trans. R. J. Deferrari (Cambridge, MA, 1951).
De archa Noe; Libellus de formatione arche, ed. P. Sicard, CCCM 176–176A (2001).
374 Bibliography
Moore, P. S. 1936, The Works of Peter of Poitiers (Notre Dame, IN).
Peter the Chanter, Summa de sacramentis et animis consiliis, ed. J.-A. Dugaquier
(Analecta mediaevalia Namurcensia 4, 7, 11, 16, 21: Louvain, Lille, 1954–63).
Peter Comestor, Historia scholastica, PL 198: 1053–1722; Petris Comestoris Scolastica
Historia, Liber Genesis, ed. A. Sylwan, CCCM 191 (2005).
Peter Lombard, Sententiae in IV Libris Distinctae, ed. I. F. Brady (2 vols., 1 in two
parts, Grottaferrata, 1971, 1981).
Peter of Poitiers, Sententiae, ed. P. S. Moore and M. Dulong (Notre Dame, IN,
1961).
Robert of Melun, Oeuvres, vol. 3: Sententiae, ed. R. M. Martin (Spicilegium
sacrum Lovaniense 21, 25: Louvain, 1947, 1952).

Secondary Sources
Baldwin, J. W. 1970, Masters, Princes and Merchants. The Social Views of Peter the
Chanter and His Circle (2 vols., Princeton, NJ).
Burman, T. E. 2007, Reading the Qur’ān in Latin Christendom, 1140–1560
(Philadelphia, PA).
Carruthers, M. 2008, The Book of Memory: A Study of Memory in Medieval Culture,
2nd edn. (Cambridge).
Colish, M. L. 1994, Peter Lombard (2 vols., Leiden).
Delano-Smith, C. 2012, ‘The exegetical Jerusalem: Maps and plans for Ezekiel
chapters 40–48’, in Imagining Jerusalem in the Medieval West, ed. L. Donkin
and H. Vorholt (Oxford), 41–75.
2013, ‘Maps and plans in medieval exegesis: Richard of St Victor’s In visionem
Ezechielis’, in From Knowledge to Beatitude: St Victor, Twelfth-Century Scholars
and Beyond. Essays in Honor of Grover A. Zinn, Jr., ed. E. A. Matter and
L. Smith (Notre Dame, IN), 1–45.
Flint, V. I. J. 1976, ‘The school of Laon: A reconsideration’, RTAM 43,
89–100.
Giraud, C. 2010, Per verba magistri: Anselme de Laon et son école au XIIe siècle
(Turnhout).
Gross-Diaz, T. 1996, The Psalms Commentary of Gilbert of Poitiers. From lectio
divina to the Lecture Room (Leiden).
Lottin, O. 1959, Psychologie et morale aux XIIe et XIIIe siècles, 5: L’Ėcole d’Anselme de
Laon et de Guillaume de Champeaux (Gembloux).
Morgan, N. J. and S. Panayotova 2015, A Catalogue of Western Book Illumination in
the Fitzwilliam Museum and the Cambridge Colleges, Part III: France, vol. 1:
c. 1000–c. 1250 (Turnhout).
Olszowy-Schlanger, J. 2009, ‘Christian Hebraism in thirteenth-century England’,
in Crossing Borders: Hebrew Manuscripts as a Meeting-Place of Cultures, ed.
P. van Boxel and S. Arndt (Oxford), 115–22.
Rudolph, C. 2004, ‘First, I Find the Center Point’: Reading the Text of Hugh of
St Victor’s The Mystic Ark (Philadelphia, PA).
 Bibliography 375
2014, The Mystic Ark. Hugh of Saint Victor, Art, and Thought in the Twelfth
Century (Cambridge).
Smalley, B. 1935–6, ‘Gilbertus Universalis, bishop of London (1128–34) and the
problem of the Glossa Ordinaria’, RTAM 7, 235–62; 8, 24–60.
1937, ‘La Glossa Ordinaria. Quelques prédécesseurs d’Anselme de Laon’, RTAM
9, 365–400.
1983, The Study of the Bible in the Middle Ages, 3rd edn. (Oxford).
1984, ‘Glossa Ordinaria’, in Theologische Realenzyklopädie 13, 452–57.
Smith, L. 2001, Masters of the Sacred Page: Manuscripts of Theology in the Latin West
to 1274 (Notre Dame, IN).
2008, ‘Medieval glossed psalters: Layout and use’, Bodleian Library Record 21,
48–61.
2009, The Glossa Ordinaria: The Making of a Medieval Bible Commentary
(Leiden).
Southern, R. W. 1995, 2001, Scholastic Humanism and the Unification of Europe, 1:
Foundations, 2: The Heroic Age (Oxford).
Van Elswijk, H. C. 1966, Gilbert Porreta: Sa vie, son oeuvre, sa pensée (Spicilegium
sacrum Lovaniense 33: Louvain).
Worm, A. 2012, ‘“Ista est Jerusalem”: Intertextuality and visual exegesis in Peter of
Poitiers’ Compendium historiae in genealogia Christi and Werner Rolevinck’s
Fasciculus temporum’, in Imagining Jerusalem in the Medieval West, ed.
L. Donkin and H. Vorholt (Oxford), 123–61.
Zier, M. A. 1993, ‘The manuscript tradition of the Glossa Ordinaria for Daniel,
and hints at a method for a critical edition’, Scriptorium 47, 3–25.

12 Logic
Boethius, De syllogismo categorico, ed. C. T. Thomsen Thornqvist (Studia Graeca
et Latina Gothoburgensia 68: Gotheburg, 2008).
Introductio ad syllogismos categoricos, ed. C. T. Thomsen Thornqvist (Studia
Graeca et Latina Gothoburgensia 69: Gotheburg, 2008).
Burnett, C. (ed.) 1993, Glosses and Commentaries on Aristotelian Logical Texts:
The Syriac, Arabic and Medieval Latin Traditions (London).
Cameron, M. 2011, ‘Abelard’s early glosses: Some questions’ in Rosier-Catach,
647–62.
Catalogue général des manuscrits latins. Tome IV (Nos3014 à 3277) (BnF, 1958).
Codices Boethiani: A Conspectus of Manuscripts of the Works of Boethius, I, ed.
M. T. Gibson, M. Passalacqua and L. Smith (London, 1995), II–III
(London and Turin, 2001), IV (London and Turin, 2010).
De Rijk, L. M. 1967, Logica Modernorum: A Contribution to the History of Early
Terminist Logic, II.1 (Assen).
Ebbesen, S. 1993, ‘Medieval Latin glosses and commentaries on Aristotelian logical
texts of the twelfth and thirteenth centuries’ in Burnett, Glosses and
Commentaries, 129–77.
376 Bibliography
2009, ‘The Aristotelian commentator’, in Marenbon, Cambridge Companion,
34–55.
2011, ‘Context-sensitive argumentation: Dirty tricks in the Sophistical
Refutations and a perceptive medieval interpretation of the text’, Vivarium
49, 75–94.
Gillespie, A. 2011, ‘Medieval books, their booklets, and booklet theory’, English
Manuscript Studies 1100–1700 16, 1–29.
Green-Pedersen, N. J. 1974, ‘William of Champeaux on Boethius’ Topics accord-
ing to Orléans Bibl. Mun. 266’, CIMAGL 13, 13–30.
1977, ‘The doctrine of “maxima propositio” and “locus differentia” in commen-
taries from the 12th century on Boethius’ “Topics”’, Studia Mediewistyczne 18,
125–63.
1984, The Tradition of the Topics in the Middle Ages. The Commentaries on
Aristotle’s and Boethius’ ‘Topics’ (Munich).
Grondeux, A. 2009, ‘Saintété et grammaire: Figures d’une mésentente. Gosvin
d’Anchin, Bernard d’Anchin et les Notae Dunelmenses’, in ‘Parva pro magnis
munera’: Études de littérature tardo-antique et médiévale offertes à François
Dolbeau par ses élèves, ed. M. Goullet (Turnhout), 883–918.
Gumbert, J. P. 1989, ‘L’unité codicologique ou: À quoi bon les cahiers?’, Gazette
du livre médiéval 14, 4–8.
1999, ‘One book with many texts: The Latin tradition’, in Codices miscella-
nearum, ed. R. Jansen-Sieben and H. van Dijk (Brussels), 27–36.
2004, ‘Codicological units: Towards a terminology for the stratigraphy of the
non-homogenous codex’, Segno e Testo 2, 17–42.
Hanna, R. 1986, ‘Booklets in medieval manuscripts: Further considerations’,
Studies in Bibliography 39, 100–11.
Hansen, H. 2005, ‘An early commentary on Boethius’s De topicis differentiis’,
CIMAGL 76, 45–130.
Hunt, R. W. 1948, ‘The introductions to the “Artes” in the twelfth century’, in
Studia mediaevalia in honorem admodum Reverendi Patris Raymundi Josephi
Martin (Bruges), 85–112; repr. R. W. Hunt, The History of Grammar in the
Middle Ages. Collected papers, ed. G. L. Bursill-Hall (Amsterdam, 1980).
Iwakuma, Y. 1993, ‘Introductiones dialecticae artis secundum magistrum
G. Paganellum’, CIMAGL 63, 45–114.
1999, ‘Pierre Abélard et Guillaume de Champeaux dans les premières années du
XIIe siècle: Une étude préliminaire’, in Langage, sciences, philosophie au XIIe
siècle, ed. J. Biard (Paris), 93–123.
2003, ‘William of Champeaux on Aristotle’s Categories’, in La tradition
médiévale des Catégories (XIIe–XIVe siècles), ed. J. Biard and I. Rosier-
Catach (Louvain-la-Neuve/Paris), 313–28.
2008, ‘Pseudo-Rabanus super Porphyrium (P3)’, AHDLMA 75, 43–196.
2009, ‘Vocales revisited’, in The Word in Medieval Logic, Theology and
Psychology, ed. T. Shimizu and C. Burnett (Turnhout), 81–171.
unpublished, ‘Prologues of Commentaries on the Logica vetus Literature in the
12th Century’.
 Bibliography 377
Iwakuma, Y. and S. Ebbesen 1992, ‘Logico-theological schools from the second
half of the twelfth century’, Vivarium 30, 173–210.
Jacobi, K. 2011, ‘William of Champeaux: Remarks on the tradition in the manu-
scripts’, in Rosier-Catach, 261–71.
Jeauneau, ‘Prologue’.
John of Salisbury, Metalogicon, ed. J. B. Hall, CCCM 98 (1991).
Kwakkel, E. 2002, ‘Towards a terminology for the analysis of composite manu-
scripts’, Gazette du Livre Médiéval 41, 12–19.
2012, ‘Late medieval text collections: A codicological typology based on single-
author manuscripts’, in Author, Reader, Book: Medieval Authorship in Theory
and Practice, ed. S. Partridge and E. Kwakkel (Toronto), 56–79.
Magee, J. and J. Marenbon 2009, ‘Appendix: Boethius’ works’, in Marenbon,
Cambridge Companion, 303–10.
Marenbon, J. 1981, From the Circle of Alcuin to the School of Auxerre: Logic, Theology
and Philosophy in the Early Middle Ages (Cambridge).
1993, ‘Medieval Latin commentaries and glosses on Aristotelian logical texts,
before c. 1150 AD’, in Burnett, Glosses and Commentaries, 77–127; repr., with
additional material, in Marenbon, Aristotelian Logic, art. II.
1997, ‘Glosses and commentaries on the Categories and De interpretatione before
Abelard’, in Dialektik und Rhetorik im früheren un hohen Mittelalter, ed.
J. Fried (Munich), 21–49; repr. Marenbon, Aristotelian Logic, art. IX.
1997a, The Philosophy of Peter Abelard (Cambridge).
2000, Aristotelian Logic, Platonism, and the Context of Early Medieval Philosophy
in the West (Aldershot).
2007, Medieval Philosophy: An Historical and Philosophical Introduction
(London and New York).
(ed.) 2009, The Cambridge Companion to Boethius (Cambridge).
2013, ‘La logique en occident latin (ca.780–ca. 1150): Le programme des études et
ses enjeux’, in ‘Ad notitiam ignoti’: The Organon in the Translatio Studiorum at
the Time of Albert the Great. Orders of Treatises, Divisions of Logic and Textual
Transmissions, ed. J. Brumberg-Chaumont (Turnhout), 173–91.
2013a, ‘The tradition of studying the Categories in the early Middle Ages
(until c. 1200): a revised working catalogue of glosses, commentaries and
treatises’, in Aristotle’s Categories in the Byzantine, Arabic and Latin
Traditions, ed. S. Ebbesen, J. Marenbon and P. Thom (Copenhagen),
139–73.
Martin, C. 2011, ‘A note on the attribution of the literal glosses in Paris, BnF, lat.
13368 to Peter Abelard’, in Rosier-Catach, 605–46.
Mews, C. J. 1985, ‘On dating the works of Peter Abelard’, AHDLMA 52, 73–134;
repr. Mews, Abelard and His Legacy (Aldershot, 2001).
2005, ‘Logica in the service of philosophy: William of Champeaux and his
influence’, in R. Berndt (ed.), Schrift, Schreiber, Schenker: Studien zur Abtei
Sankt Viktor in Paris und den Viktorinern (Berlin), 77–117.
Minio-Paluello, L. 1958, Twelfth Century Logic, Texts and Studies, II: Abaelardiana
inedita (Rome).
378 Bibliography
Munk Olsen, B. 1998, ‘L’élément codicologique’, in Recherches de codicologie
comparée: La composition du codex au Moyen Âge, en Orient et en Occident,
ed. P. Hoffmann (Paris), 105–29.
Peter Abelard, Dialectica, ed. L. M. de Rijk, 2nd edn. (Assen, 1970).
Poirel, D. 2011, ‘Datation des textes et traitement des recensions multiples’, in
Rosier-Catach, 249–59.
Robinson, P. R. 1980, ‘The ‘booklet’: A self-contained unit in composite manu-
scripts’, in Codicologica 3. Essais typologiques, ed. A. Gruys and J. P. Gumbert
(Leiden), 46–69.
Rosier-Catach, I. 2011, Arts du langage et théologie aux confins des XIe–XIIe siècles:
Textes, maîtres, débats (Turnhout).
Thomsen Thornqvist, C. 2010, ‘The “Anonymous Aurelianensis III” and the
reception of Aristotle’s Prior Analytics in the Latin West’, CIMAGL 79,
25–41.
Thomson, R. M. 2011, 2013, Catalogue of Medieval Manuscripts of Latin Commentaries
on Aristotle in British Libraries, I: Oxford; II: Cambridge (Turnhout).

13 Old Texts in New Contexts

Baswell, C. 1999, ‘Latinitas’, in The Cambridge History of Medieval English
Literature, ed. D. Wallace (Cambridge), 122–51.
Bozzolo, C. and E. Ornato 1983, Pour une histoire du livre manuscrit au Moyen Âge:
Trois essais de codicologie quantitative (Paris).
Brown, G. 1994, ‘Introduction: The Carolingian renaissance’, in Carolingian
Culture: Emulation and Innovation, ed. R. McKitterick (Cambridge), 1–51.
Brown, V. 1972, The Textual Transmission of Caesar’s Civil War (Leiden).
Burnett, C. 1984, ‘The content and affiliation of the scientific manuscripts written
at, or brought to, Chartres in the time of John of Salisbury’, in The World of
John of Salisbury, 127–60.
Contreni, J. 1995, ‘The Carolingian renaissance: Education and literary culture’, in
The New Cambridge Medieval History, II: c. 700–c. 900, ed. R. McKitterick
(Cambridge).
Copeland, R. and I. Sluiter 2009, Medieval Grammar and Rhetoric: Language Arts
and Literary Theory, AD 300–1475 (Oxford).
Dronke, P. 1988, ‘Thierry of Chartres’, in A History of Twelfth-Century Western
Philosophy, ed. P. Dronke (Cambridge), 358–85.
Hunt, R. W. 1948, ‘The introductions to the “Artes” in the twelfth century’, in
Studia mediaevalia in honorum admodum Reverendi Patris Raymundi Josephi
Martin (Bruges), 85–112.
Hunt, T. 1991, Teaching and Learning in Thirteenth-Century England, I: Texts
(Cambridge).
Huygens, R. B. C. 1970, Accessus ad Auctores (Leiden).
Jaeger, C. S. 2003, ‘Pessimism in the twelfth-century “Renaissance”’, Speculum 78,
151–83.
 Bibliography 379
Ker, J. 2009, The Many Deaths of Seneca (Oxford).
Kwakkel, E. 2015, ‘Decoding the material book: Cultural residue in medieval
manuscripts’, in The Medieval Manuscript Book: Cultural Approaches, ed. M.
Johnston and M. Van Dussen (Cambridge), 60–76.
Lowe, E. A. 1925, ‘Some facts about our oldest Latin manuscripts’, Classical
Quarterly 19, 197–208; repr. in his Palaeographical Papers, ed. L. Bieler
(2 vols., Oxford, 1972), 1. 187–202.
Martianus Capella, The Marriage of Philosophy and Mercury, trans. W. H. Stahl, R.
Johnson and E. L. Burge, Martianus Capella and the Seven Liberal Art (2 vols.,
New York, 1971), vol. 2.
Martin, J. 1984, ‘John of Salisbury as a classical scholar’, in The World of John of
Salisbury, 179–201.
Minnis, A. 1988, Medieval Theory of Authorship: Scholastic Literary Attitudes in the
Later Middle Ages, 2nd edn. (Philadelphia, PA).
Munk Olsen, B. 1982–2014, L’étude des auteurs classiques latins aux XIe et XIIe siècles
(4 vols., Paris).
1979, 1980, ‘Les classiques latins dans les florileges médiévaux anterieurs au XIIIe
siècle’, Revue d’histoire des textes 9, 47–21, 10, 115–64.
1984, ‘The Cistercians and classical culture’, CIMAGL 47, 64–102; repr. in La
réception de la littérature classique au Moyen Âge, ed. K. Friis-Jensen
(Copenhagen, 1995), 95–132.
1996, ‘The production of the classics in the eleventh and twelfth centuries’, in
Chavannes-Mazel and Smith, Latin Classics, 1–18.
Parkes, M. B. 1992, Pause and Effect: An Introduction to the History of Punctuation
in the West (Aldershot).
Quain, E. A. 1945, ‘The medieval Accessus ad Auctores’, Traditio 3, 215–64.
Reynolds, L. D. 1965, The Medieval Tradition of Seneca’s Letters (Oxford).
1983, Texts and Transmission: A Survey of the Latin Classics (Oxford).
Reynolds, L. D. and N. G. Wilson 1991, Scribes and Scholars: A Guide to the
Transmission of Greek and Latin Literature, 3rd edn. (Oxford).
Reynolds, S. 1996, Medieval Reading: Grammar, Rhetoric and the Classical Text
(Cambridge).
Reijnders, H. F. 1972, ‘Aimericus, Ars lectoria (3)’, Vivarium 10, 124–76.
Rouse, M. A. and R. H. Rouse 1975, ‘The Florilegium Angelicum: Its origin, content,
and influence’, in Hunt Essays, 66–114; repr. in their Authentic Witnesses, 101–52.
1979, Preachers, Florilegia and Sermons: Studies in the ‘Manipulus Florum’ of
Thomas of Ireland (Toronto).
Sears, E. 2002, ‘Portraits in counterpoint: Jerome and Jeremiah in an Augsburg
manuscript’, in Reading Medieval Images: The Art Historian and the Object,
ed. E. Sears and T. K. Thomas (Ann Arbor, MI), 61–74.
Stump, E. 1978, Boethius’s De topicis differentiis: Translated with Notes and Essays on
the Text (Ithaca, NY).
Tahkokallio, J. 2015, ‘The classicization of the Latin curriculum and “The
Renaissance of the Twelfth Century”: A quantitative study of the codicolo-
gical evidence’, Viator 46, 129–54.
380 Bibliography
Thierry of Chartres, ‘Commentarius super libros de inventione’, in The Latin Rhetorical
Commentaries by Thierry of Chartres, ed. K. M. Fredborg (Toronto, 1988).
Thomson, R. M. 2011, A Descriptive Catalogue of the Medieval Manuscripts of
Corpus Christi College, Oxford: Western Manuscripts (Cambridge).
Ward, J. O. 2003, ‘Ciceronian rhetoric and oratory from St. Augustine to Guarino
da Verona’, in Cicero Refused to Die: Ciceronian Influence through the
Centuries, ed. N. van Deusen (Leiden), 163–96.
Wieland, G. 1999, ‘Interpreting the interpretation: The polysemy of the Latin
gloss’, JML 8, 59–71.
Zetzel, J. 2005, Marginal Scholarship and Textual Deviance: The Commentum
Cornuti and the Early Scholia on Persius (Bulletin of the Institute of
Classical Studies of the University of London Supplement 84: London).

14 Reading the Sciences

Behrends, F. (ed.) 1976, The Letters and Poems of Fulbert of Chartres (Oxford).
Borelli, A. 2008, Aspects of the Astrolabe: ‘Architectonica Ratio’ in Tenth- and
Eleventh-Century Europe (Stuttgart).
Burnett, C. 1978, ‘Arabic into Latin in twelfth-century Spain: The works of
Hermann of Carinthia’, MJ 13, 99–134.
1984, ‘The contents and affiliation of the scientific manuscripts written at, or
brought to, Chartres in the time of John of Salisbury’, in The World of John of
Salisbury, 127–60.
1996, ‘Algorismi vel helcep decentior est diligentia: The arithmetic of Adelard of
Bath and his circle’, in Mathematische Probleme im Mittelalter: Der lateinische
und arabische Sprachbereich, ed. M. Folkerts (Wiesbaden), 221–331; repr. in
Burnett, Numerals and Arithmetic, art. III.
1997, ‘The instruments which are the proper delights of the quadrivium:
Rhythmomachy and chess in the teaching of arithmetic in twelfth-century
England’, Viator 28, 175–201.
2002, ‘The abacus at Echternach in ca. 1000 A.D.’, Sciamus 3, 91–108; repr. in
Burnett, Numerals and Arithmetic, art. I.
2002a, ‘Physics before the Physics: Early translations from Arabic of texts
concerning nature in MSS British Library, Additional 22719 and Cotton
Galba E. IV’, Medioevo 27, 53–109; repr. in Burnett, Arabic into Latin in the
Middle Ages, art. II.
2002b, ‘Indian numerals in the Mediterranean basin in the twelfth century,
with special reference to the “Eastern Forms”’, in From China to Paris: 2000
Years’ Transmission of Mathematical Ideas, ed. Y. Dold-Samplonius, J. W.
Dauben, M. Folkerts and B. van Dalen (Stuttgart), 237–88; repr. in Burnett,
Numerals and Arithmetic, art. V.
2003–4, ‘The use of Arabic among the three language cultures of Norman
Sicily’, in Art and Form in Norman Sicily, ed. D. Knipp, Römisches Jahrbuch
der Bibliotheca Hertziana 35, 39–48.
 Bibliography 381
2004, ‘Lunar astrology: The varieties of texts using lunar mansions, with
emphasis on Jafar Indus’, Micrologus 12, 43–133.
2004a, ‘Arabic and Latin astrology compared in the twelfth century: Firmicus,
Adelard of Bath and “Doctor Elmirethi” (“Aristoteles Milesius”)’, in Studies
in the History of the Exact Sciences in Honour of David Pingree, ed. C. Burnett,
J. P. Hogendijk, K. Plofker and M. Yano (Leiden), 247–63.
2006, ‘A Hermetic programme of astrology and divination in mid-twelfth-
century Aragon: The hidden preface in the Liber novem iudicum’, in Magic
and the Classical Tradition, ed. W. F. Ryan and C. Burnett (London), 99–118.
2007, ‘The “translation” of diagrams and illustrations from Arabic into Latin’,
in Arab Painting: Text and Image in Illustrated Arabic Manuscripts, ed.
A. Contadini (Leiden), 161–76.
2009, Arabic into Latin in the Middle Ages: The Translators and Their Intellectual
and Social Context (Farnham).
2010, Numerals and Arithmetic in the Middle Ages (Farnham).
2011, ‘Al-Qabīsī’s Introduction to Astrology: From courtly entertainment to
˙
university textbook’, in Studies in the History of Culture and Science: A
Tribute to Gad Freudenthal, ed. R. Fontaine, R. Glasner, R. Leicht and
G. Veltri (Leiden), 43–89.
2013, ‘The geometry of the Liber Ysagogarum Alchorismi’, Sudhoffs Archiv 97,
143–73.
2013a, ‘Simon of Genoa’s use of the Breviarium of Stephen, the Disciple of
Philosophy’, in Simon of Genoa’s Medical Lexicon, ed. B. Zipser (London),
67–78.
2017 ‘Béziers as an astronomical centre for Jews and Christians in the mid-
twelfth century’, Aleph 17.
Burnett, C. and D. Jacquart (eds.). 1994, Constantine the African and ‘Ali ibn al-
’Abbas al-Magusi: The ‘Pantegni’ and Related Texts (Leiden).
Burnett, C. and P. Jones 2008, ‘Scientific and medical writings’, in CHBB II.
446–62.
Carey, H. M. 2004, ‘Astrological medicine and the medieval English folded
almanac’, Social History of Medicine 17, 345–63.
Classen, P. 1974, Burgundio von Pisa, Richter – Gesandter – Übersetzer
(Heidelberg).
Endress, G. 2001, ‘Philosophische Einband-Bibliotheken aus Isfahan’, Oriens 36,
10–58.
Evans, G. 1983, ‘The uncompleted Hepteuchon’, History of Universities 3, 1–13.
Falmagne, T. 2009, Die Echternacher Handschriften bis zum Jahr 1628 in den
Beständen der Bibliothèque nationale de Luxembourg (2 vols., Wiesbaden).
Folkerts, M. 1970, ‘Boethius’ Geometrie II (Wiesbaden).
1996, ‘Frühe Darstellungen des Gerbertschen Abakus’, in Itinera mathematica:
Studi in onore di Gino Arrighi per il suo 90a compleanno (Siena), 23–43.
Grupe, D. (ed. and trans.), Stephen the Philosopher, Liber Mamonis (in preparation).
382 Bibliography
Gumbert, J. P. 1993, ‘Sizes and formats’, in Ancient Book Materials and Techniques,
ed. M. Maniaci and P. Munafò (Studi e Testi 358: 2 vols., Vatican City), 1.
227–63.
Haskins, C. H. 1927, Studies in the History of Medieval Science, 2nd edn.
(Cambridge, MA).
Irblich, E. 1981, ‘Einfluss von Vorlage und Text im Hinblick auf kodikologische
Erscheinungsformen am Beispiel der Überlieferung der “Chirurgie” des
Abu’l Qasim Halaf Ibn ’Abbas al-Zahraui vom 13. Jahrhundert bis 1500’,
Münchener Beiträge zur Mediävistik und Renaissance-Forschung 32, 209–31.
Jacquart, D. 1994, ‘Le sens donné par Constantin l’Africain à son œuvre: Les
chapitres introductifs en arabe et en latin’, in Burnett and Jacquart,
Constantine, 71–89.
Jeauneau, E. 1954, ‘Le Prologus in Eptateuchon de Thierry de Chartres’, Mediaeval
Studies 14, 171–5, repr. in his ‘Lectio philosophorum’: Recherches sur l'école de
Chartres (Amsterdam, 1973), 87–91.
Juste, D. 2007, Les Alchandreana primitifs (Leiden).
2015, Les manuscrits astrologiques latins conservés à la Bibliothèque nationale de
France (Paris, 2015).
Juste, D. and C. Burnett 2016, ‘A new catalogue of medieval translations into Latin
of texts on astronomy and astrology’, in Medieval Textual Culture: Agents
of Transmission, Translation and Transformation, ed. F. Wallis and
R. Wisnovsky (Berlin and Boston), 63–76.
Kunitzsch, P. 1998, ‘Traces of a tenth-century Spanish-Arabic astrolabe’, Zeitschrift
für Geschichte der arabisch-islamischen Wissenschaften 12 (1998), 113–20.
Kwakkel, E. and F. Newton, Medicine in Monte Cassino: Constantine the African
and the Oldest Manuscript of his Pantegni (Turnhout: Brepols, in press).
Leino, M. and C. Burnett 2003, ‘Myth and astronomy in the frescoes at
Sant’Abbondio in Cremona’, JWCI 66, 273–88.
Mandosio, J.-M. 2010, ‘L’Humanisme ou barbarie? Formes de la latinité et
mémoire de l’Antiquité dans quelques traductions médiévales de textes
philosophiques arabes’, in Écritures latines de la mémoire de l’Antiquité au
XVIe siècle, ed. H. Casanova-Robin and P. Galand (Paris), 227–63.
Maurach, G. (ed.) 1979, ‘Daniel of Morley, Philosophia’, MJ 14, 204–55.
Millàs Vallicrosa, J. 1931, Assaig d’Història de les idees físiques i matemàtiques a la
Catalunya Medieval (Barcelona).
Newton, F. 1998, The Scriptorium and Library of Montecassino, 1058–1105
(Cambridge).
North, J. D. 1987, ‘Some Norman horoscopes’, in Adelard of Bath: An English
Scientist and Arabist of the Early Twelfth Century, ed. C. Burnett (London),
147–61.
Nothaft, P. (ed.) 2017, Walcher of Malvern, De lunationibus and De Dracone:
Study, Edition, Translation, and Commentary (Turnhout).
Ragep, J. and E. S. Kennedy 1981, ‘A description of Zahiriyya (Damascus) MS
4871: A philosophical and scientific collection’, Journal for the History of
Arabic Science 5, 85–108.
 Bibliography 383
Reeve, M. 1980, ‘Some astronomical manuscripts’, Classical Quarterly 30, 508–22.
Robinson, P. R. 1980, ‘The “booklet”: A self-contained unit in composite manu-
scripts’, in Codicologica 3, ed. A. Gruys. and J. P. Gumbert (Leiden), 46–69.
Schramm, M., et al. 2006–7, ‘Der Astrolabtext aus der Handschrift Codex 196,
Burgerbibliothek Bern’, Zeitschrift für Geschichte der arabisch-islamischen
Wissenschaften 17, 199–300.
Schupp, F. (ed.) 2005, Al Farabi, Über die Wissenschaften/De scientiis (Hamburg).
Sezgin, F., with J. Hogendijk and F. Käs 2010, Codex Ayasofya 4832: A Collection of
Mathematical, Philosophical, Meteorological, and Astronomical Treatises
(Frankfurt).
Toneatto, L. 1994–5, Codices artis mensoriae: I manoscritti degli antichi opuscoli
latini d’agrimensura (V–XIX secolo) (3 vols., Spoleto).
Van Koningsveld, P. 1977, The Latin-Arabic Glossary of the Leiden University
Library (Leiden).
Vuillemin-Diem, G. and M. Rashed 1997, ‘Burgundio de Pise et ses manuscrits
grecs d’Aristote: Laur. 87.7 et Laur. 81.18’, RTAM 64, 136–98.

15 Medical Books
Barker-Benfield, B. C., 2008 St Augustine's Abbey, Canterbury, CBMLC 13, 3 vols.
Beccaria, A. 1956, I Codici di medicina del periodo presalernitano (secoli IX, X e XI)
(Rome).
Bos, G. 1994, ‘Ibn al-Ğazzār’s Risāla fī n-nisyān and Constantine’s Liber de
oblivione’ [with an edition of the Latin De oblivione by C. Burnett], in
Burnett and Jacquart, Constantine, 203–32.
Cahn, W. 1996, Romanesque Manuscripts: The Twelfth Century (a survey of manu-
scripts illuminated in France: 2 vols., London).
D’Aronco, M. A. 1998, The Old English Illustrated Pharmacopoeia: British Library
Cotton Vitellius C III (Copenhagen).
Ferraces Rodríguez, A. 2013, ‘Ars medicinalis de animalibus. Estudio y edición
crítica de un anecdotum de zooterapia altomedieval’, Myrtia 28, 175–241.
Folkerts, M. 1972, ‘Pseudo-Beda: De arithmeticis propositionibus, eine mathema-
tische Schrift aus der Karolingerzeit’, Sudhoffs Archiv 56/1, 22–43.
Freudenthal, G. 2013, ‘The father of the Latin-into-Hebrew translations: “Doeg
the Edomite”, the twelfth-century repentant convert’, in Latin-into-Hebrew:
Texts and Studies, ed. G. Veltri (2 vols., Leiden), 1. 105–20.
Galdi, A. 2014, ‘S. Benedetto tra Montecassino e Fleury (VII–XII secolo)’, in
Mélanges de l’École française de Rome – Moyen Âge 126–2, available at http://
mefrm.revues.org/2047.
Green, M. H. 2008, ‘Rethinking the manuscript basis of Salvatore De Renzi’s
Collectio Salernitana: The corpus of medical writings in the “long” twelfth
century’, in La ‘Collectio Salernitana’ di Salvatore De Renzi, ed. D. Jacquart
and A. Paravicini Bagliani (Florence), 15–60.
384 Bibliography
2011, ‘Moving from philology to social history: The circulation and uses of
Albucasis’s Latin Surgery in the Middle Ages’, in Between Text and Patient:
The Medical Enterprise in Medieval and Early Modern Europe, ed. F. E. Glaze
and B. Nance (Florence), 331–72.
Hoffmann, H. (ed.) 1980, Die Chronik von Montecassino, MGH Scriptores
XXXIV.
Knight, V. 2015, ‘The De podagra (On Gout): A pre-Gariopontean treatise excerpted
from the Latin translation of the Greek Therapeutica by Alexander of Tralles’,
PhD thesis, University of Manchester.
Kwakkel, E. and F. Newton, Medicine in Monte Cassino: Constantine the
African and the Oldest Manuscript of his Pantegni (Turnhout: Brepols, in
press).
Langslow, D. 2006, The Latin Alexander Trallianus: The Text and Transmission of
a Late Latin Medical Book (Journal of Roman Studies Monographs 10:
London).
Manzanero Cano, F. 1996, Liber Esculapii (Anonymus Liber Chronicorum). Edición
crítica y estudio. PhD Thesis. Universidad Complutense de Madrid.
Mostert, M. 1989, The Library of Fleury: A Provisional List of Manuscripts
(Hilversum).
Newton, F. 1994, ‘Constantine the African and Monte Cassino: New elements and
the text of the Isagoge’, in Burnett and Jacquart, Constantine, 16–47.
Orofino, G. 2015, ‘Gemelli diversi: Trasmissione e circolazione degli erbari in età
sveva’, in Medioevo: Natura e Figura. La raffigurazione dell’uomo e della natura
nell’arte medievale, ed. A. C. Quintavalle (Milan), 505–16.
Pradel-Baquerre, M. 2013, ‘Ps.-Apulée, Herbier, introduction, traduction et com-
mentaire’, PhD Thesis, Université Paul Valéry – Montpellier III.
Reeve, M. D. 2000, ‘The transmission of Vegetius’ Epitoma Rei Militaris’, Aevum
74, 243–354.
Scheller, R. W. 1995, Exemplum: Model-Book Drawings and the Practice of Artistic
Transmission in the Middle Ages (ca. 900–ca. 1470), trans. M. Hoyle
(Amsterdam).
Schipke, R. 2007, Die lateinischen Handschriften in Quarto der Staatsbibliothek zu
Berlin - Preußischer Kulturbesitz, Teil 1: Ms. lat. quart. 146-406 (Kataloge der
Handschriftenabteilung, Erste Reihe 6.1: Wiesbaden).
Stones, A. 2014, Gothic Manuscripts, 1260–1320 (2 vols., London).
Sudhoff, K. 1920, ‘Die Salernitaner Handschrift in Breslau’, Archiv für Geschichte
der Medizin 12, 101–48.
Ventura, I. (ed.) 2009, Ps. Bartholomaeus Mini de Senis: Tractatus de herbis
(Ms London, British Library, Egerton 747) (Florence).

16 Law Books
Austin, G. 2009, Shaping Church Law around the Year 1000 (Farnham).
 Bibliography 385
Brett, M. and K. G. Cushing (eds.) 2009, Readers, Texts and Compilers in the
Earlier Middle Ages: Studies in Medieval Canon Law in Honour of Linda
Fowler-Magerl (Farnham).
Ciaralli, A. 2000, ‘Ancora sul manoscritto pistoiese del Codex (Arch. Cap. C 106).
Note paleografiche e codicologiche’, Scrittura e Civiltà 24, 173–226.
Colli, V. (ed.) 2002, Juristische Buchproduktion im Mittelalter (Ius Commune,
Sonderheft 155: Frankfurt).
Cushing, K. G. 1998, Papacy and Law in the Gregorian Revolution. The Canonistic
Work of Anselm of Lucca (Oxford).
Dolezalek, G. 1972, Verzeichnis der Handschriften zum römischen Recht bis 1600
(4 vols., Frankfurt).
1984, ‘Index manuscriptorum veterum Infortiati’, Ius Commune 11, 281–7.
1985, Repertorium Manuscriptorum Veterum Codicis Iustiniani (Ius Commune,
Sonderheft 23: Frankfurt).
1989, ‘La pecia e la preparazione dei libri giuridici nei secoli XII–XII’, in Luoghi e
metodi d’insegnamento nell’Italia medioevale (secoli XII–XIV): Atti del convegno
internazionale di studi (Lecce-Otranto 6–8 ottobre 1986), ed. L. Gargan and
O. Limone (Galatina), 201–17.
Gobbitt, T. 2014, ‘Codicological features of a late-eleventh-century manuscript of
the Lombard Laws’, Studia Neophilologica 86, Supplement 1, 48–67.
Landau, P. 2008, ‘Gratian and the Decretum Gratiani’, in The History of Medieval
Canon Law in the Classical Period, 1140–1234: From Gratian to the Decretals of
Pope Gregory IX, ed. K. Pennington and W. Hartmann (Washington, DC),
22–54.
Liber Papiensis, ed. A. Boretius, MGH Leges IV.
McKitterick, R. 1989, The Carolingians and the Written Word (Cambridge).
Meyer, C. H. F. 2003, ‘Langobardisches Recht nördlich der Alpen unbeachtete
Wanderungen gelehrten Rechts im 12.–14. Jahrhundert’, Tijdschrift voor
Rechtsgeschiednis 71, 387–402.
Mordek, H. 1995, Bibliotheca capitularium regum Francorum manuscripta:
Überlieferung und Traditionszussammenhang der fränkischen Herrschererlasse,
MGH: Hilfsmittel 15 (Munich).
Orlandelli, G. 1956–7, ‘Ricerche sulla origine della “littera bononiensis”: Scritture
documentarie bolognesi del secolo XII’, Bollettino dell’Archivio Paleografico
Italiano 2–3.2, 179–214.
Pennington, K. 2009, ‘Decretal collections, 1190–1234’, in K. Pennington and
Hartmann, History of Medieval Canon Law (Washington, DC), 293–317.
Petrucci, A. and C. Romeo 1992, ‘Scrivere “in iudicio” nel “Regnum Italiae”’, in
Scriptores in urbibus: Alfabetismo e cultura scritta nell’Italia altomedievale
(Bologna), 195–236.
1995, Writers and Readers in Medieval Italy, trans. C. Radding (New Haven, CT
and London)
Radding, C. 1988, The Origins of Medieval Jurisprudence: Pavia and Bologna,
850–1150 (New Haven, CT).
386 Bibliography
1997, ‘Petre te appellat Martinus. Eleventh-century judicial procedure as seen
through the glosses of Walcausus’, in La Giustizia nell’Alto medioevo II (secoli
IX–XI), XLIVa Settimana di studio sull’Alto Medioevo, Spoleto, 11–17 aprile
1996 (Spoleto), 827–61.
2013, Le origini della giurisprudenza medieval. Una storia culturale, trans.
A. Ciaralli (Rome).
Radding, C. and A. Ciaralli 2006, The Corpus Iuris Civilis in the Middle Ages:
Manuscripts and Transmission from the Sixth Century to the Juristic Revival (Leiden).
Rolker, C. 2010, Canon Law and the Letters of Ivo of Chartres (Cambridge).
Rouse, M. A. and R. H. Rouse 1994, ‘The dissemination of texts in pecia at
Bologna and Paris’, in Rationalisierung der Buchherstellung im Mittelalter und
in der frühen Neuzeit, ed. P. Rück and M. Boghardt (Marburg), 69–77.
Savigny, F. C. 1834–51, Geschichte des römischen Rechts im Mittelalter, 2nd edn.
(7 vols., Heidelberg).
Soetermeer, F. 2005, ‘Between codicology and legal history: Pecia manuscripts of
legal texts’, Manuscripta 49, 247–67.
Winroth, A. 2000, The Making of Gratian’s Decretum (Cambridge).
2013, ‘Where Gratian slept: The life and death of the father of canon law’,
Zeitschrift der Savigny-Stiftung für Rechtsgeschichte. Kanonistische Abteilung
99, 105–28.

17 Vernacular Manuscripts I

Primary Sources
Chretien de Troyes, Le Chevalier au Lion (Yvain), ed. M. Roques (Paris, 1965).
Curley, M. J. 1982, ‘A new edition of John of Cornwall’s Prophetia Merlini’,
Speculum 57, 217–49.
Dante, De Vulgari Eloquentia, ed. and trans. S. Botterill (Cambridge, 2005).
Denis Piramus, La vie seint Edmund le rei, ed. H. Hjellman (Gothenburg, 1935).
Geffrei Gaimar, Estoire des Engleis/History of the English, ed. and trans. I. Short
(Oxford, 2009).
Geoffrey of Monmouth, The History of the Kings of Britain, ed. M. D. Reeve, trans.
N. Wright (Woodbridge, 2007).
Gerald of Wales, Expugnatio Hibernica, ed. A. B. Scott and F. X. Martin (Dublin,
1978).
Greenway, D. E. (ed.) 1972, Charters of the Honour of Mowbray, 1107–1191
(London).
Hartmann von Aue, Der arme Heinrich, ed. H. Paul, 18th edn. (rev. K. Gärtner,
Berlin, 2010).
Henry Archdeacon of Huntingdon, Historia Anglorum, ed. and trans. D. E.
Greenway (Oxford, 1996).
Hue de Rotelande, Protheselaus, ed. A. J. Holden (London, 1991).
 Bibliography 387
Wace, Roman de Rou, ed. and trans. G. S. Burgess, A. J. Holden and E. Van Houts,
(Jersey, 2002).
Wright, N. (ed.) 1984, The Historia Regum Britannie of Geoffrey of Monmouth I:
Bern Burgerbibliothek MS 568 (Cambridge).

Modern Literature
Asperti, S. 2006, Origini romanze: Lingue, testi antichi, letterature (Rome).
Aurell, M. 2011, Le Chevalier lettré: Savoir et conduite de l’aristocratie aux XIIe et
XIIIe siècles (Paris).
Avril, F. and M.-Th. Gousset 1984, Manuscrits enluminés d’origine italienne II:
XIIIe siècle (Paris).
Bäuml, F. H. 1980, ‘Varieties and consequences of medieval literacy and illiteracy’,
Speculum 55, 237–65.
Beer, J. 1968, Villehardouin: Epic Historian (Geneva).
Bepler, J., P. Kidd and J. Geddes 2008, The St Albans Psalter (Albani Psalter):
Facsimile and Commentary (2 vols., Simbach am Inn).
Blaess, M. 1973, ‘Les manuscrits français dans les monastères anglais au Moyen
Âge’, Romania 94, 321–58.
Bogaert, P. 1992, ‘Bible française’, in Dictionnaire des lettres françaises: Le Moyen
Âge, ed. G. Hasenohr and M. Zink (Paris), 189–92.
Brooke, C. N. L. 1971, The Twelfth-Century Renaissance (London).
Bumke, J. 1991, Courtly Culture: Literature and Society in the High Middle Ages,
trans. T. Dunlap (Berkeley, CA).
Busby, K., 2002, Codex and Context: Reading Old French Verse Narratives in
Manuscript (2 vols., Amsterdam and New York).
Busby, K., et al. 1993, The Manuscripts of Chrétien de Troyes (Amsterdam).
Careri, M., et al. 2001, Album de manuscrits français du XIIIe siècle: Mise en page et
mise en texte (Rome).
Careri, M., C. Ruby and I. Short 2011, Livres et écritures en français et en occitan au
XIIe siècle: Catalogue illustré (Rome).
Cazal, Y. 1998, Les Voix du peuple, Verbum Dei: Le bilinguisme latin-langue vulgaire
au Moyen Âge (Geneva).
Clanchy, M. 2012, From Memory to Written Record: England 1066–1307, 3rd edn.
(Chichester).
Damian-Grint, P. 1999, The New Historians of the Twelfth-Century Renaissance:
Inventing Vernacular Authority (Woodbridge).
Da Rold, O., T. Kato, M. Swan and E. Treharne, The Production and Use of
English Manuscripts, 1060–1220, available at www.le.uk/ee/em1060to1220/
[2010, last update 2013].
Davies, J. R. 2003, The Book of Llandaf and the Norman Church in Wales
(Cambridge).
Dean, R. J., with M. Boulton 1999, Anglo-Norman Literature: A Guide to Texts and
Manuscripts (London).
388 Bibliography
Duncan, E. 2008, Catalogue of Latin and Vernacular Manuscripts Written by Gaelic
Scribes: AD 1000–1200 (Aberdeen).
Edel, D. 2003, ‘The status and development of the vernacular in early medieval
Ireland’, in Goyens and Verbeke.
Emmerson, R. K. 2007, ‘Visualizing the vernacular: Middle English in early
fourteenth-century bilingual and trilingual manuscript illustrations’, in
Tributes to Lucy Freeman Sandler: Studies in Illuminated Manuscripts, ed.
K. A. Smith et al. (Turnhout), 187–204.
Faulkner, M. 2012, ‘Rewriting English literary history, 1042–1215’, in Literature
Compass 9/4, 275–91.
Frank, B. 1994, Die Textgestalt als Zeichen: Lateinische Handschriftentradition und
die Verschriftlichung der romanischen Sprachen (Tübingen).
Frankis, J. 1986, ‘The social context of vernacular writing in the thirteenth century:
The evidence of the manuscripts’, in Thirteenth-Century England I, ed. P. R.
Coss and S. D. Lloyd (Woodbridge), 175–84.
Garrison, M., et al. (eds.) 2013, Spoken and Written Language: Relations
between Latin and the Vernacular Languages in the Earlier Middle Ages
(Turnhout).
Gerry, K. B. 2009, ‘The Alexis Quire and the cult of saints at St Albans’, in
Historical Research 82, 593–612.
Goyens, M. and W. Verbeke (eds.) 2003, The Dawn of the Written Vernacular in
Western Europe (Louvain).
Gransden, A. 1974, Historical Writing in England, c. 550 to c. 1307 (London).
Gunnlaugsson, G. M. 2008, ‘The origin and development of Icelandic script’, in
O. Kresten and F. Lackner (eds.), Régionalisme et internationalisme: Problèmes
de Paléographie et de Codicologie du Moyen Âge. Actes du XVe colloque du
Comité international de paléographie latine (Vienna), 87–94.
Huws, D. 2000, Medieval Welsh Manuscripts (Cardiff).
Kittay, J. and W. Godzich 1987, The Emergence of Prose: An Essay in Prosaics
(Minneapolis, MN).
Lacy, N. J. 2000, ‘The evaluation and legacy of French prose romance’, in The
Cambridge Companion to Medieval Romance, ed. R. L. Krueger (Cambridge),
167–82.
Laing, M. 1993, Catalogue of Sources for a Linguistic Atlas of Early Medieval English
(Cambridge).
Legge, M. D. 1950, Anglo-Norman in the Cloisters: The Influence of the Orders upon
Anglo-Norman Literature (Edinburgh).
1963, Anglo-Norman Literature and its Background (Oxford, 1963).
Lusignan, S. 1986, Parler vulgairement: Les intellectuels et la langue française aux
XIIIe et XIVe siècles (Paris).
Maiden, M., et al. (eds.) 2013, The Cambridge History of the Romance Languages, II:
Contexts (Cambridge).
McTurk, R. (ed.) 2008, A Companion to Old Norse: Icelandic Literature and
Culture (Oxford).
 Bibliography 389
O’Brien, B. R. 2011, Reversing Babel: Translation among the English during an Age of
Conquests, c. 800 to c. 1200 (Newark, DE).
Pelteret, D. A. E. 1990, Catalogue of English Post-Conquest Vernacular Documents
(Woodbridge).
Raible, W. 1998, ‘Die Anfänge der volkssprachlichen Schriftkultur in der Romania
oder: Die Eroberung konzeptueller Räume’, in Verschriftung und
Verschriftlichung: Aspekte des Medienwechsels in verschiedenen Kulturen und
Epochen, ed. C. Ehler and U. Schaefer (Tübingen), 156–73.
Richter, M. 1976, ‘Kommunikationsprobleme im lateinischen Mittelalter’,
Historische Zeitschrift 222, 43–80.
1979, Sprache und Gesellschaft im Mittelalter: Untersuchungen zur mündlichen
Kommunikation in England von der Mitte des elften bis zum Beginn des
vierzehnten Jahrhunderts (Stuttgart).
Rigg, A. G. 1992, A History of Anglo-Latin Literature (Cambridge).
Schon, P. M. 1960, Studien zum Stil der frühen französischen Prosa: Robert de Clari,
Geoffroy de Villehardouin, Henri de Valenciennes (Frankfurt).
Short, I. 1992, ‘Patrons and polyglots: French literature in twelfth-century
England’, in ANS 14 (1992), 229–49.
1994, ‘Gaimar’s Epilogue and Geoffrey of Monmouth’s Liber vetustissimus’,
Speculum 69, 323–43.
1996, ‘Tam Angli quam Franci: Self-definition in Anglo-Norman England’, in
ANS 18, 153–75.
2007, ‘Denis Piramus and the truth of Marie’s Lais’, Cultura Neolatina 67,
319–40.
2009, ‘Verbatim et literatim: Oral and written French in 12th-century Britain’,
Vox Romanica 68, 156–68.
2010, M. Careri, C. Ruby, ‘Les Psautiers d’Oxford et de St Albans: Liens de
parenté’, Romania 128 (2010), 29–45.
2013, Manual of Anglo-Norman, 2nd edn. (Oxford), 17–42.
(ed.) 2015, The Oxford Psalter (Bodleian MS Douce 320) (Oxford).
Smith, C. 1999, ‘The vernacular’, in The New Cambridge Medieval History, 5
(Cambridge), 71–83.
Smith, D. M. 2004, ‘Alexander (d. 1148)’, Oxford Dictionary of National Biography
(Oxford), available at www.oxforddnb.com/view/article/324.
Southern, R. W. 1970, Medieval Humanism and Other Studies (Oxford).
Stanton, R. 2002, The Culture of Translation in Anglo-Saxon England
(Cambridge).
Stock, B. 1983, The Implications of Literacy: Written Language and Models of
Interpretation in the Eleventh and Twelfth Centuries (Princeton, NJ).
Swan, M. and E. Treharne 2000, Rewriting Old English in the Twelfth Century
(Cambridge).
Taylor, A. 2003, ‘From manual to miscellany: Stages in the commercial copying of
vernacular literature in England’, Yearbook of English Studies 33, 1–17.
Toswell, M. J. 2014, The Anglo-Saxon Psalter (Turnhout).
Treharne, E. 2012, Living through Conquest: The Politics of Early English (Oxford).
390 Bibliography
and P. Pulsiano 2001, ‘An introduction to the corpus of Anglo-Saxon vernacular
literature’, in A Companion to Anglo-Saxon Literature (Oxford), 3–10, 403–14.
Turner, R. V. 1978, ‘The miles literatus in twelfth- and thirteenth-century
England: How rare a phenomenon?’, American Historical Review 83, 928–45.
Tyler, E. M. (ed.) 2011, Conceptualizing Multilingualism in England, c. 800–c. 1250
(Turnhout).
West, J. 2008, ‘A taste for the antique: Henry of Blois and the arts’, ANS 30, 213–30.
Wogan-Browne, J., N. Watson, A. Taylor and R. Evans (eds.) 1999, The Idea of the
Vernacular: An Anthology of Middle English Literary Theory, 1280–1520
(Exeter).
Wright, R. A Sociophilological Study of Late Latin (Turnhout, 2002).
Zink, M. 1981, ‘Une mutation de la conscience littéraire: Le langage romanesque à
travers des exemples français du XIIe siècle’, Cahiers de civilisation médiévale
24, 3–27.
1985, La subjectivité littéraire: Autour du siècle de saint Louis (Paris).
Zumthor, P. 1984, La poésie et la voix dans la civilisation médiévale (Paris).
1987, La lettre et la voix: De la ‘littérature’ médiévale (Paris).

18 Vernacular Manuscripts II

Primary Sources
Gottfried von Strassburg Tristan und Isold, ed. W. Haug and M. G. Scholz (2 vols.,
Berlin, 2011).
Hartmann von Aue, Der arme Heinrich, ed. H. Paul, 18th edn. (rev. K. Gärtner,
Berlin and New York, 2010).
Iwein. Text der siebenten Ausgabe von G. F. Benecke, K. Lachmann und L. Wolff,
trans. and with notes by T. Cramer, 4th edn. (Berlin, 2001).
Iwein. Handschrift B, facsimile ed. H. Matthias Heinrichs (Cologne and Graz,
1964).
Heinrich der Glîchezâre, Reinhart Fuchs. Mittelhochdeutsch/Neuhochdeutsch, ed.
K.-H. Göttert (Stuttgart, 1976).
Heinric van Veldeken, Sente Servas. Mitteniederländisch/Neuhochdeutsch, ed.
J. Goossens et al. (Münster, 2008).
Eneasroman. Mittelhochdeutsch/Neuhochdeutsch, ed. and trans. D. Kartschoke
(Stuttgart, 1986).
Hellgardt, E. (ed.), Die spätalthochdeutschen ‘Wessobrunner Predigten’ im
Überlieferungsverbund mit dem ‘Wiener Notker’: Eine neue Ausgabe (Berlin, 2014).
Herbort’s von Fritslâr liet von Troye, ed. G. K. Frommann (Quedlinburg and
Leipzig, 1837; repr. Amsterdam, 1966).
König Rother. Mittelhochdeutsch/Neuhochdeutsch, ed. P. K. Stein and I. Bennewitz
(Stuttgart, 2000).
Der deutsche ‘Lucidarius’, 1: Kritischer Text nach den Handschriften, ed.
D. Gottschall and G. Steer (Tübingen, 1994).
 Bibliography 391
Die Millstätter Predigten, ed. R. Schiewer, (Berlin and Boston, 2015).
Das Rolandslied des Pfaffen Konrad. Mittelhochdeutsch/Neuhochdeutsch, ed.
D. Kartschoke (Stuttgart, 1993).
Das St. Trudperter Hohelied. Eine Lehre der liebenden Erkenntnis, ed. F. Ohly
(Frankfurt, 1998).
Ulrich von Zatzikhoven, Lanzelet, ed. and trans. F. Kragl (2 vols., Berlin and New
York, 2006).
Wolfram von Eschenbach, Parzival, ed. E. Nellmann, transl. D. Kuhn (Frankfurt,
1994).
Willehalm, ed. J. Heinzle, with the miniatures of the Wolfenbüttel manuscript
and an essay by P. and D. Diemer (Frankfurt, 1991).

Modern Literature
2VL ‒ Die deutsche Literatur des Mittelalters. Verfasserlexikon, 2nd edn. (ed. K. Ruh
et al., 14 vols., Berlin, 1978–2008).
Beach, Manuscripts and Monastic Culture.
Bertelsmeier-Kierst, C. 2000, ‘Aufbruch in die Schriftlichkeit: Zur volkssprachli-
chen Überlieferung im 12. Jahrhundert’, Wolfram-Studien 16, 157–74.
2003, ‘Fern von Braunschweig und fern von Herzogen Heinriche? Zum A-Prolog
des “Lucidarius”’, Zeitschrift für deutsche Philologie 122, 20–47.
2003a, ‘Verortung im kulturellen Kontext: Eine andere Sicht auf die Literatur um
1200’, in Eine Epoche im Umbruch: Volkssprachliche Literalität, 1200–1300.
Cambridger Symposium 2001, ed. C. Bertelsmeier-Kierst et al. (Tübingen), 23–44.
Bumke, Mäzene.
Bumke, J. 1986, Höfische Kultur: Literatur und Gesellschaft im hohen Mittelalter
(2 vols., Munich).
1991, Courtly Culture: Literature and Society in the High Middle Ages, English
transl. of Bumke 1986 by T. Dunlap (Berkeley, CA).
1995, ‘Heinrich der Löwe und der “Lucidarius”-Prolog’, Deutsche
Vierteljahrsschrift für Literaturwissenschaft und Geistesgeschichte 69, 603–33.
1996, Die vier Fassungen der ‘Nibelungenklage’: Untersuchungen zur
Überlieferungsgeschichte und Textkritik der höfischen Epik im 13. Jahrhundert
(Berlin and New York).
Fingernagel, A. and N. Henkel (eds.) 1992, Heinrich von Veldeke: Eneas-Roman.
Vollfaksimile des Ms. germ. fol. 282 der Staatsbibliothek zu Berlin‒Preußischer
Kulturbesitz (Wiesbaden).
Gärtner, K. 1988, ‘Zu den Handschriften mit dem deutschen Kommentarteil des
Hoheliedkommentars Willirams von Ebersberg’, in Honemann and Palmer,
1–34.
Goossens, J. 1991, ‘Die Servatiusbruchstücke: Mit einer Untersuchung und
Edition der Fragmente Cgm 5249/18, 1b der Bayer. Staatsbibliothek
München’, Zeitschrift für deutsches Altertum 120, 1–65.
Grote, M. 2012, ‘Frühe deutschsprachige Sammelhandschriften’, PhD
Dissertation, Universität Paderborn (publication pending).
392 Bibliography
Grubmüller, K. 2000, ‘Die Vorauer Handschrift und ihr Alexander. Die kodiko-
logischen Befunde: Bestandsaufnahme und Kritik’, in Alexanderdichtungen
im Mittelalter, ed. J. Cölln et al. (Göttingen), 208–21.
Gutfleisch-Ziche, B. 1997, Volkssprachliches und bildliches Erzählen biblischer
Stoffe: Die illustrierten Handschriften der ‘Altdeutschen Genesis’ und des
‘Leben Jesu’ der Frau Ava (Frankfurt).
Heinzer, F. (ed.) 2008: Klosterreform und mittelalterliche Buchkultur im deutschen
Südwesten (Leiden).
Heinzle, J. (ed.) 2011, Wolfram von Eschenbach: Ein Handbuch (2 vols., Berlin and
Boston).
Hellgardt, E. 1988, ‘Die deutschsprachigen Handschriften im 11. und 12.
Jahrhundert: Bestand und Charakteristik im chronologischen Aufriß’, in
Honemann and Palmer, 35–81.
1992, ‘Lateinisch-deutsche Textensembles in Handschriften des 12.
Jahrhunderts’, in Latein und Volkssprache im deutschen Mittelalter,
1100–1500, ed. N. Henkel and N. F. Palmer (Tübingen), 19–31.
Holznagel, F.-J. 1995, Wege in die Schriftlichkeit: Untersuchungen und Materialien
zur Überlieferung der mittelhochdeutschen Lyrik (Tübingen and Basel).
Honemann, V. and N. F. Palmer (eds.) 1988: Deutsche Handschriften 1100–1400:
Oxforder Kolloquium 1985 (Tübingen).
Kartschoke, D. 1989, ‘“In die latîne bedwungin”: Kommunikationsprobleme im
Mittelalter und die Übersetzung der Chanson de Roland durch den Pfaffen
Konrad’, Beiträge zur Geschichte der deutschen Sprache und Literatur 111,
196–209.
Klein, K. 2000, ‘Französische Mode? Dreispaltige Handschriften des deutschen
Mittelalters’, in Scrinium Berolinense: Tilo Brandis zum 65. Geburtstag, ed.
P. J. Becker et al. (2 vols., Berlin), 1. 180–201.
Klein, T. 1988, ‘Ermittlung, Darstellung und Deutung von Verbreitungstypen in
der Handschriftenüberlieferung mittelhochdeutscher Epik’, in Honemann
and Palmer, 110–67.
Kuhn, H. 1968, ‘Aspekte des 13. Jahrhunderts in der deutschen Literatur
(Akademievortrag)’; repr. H. Kuhn, Entwürfe zu einer Literatursystematik
des Spätmittelalters (Tübingen, 1980), 1–18.
McLelland, N. 2000, Ulrich von Zatzikhoven’s ‘Lanzelet’: Narrative Style and
Entertaintment (Woodbridge).
Müller, S. 2005, ‘“Erec” und “Iwein” in Bild und Schrift. Entwurf einer medie-
nanthropologischen Überlieferungs- und Textgeschichte am Beispiel der
frühesten Zeugnisse der Artusepen Hartmanns von Aue’, Beiträge zur
Geschichte der deutschen Sprache und Literatur 128, 414–35.
Okken, L. 1974, Hartmann von Aue: ‘Iwein’. Ausgewählte Abbildungen und
Materialien zur handschriftlichen Überlieferung (Göppingen).
Palmer, N. F. 1993, German Literary Culture in the Twelfth and Thirteenth
Centuries: An Inaugural Lecture Delivered before the University of Oxford on
4 March 1993 (Oxford).
 Bibliography 393
2005, ‘The High and Later Middle Ages (1100–1450)’, in The Cambridge History
of German Literature, ed. H. Watanabe-O’Kelly, 2nd edn. (Cambridge),
40–91, 521–37.
2005a, ‘Manuscripts for reading: The material evidence for the use of manu-
scripts containing Middle High German narrative verse’, In Orality and
Literacy in the Middle Ages: Essays on a Conjunction and Its Consequence in
Honour of D. H. Green, ed. M. Chinca and C. Young (Turnhout), 67–102.
2010, ‘The Houghton Library Lanzelet fragment’, Harvard Library Bulletin 21,
53–72.
2011, ‘A fragment of “König Rother” in the Charles E. Young Research Library
in Los Angeles’, in: Mittelhochdeutsch: Beiträge zur Überlieferung, Sprache und
Literatur. Festschrift für Kurt Gärtner zum 75. Geburtstag, ed. R. Plate and
M. Schubert (Berlin), pp. 22–41.
2017, ‘Manuscripts of the earliest Middle High German prayers, 1150–1250’, in
Vernacular Manuscript Culture, ed. E. Kwakkel (Leiden).
Petzet, E. and O. Glauning 1910–30, Deutsche Schrifttafeln des IX. bis XVI.
Jahrhunderts aus Handschriften der K. Hof- und Staatsbibliothek in München
(5 parts, Munich etc.).
Putzo, C. 2009, ‘Die Frauenfelder Fragmente von Konrad Flecks “Flore und
Blanscheflur”. Zugleich ein Beitrag zur alemannischen
Handschriftenüberlieferung des 13. Jahrhunderts’, Zeitschrift für deutsches
Altertum 138, 312–43.
2011, ‘Mehrsprachigkeit im europäischen Kontext. Zu einem vernachlässigten
Forschungsfeld interdisziplinärer Mediävistik’, in Mehrsprachigkeit im
Mittelalter: Kulturelle, literarische, sprachliche und didaktische Konstellationen
in europäischer Perspektive, ed. M. Baldzuhn and C. Putzo (Berlin and New
York), 3–34.
Scheepsma, W. 2013, ‘Maastricht’, in Schreiborte des deutschen Mittelalters:
Skriptorien ‒ Werke ‒ Mäzene ed. M. Schubert (Berlin and Boston), 307–28.
Schneider, K. 1987, Gotische Schriften in deutscher Sprache, I: Vom späten 12.
Jahrhundert bis um 1300 (2 vols., Wiesbaden).
Schröder, E. 1909, ‘Zur Überlieferung des Herbort von Fritzlar’, Göttinger Gelehrte
Anzeigen, phil.-hist. Klasse, 92–102.
Steer, G. 1990, ‘Der deutsche Lucidarius: Ein Auftragswerk Heinrichs des
Löwen?’, Deutsche Vierteljahrsschrift für Literaturwissenschaft und
Geistesgeschichte 64, 1–25.
Wetzel, R. 1992, Die handschriftliche Überlieferung des ‘Tristan’ Gottfrieds von
Straßburg untersucht an ihren Fragmenten (Fribourg).
Wiesinger, P. 1978, ‘Ein Fragment von Hartmanns “Iwein” aus Kremsmünster’,
Zeitschrift für deutsches Altertum 107, 193–203.
1984, ‘Nachträge zum wiedergefundenen Kremsmünsterer Fragment von
Hartmanns “Iwein”’, Zeitschrift für deutsches Altertum 113, 239–41.
Wolf, J. 2008, Buch und Text. Literatur- und kulturhistorische Untersuchungen zur
volkssprachigen Schriftlichkeit im 12. und 13. Jahrhundert (Tübingen).
 Index of Manuscripts

Aberystwyth, National Libr. of Wales 17110B II 2425 41

(‘Book of Llandaff ’) 21, 314 8486–91 266
Peniarth 540 314
Admont, Stiftsbibl. 434 41 Cambrai, Médiathèque mun. 168 274
742 235 Cambridge
Angers, Bibl. mun. 304 (295) 323 Corpus Christi Coll. 2 (‘Bury Bible’) 10, 13, 49,
Assisi, Bibl. del sacro conv. 573 222–3, 232, 236 Fig. 3.3, 57, 83
Avesnes, Société Archéologique, s. n. 49 3–4 (‘Dover Bible’) 46, 49
Avranches, Bibl. mun. 72 58 Fitzwilliam Museum 24 13
91 41 Maclean 165 232
128 73 Gonville & Caius Coll. 2/2 24
229 234 3/3 24
6/6 24
Bamberg, Staatsbibl. Class. 10 234 7/7 24
Class. 15 220, 232 10/10 24
Class. 21 248 12/128 24
Patr. 5 11, 19, 74 14/130 24
Barcelona, Archivo de la Corona de Aragón, 15/131 24
Ripoll 78 310 16/132 24
Basel, Universitätsbibl. N I 2, 83 323 17/133 24
Beirut, Université de St Joseph 223 275 18/134 24
Berlin, Staatsbibl. germ. fol. 282 63 19/135 24
lat. fol. 74 263 123/60 320
lat. fol. 252 245 427/427 36
lat. fol. 272 302 456/394 275
lat. fol. 273 301–2 Jesus Coll. Q. D. 2 (44) 266, 321
lat. qu. 198 283–6 Pembroke Coll. 59 169
Phillipps 1925 322 113 320
Bern, Burgerbibl. 79 322 Peterhouse 229 23
120 63 St John’s Coll. B. 13 (35) 119
196 273 G. 15 (183) 13
332 232 Trinity Coll. B. 5. 4 (150) 201, Fig. 11.3,
337 291 208, 213
702 232 B. 11. 10 (249) Fig.10.2
Bethesda (Maryland), National Libr. of Medicine B. 14. 33 (317) 170
E. 78 288 O. 3. 55 (1227) Fig. 10.1
Bologna, Bibl. Universitaria 2208–9 161, O. 7. 9 (1337) 234
164–5, 171 O. 7. 13 (1341) 13
4228 219, 234 R. 15. 16 (940) 269, Fig. 14.1, 273
Brussels, Bibl. royale II. 139/3 323 R. 17. 1 (987) (‘Eadwine Psalter’) 71, 315,
II 181 frag. 3 320 319, 322

394
 Index of Manuscripts 395
University Library Add. 4166 frag. 9 321 S. Marco 114 233
Ii. 1. 33 326 S. Marco 124 233
Ii. 3. 33 41 S. Marco 125 232
Ii. 6. 20 244 S. Marco 130 233
Ii. 6. 32 314 S. Marco 166 231
Charleville, Bibl. mun. 187 219, 235 Bibl. Naz. II. I. 7 161, 169
Chartres, Bibl. mun. 92 234 II. III. 214 267
190 234 Bibl. Riccardiana Ricc. 126 235
497–8 219, 231, 245, 267 Frankfurt, Stadt- und Universitätsbibl. Barth. 42
Cologne, Dombibl. 59 46 57, 83
Cologny-Genève, Bibl. Bodmeriana 9 234 Freiburg, Augustinermuseum, G 23/1a 54
11 320
17 320 Geneva, Bibl. de Genève Com. Lat. 183
127 48 319
Copenhagen, Konigl. Bibl., Gamle Kgl. Samling Ghent, Universiteitsbibl. 92 62
1653 291 Gießen, Universitätsbibl. Hs. 89 Fig. 18.4,
Universtetsbibl. AM 618 4° 319 338–9
Córdoba, Bibl. de la Catedral 153 233 Göttingen, Universitätsbibl., Apparat. Diplom.
Cracow Cath. 208 46 10E. Mappe IV.1 233
Graz, Universitãtsbibl. 1703/137 41
Damascus, Zahiriyya 4871 266
Darmstadt, Landesbibl. 1410 267 The Hague, Koninklijke Bibl. 73 J 6 150, 261
2282 217, Fig. 12.1, 231 73 J 7 24, 150
Dijon, Bibl. mun. (‘Bible of Stephen Harding’) 76 E 15 41
12–15 83 Hereford Cathedral, Dean & Chapter Archives
114 41 990 22
Dublin, Royal Irish Academy 1229 (23 E 25) 314 Library O. III. 2 73
Trinity Coll. 494 223 O. III. 15 325
1339 314 O. V. 10 82
Durham Cathedral A. II. 1 (‘Puiset Bible’) 20–1 O. V. 14 82
A. II. 4 (‘Carilef Bible’) 83 O. VI. 10 82
A. II. 11 233 O. VIII. 6 233
A. IV. 34 20, 23 P. I. 12 82
C. IV. 27 320 P. II. 14 82
P. III. 7 82
El Escorial, Real Bibl. E. IV. 24 239 P. V. 3 82
Epinal, Bibl. mun. 58 317, 322 P. V. 4 82
Erfurt, Universitätsbibl. Dep. Erf. Cod. Amplon. Heidelberg, Universitätsbibl. Cod. Pal. germ. 112
8° 5 237 332, Fig. 18.3
Cod. Amplon. 8° 32 320 Hildesheim, Dombibl. St. Godehard 1 61, 319, 321
Cod. Amplon. 8° 66 233 Dom-Museum DS 37 61
Cod. Amplon. 4° 365 275
Erlangen, Universitätsbibl. 191 231 Istanbul, Ayasofya 4832 266
Eton Coll. 161 266
204 291 Jena, Universitätsbibl. Bos. 9. 6 63
Jerusalem, Nat. Libr. of Israel Heb. 4°5827 161,
Florence, Bibl. Laurenziana, Ashburnham 1459 165, Fig. 9.2, 167
(1382) 234
Plut. 11. 9 234 Klosterneuburg, Stiftsbibl. 671 234
Plut. 45. 2 251 1098 231
Plut. 64. 35 323 Kues, St. Nikolaus-Hospital 190 233
Plut. 71. 21 233
Plut. 73. 16 292 Laon, Bibl. mun. 224 (455) 322
S. Marco 102 233 435 ter 235
S. Marco 113 233 Le Havre, Bibl. mun. 332 118
396 Index of Manuscripts
Leiden, Universiteitsbibl. BPG 49a 168 Harl. 2798–9 13
BPL 5 155 Harl. 2803–4 13
BPL 20 41 Harl. 3509 248
BPL 30 149 Harl. 3534 23
BPL 35 155 Harl. 3859 314
BPL 38D 257 Harl. 4070 325–6
BPL 43 155 Harl. 4388 317, 321
BPL 64 143–5 Harl. 4733 320
BPL 91 146, 148 Harl. 5786 264
BPL 92 146, 148 Royal 3 A. xii 82
BPL 92A 141–2, 155 Royal 4 D. vii 23
BPL 144 142–3, Fig. 8.1 Royal 7 D. xxv 223
BPL 189 145–6, Fig. 8.2, 254 Royal 7 F. vi 82
BPL 196 Fig. 2.1, 41 Royal 10 C. iv 23
BPL 1048 155 Royal 12 C. i 42
BPL 1925 232 Royal App. 85 275
BUR Q 1 248-51, Fig. 13.1, 250–1 Sloane 1975 291
d’Ablaing 1 Fig. 16. 2 Sloane 2030 276
SCA Hebr. 8 (Or. 4725) 168 Lambeth Palace 3 (‘Lambeth Bible’) 49, 60, 83
VLF 39 148 73 322
VLF 85 291 339 217, 231
VLQ 12 41 Society of Antiquaries 716 320
VLQ 42 155 National Archives C 146/10018 319
VLQ 51 Fig 2.4, 41 private owner, s. n. 232
VLQ 103 248 Sotheby & Co., Valmadonna Trust Libr. 1
VUL 46 41 (formerly Sassoon Coll. 282) 161, 165,
Liège, Bibl. universitaire 77 100 167, 172
Lincoln Cath. 199 320 Wellcome Libr. 4 266
London Los Angeles, J. Paul Getty Museum 64 62
BL Add. 1695 61 Lucca, Bibl. Statale 1405 252
Add. 11639 173 Lunel, Bibl. mun. 6 237-8
Add. 17737–8 (‘Floreffe Bible’) 13, 46, Luxembourg, Bibl. nat. 138 244
Fig. 3.2, 60 770 273
Add. 18342 232
Add. 22719 261, 271 Maidstone, Kent County Archives Fa Z 1 319
Add. 28106–7 13 Museum P. 5 49, 83
Add. 46847 (‘Sherborne Cartulary’) 21 Milan, Bibl. Ambrosiana I. 195 235
Add. 49366 317, 319 M. 62 sup. 231
Add. 89000 19 M. 63 sup. 226, 237
Arundel 230 319–20 O. 53 sup. 296
Arundel 268 274 O. 55 sup. 296
Arundel 348 233, 255 Bibl. Capitolare della Basilica Ambrosiana M
Arundel Or. 2 161, 165, 168 2 237
Arundel Or. 51 161, 169 Monte Cassino, Archivio della Badia 174 96
Burney 161 41 225 261
Cotton Faustina A. X 101 Montpellier, Bibl. Universitaire, Section de
Cotton Nero A. V 321 Médicine H. 82 302
Cotton Nero C. IV 319 Munich, BSB
Cotton Titus D. XXIV 322 Cgm 17 340
Cotton Vitellius E. IX 319 Cgm 77 340
Cotton App. 56 320–1 Cgm 191 339
Egerton 3661 41 Cgm 5248/7 37, 40
Harl. 270 320 Clm 331 233
Harl. 1585 291 Clm 4622 291
Harl. 2713 233 Clm 4660 (‘Carmina Burana’) 99
 Index of Manuscripts 397
Clm 13002 Fig. 3.4 Douce 320 317, Fig. 17.1, 319, 322
Clm 14458 223 Douce 381 326
Clm 14503 232 French d. 16 320
Clm 14779 223 Hatton 92 101
Clm 16123 235 Lat. liturg. f. 1 13
Clm 19475 100 Lat. th. d. 20 41
Clm 19490 100 Laud. lat. 67 223
Clm 22009 Fig. 4.1 Laud. misc. 91 325
Clm 22292 235 Laud. or. 168 161, 163
Clm 29384 235 Opp. 627 161
Clm 29246 232 Opp. 717 161
Clm 29384 232 Or. 6 161
Clm 30055 46 Or. 621 172
Cod. gall. 29 323 Rawl. B. 502 314
Rawl. B. 503 314
Nantes, Musée Dobrée V 317, 322 Rawl. C. 641 320
Naples, Bibl. Naz. Vindob. 47 (Martini 7) 232 Rawl. D. 913 320
New York, Jewish Theological Seminary 8092 161 Rawl. Q. f. 8 23
Pierpont Morgan Libr. M. 338 322 Savile 15 275
M. 736 49 Selden supra 24 272–3
Corpus Christi Coll. 6 161, 169
Orléans, Bibl. mun. 80 233 57 244
265 217, 231 133 161, 169
266 223–4, 228, 237 165 161
269 233 Jesus Coll. 26 82, 326
283 291 52 82
285 291 53 82
286 291 62 82
301 291 67 82
Oslo, Martin Schøyen Coll. 206 167 68 82
Oxford 70 82
Balliol Coll. 36 206, 213 Merton Coll. 291 257
Bodl. Libr. Auct. D. 1. 13 208 St John’s Coll. HB4/4. a. 4. 21 (I. subt. 1. 47) 319
Auct. E. inf. 1–2 46 49 102
Auct. E. inf. 6 13 Trinity Coll. 47 232
Auct. E. inf. 7 Fig. 11.2 65 121
Auct. F. 1. 9 268–9, 275 University Coll. 165 51
Auct. F. 2. 13 63 191 119
Auct. F. 6. 3 290
Barlow 40 252, Fig. 13.2 Padua, Bibl. Antoniana 550 22
Bodley 444 Fig. 11.4 553 231
Bodley 614 64 Bibl. Universitaria 1688 232
Bodley 672 13 2087 238
Bodley 717 Fig. 3.1, 48 Palermo, Bibl. com. 4 Qq. A.10 267
Canon. Class. lat. 41 41 Paris
Canon. Pat. lat. 148 41 Alliance Israélite Universelle 147 163
E. D. Clarke 35 Fig. 11.1 Archives nationales AB XIX 1734 (dossier Orne
Digby 13 316, 321 n. 1) 319
Digby 23 317, 319, 322, 326 138 (‘Mortuary Roll of Vitalis, abbot of
Digby 40 275 Savigny’) 69
Digby 51 271–2 Bibl. de l’Arsenal 256 94, 96
Digby 53 326 711 246
Digby 56 36 910 223, 238
Digby 79 286 939 102
Digby 159 275 1162 212
398 Index of Manuscripts
Paris (cont.) nouv. acq. lat. 873 317, 321
Bibl. Maz. 54 (70) 319, 321 nouv. acq. lat. 1628 291
BnF fr. 900 322 nouv. acq. lat. 1670 319
fr. 24766 321 nouv. acq. lat. 1761 310
fr. 24768 322 Bibl. de la Sorbonne 632 323
hébr. 113 172 Victor Klagsbald Collection (formerly Sassoon
hébr. 326 163 Coll. 535) 161, 169
hébr. 633 173 Parma, Bibl. Palatina 2574 163
lat. 258 136 Pesaro, Bibl. Oliveriana 26 301
lat. 768 319 Pistoia, Bibl. Cap. C. 77 233
lat. 943 100 C. 106 300–1, 309
lat. 1835 13 Pommersfelden, Schlossbibl. 16/27664 232, 236
lat. 1954 233 Prague, Metropolitan Libr. A. 21. 1 57
lat. 2171 136 L. 54 233
lat. 2770 13
lat. 2858 233 Reims, Bibl. mun. 23 242
lat. 3237 224 Rome, BAV
lat. 3853 13 Borgh. 131 231
lat. 3884 310 Ottob. lat. 1474 314
lat. 4516 301 Ottob. lat. 1974 232
lat. 4613 295 Pal. lat. 772 309
lat. 5764 244, 257 Pal. lat. 1371 262
lat. 5783 257 Pal. lat. 1547 247
lat. 6398 234 Pal. lat. 1685 245
lat. 6400C 232–3 Pal. lat. 1971 320
lat. 6400D 232–3 Reg. lat. 230 223, 238
lat. 6400E 232, 234 Reg. lat. 1244 321
lat. 6400F 220, 232–3 Reg. lat. 1285 264
lat. 6400G 232 Reg. lat. 1649 232
lat. 7102 286 Rossi 537 234
lat. 7246 102 Rossi 579 265–6
lat. 7373 262 Vat. ebr. 109 163, 169
lat. 7412 273 Vat. ebr. 113 163, 169
lat. 7493 238 Vat. ebr. 468 161, Fig 9.1, 164–5, 173
lat. 8216 23 Vat. ebr. 482 161, 164–5, 167–8, 173
lat. 8846 319 Vat. lat. 566 233
lat. 9335 265–7 Vat. lat. 567 235
lat. 9656 293–4, 297, Fig. 16.1, 299 Vat. lat. 1406 302
lat. 10062 117 Vat. lat. 1722 232
lat. 11248 273 Vat. lat. 2978 231
lat. 11575–6 13, 49 Vat. lat. 2982 231
lat. 12959 232 Vat. lat. 5334 319, 323
lat. 13368 223–4, 237 Vat. lat. 10281 277, 281, 290
lat. 13951 275 Rome, San Paolo fuori le Mura, s.n. 60
lat. 14314 13 Rouen, Bibl. mun. 1 Fig. 7.1–2
lat. 14470 317, 323 8 58
lat. 14673 121 31 42
lat. 14793 88 57 41
lat. 15104 233 477 41
lat. 15406 97 932 234
lat. 16743–6 49 1406 41
lat. 17813 222, 226, Fig. 12.2, 238 1409 41
nouv. acq. fr. 4503 317, 319
nouv. acq. fr. 18217 323 San Marino (Cal.), Huntington Libr. HM 62
nouv. acq. lat. 214 41 (‘Gundulf Bible’) 316, 326
 Index of Manuscripts 399
Sankt Gallen, Stiftsbibl. 69 98 Verdun, Bibl. mun. 72 323
134 223 Vienna, Österreichisches Nationalbibl. 93 292
715 310 95 244
831 234 471 299
904 155 1568 42
Sarnen, Bibliothek des Benediktinerkollegiums 2499 239
Cod. membr. 69 Fig. 18.2 2686 Fig. 18.1
Soest, Stadtbibl. 24 276 2721 340
Soissons, Bibl. mun. 9 41 theol. gr. 336 54
Stockholm, Kungliga Bibl. A. 144 57 Vollbehr, Dr. O. (Washington DC), olim,
A. 148 279 present whereabouts unknown 233
Strasbourg, Bibl. du Grand Séminaire 37 48
Stuttgart, Württembergische Landesbibl., HB II Winchester Cath. 1 51, 83
24 61 Wolfenbüttel, Herzog August Bibl. Guelf. 34. 6
Aug. 2° 320
Tournai, Bibl. mun. 74bis 233 Guelf. 61. 2 Aug. 8° 54
Bibl. du Séminaire 1 60 Guelf. 105 noviss. 2° (‘Gospels of Henry the
Tours, Bibl. mun 85 96 Lion’) 46, 48, 61
676 231 Worcester Cath. F. 40 292
903 323 Q. 12 236
927 22 Q. 44 20
Trier, Bistumsarchiv. 6 232, 236 Q. 60 292
Troyes, Médiathèque 900 97 Wrocław, Stadtbibl. 1302 292

Utrecht, Bibliotheek der Rijksuniversiteit 32 61 Yale University, Beinecke Libr. 590 321
Marston 45 247
Valenciennes, Bibl. mun. 501 65 York Minster XVI. M. 6 248, 252, 254
502 65 XVI. M. 7 252, 254
Venice, Bibl. Naz. Marciana lat. Z 273 (1574) 231
Vercelli, Bibl. Capitolare CXCIX (35) 235 Zürich, Zentralbibl. cod. C 58 340
 General Index

Ælfric 326 ‘Alī ibn al-‘Abbās 261, 263, 279, 288

‘Abd al-Masīh of Winchester 274 Alsace 331
Aberdeenshire˙ 314 Amalakite, the 60
Abingdon, Abbey of 70, 105, 110, 113 Amalarius of Metz 104, 190
And see Faricius Amandus, St 65
Abraham Ibn Ezra 170, 272 Amazons 88
Abraham ben Moshe 173 Ambrose, St 205, 242
Abu’l-Qāsim al-Zahrāwī 265, 289 Ambrosius, a judge 297
Abū Ma‘shar 264 Amiens 57
Adam of Balsham/Petit-Pont 217, 226, 228 Anaxagoras 268
Adam of Ross 326 Anchin, Abbey of 82
Adelaide, Princess 132 Andreas Frisner 201
Adelard of Bath 92, 101, 262, 266, 268, 274, 276 Andrew of St-Victor 95, 170
Admont, Abbey of 76, 80, 329, 341–2 Angers 161, 323
Adelheit, nun of 76 Angier, Frère 317, 321
Ado of Vienne 149 Anjou 322
Adolph Rusch 200 Anselm of Bec/Canterbury 62, 91, 100, 110, 123,
Aesop 241 131–3, 138, 213
Agnes of Loon 331, 338 Anselm of Laon 15, 93–5, 193–6, 200, 203,
Agnes of Poitou 131, 138 212
Agnes, wife of Count Otto of Scheyern 343 Anselm of Lucca 305–6
agrimensores 260 Antidotarium magnum Fig. 15.1, 282–3, 287
Ailred of Rievaulx 110 Antioch 263, 288
Aimeric of Angoulême 241–3 Anton Koberger 214
al-Hakim, Caliph 91 Apollonius, Ps. 275
al-Khwarizmi 92, 267–8 Apuleius 216, 219, 230, 234–5
al-Qabīsī 264 Apuleius, Ps. Fig. 15.1, 282, 285, 289
al-Razi ˙289 Aquitaine 185
Albéric de Besançon 323, 326 ‘arabesque’ initials 55–7, 74
Alcuin 216, 230, 234 Arabic 1, 92, 140, 167, 259, 261–7, 269, Fig. 14.1,
Aldobrandino of Siena 290 271–2, 274–6, 278–9, 282, 289, 291
Alexander III, Pope 94, 308 Aramaic 164–5, 167, 173
Alexander, abt. of Jumièges 100 Arator 242
Alexander of Hales 204, 213 Aratus 64
Alexander Neckam 243 Argenteuil 100
Alexander of Tralles 280, Fig. 15.1, 282 ‘Arīb ibn Sa’d 284
Alexander de Villedieu 155 Aristotle 89–92, 98, 100, 144, 215–17, 219–20,
‘Alexis Master’ 49 228–31, 268, 272, 276, 294
Alfanus, abp. of Salerno 91, 261, 286 Arnstein, Abbey of 341
Alfonse of Poitiers 164 Arrouaise, Abbey of 106, 119–20
Alfred of Shareshill 276 Arrouasians 106, 120

400
 General Index 401
Arthurian romance 335, 338–9 Bibles 13, 79, 176–7, 192–3, 203, 209, 211–13,
articella 91, 277–8, Fig. 15.1, 283, 286, 288 242–3, 262
Ashkenaz, Ashkenazi 159, 163, 168–9, 171, 174 ‘Giant’ 13–14, 58–60, 74, 128
Augsburg 79 Glossed books 5, 14, 20, 41, 59, 81, 93, 108, 110,
Augustine, St 62, 88, 90, 93–4, 96, 110, 131, 145, 193–4, Fig. 11.1, 196–7, Fig. 11.2, 200–1,
194, 205–6, 213, 219, 235, 242, 294 Fig. 11.3, 203, 207, 211–12
Augustine, Ps. 216, 219, 230, 235 And see Glossa Ordinaria
Augustinians 72, 74, 76, 95, 106, 114, 118, 133, 135, Hebrew 161, Fig. 9.1, 163–5, 167–8, 170–3
207, 329 binding 18–21
Austria 33, 75–6, 80, 117, 329 Black Death 5
Avianus 97–8, 241 Black Forest 181, 329
Avicenna (Ibn Sina) 92, 272, 276, 289 Boethius 88–9, 98, 141–3, 152, 200, 215–17, 219–22,
Avicenna, Ps. 272 229–30, 235–6, 241–2, 248, 254, 260, 262,
Avranches 92 267, 269
Azo, Master 239 Boethius, Ps. 219
Bologna 62, 71, 171, 308, 310
Babel, Tower of 284–5 San Salvatore, Abbey of 164, 171
Babylonians 208 Bonaparte, Napoleon 172
Baghdad 266 Boquen, Abbey of 107
Bamberg, St Michael’s Abbey (Michelsberg) 74, Bordesley Abbey 118
82, 105 Braunschweig 332, 335
And see Ruotger, Wulfram Brendan 320, 326
Bartholomeus Fig. 15.1 Bridlington Priory 118
Bath Abbey 271 Britain 10, 117, 312–14, 316, 319, 322–3
Baudouin, Master 100 Brittany, Breton 313
Baudry of Bourgeuil 89 Brutus 320
Bavaria 46, 64, 75, 89, 331–2, 338 Buildwas Abbey 93
Bayeux 74 Burchard of Worms 93, 304–7, 310
And see Philip of Harcourt Burgundio of Pisa 262, 272
Beatrix of Savoy 290 Bury St Edmunds 161
Beatus of Liébana 61 Abbey 10, 49, 119, 161, 244
Beauvais Cathedral 88–9, 99 Byzantine art 53–5
St-Quentin 92
St-Vaast 89 Cairo 169
And see Fulco, Guy, Ivo, Roscelin Calcidius 58, 141, 143–4, 241
Bec, Abbey of 39, 89, 91–2, 99, 132 Cambridge 2
Bede 51, 64, 110, 194, 206, 242, 314 Canterbury 61, 70, 168, 315
Bede, Ps. 285–6 abps. of: see Anselm, Lanfranc, Stephen
Belgium 117 Langton, Thomas Becket
Benedeit 320 Christ Church Cathedral Priory 35, 39, 41, 46,
Benedict, St 135–6, 283 70, 81, 84, 108–9, 112–13, 119, 319
And see Rule of script 321
Benediktbeuern, Abbey of 89, 99 And see Eadwine
Beneit of St Albans 320–1 St Augustine’s Abbey 81, 84, 95, 267
Beneventan script 25, 40, 80, 84, 261, 280, And see Salomon
283, 288 Capitula Angilrami 88
Benoît de St-Maure 323 capitularies 295–7
Bernard of Chartres 90–1 Carmina burana 99
Bernard of Clairvaux 2, 58–9, 96, 98, 110, 181, 192, Caroline minuscule script 25–6, 28–9, 31–2, 36,
317, 322 40, 81, 84, 261, 269, 288, 297
Bernard of Pavia 308 Carolingians 1, 3, 20, 25, 46, 56, 58–61, 63–5, 70,
Bernard Silvester 13, 58, 64 73, 75, 89, 103, 106, 114, 116, 140, 148–9,
Bernard of Utrecht 89 152–4, 156, 160, 175, 190, 240, 296
Besançon 89, 100, 289 Carthusians 46, 106, 135–6, 181, 329
bestiary 63 Cassiodorus 13, 154, 212
402 General Index
Castilian 314 Damascus 208
Catalan 314 Daniel of Morley 265, 272
Catiline 252, Fig. 13.2 Dante 311
Cato 97, 251–2, Fig. 13.2 David, King 60–1, 207
Celestine II: see Guido Deidemia to Achilles 89
Celtic (regions, languages) 10, 80, 311–14, 323 Denis the Areopagite 90
Champagne 322 Denis Piramus, monk of Bury 313
Chanson d’Alexis 317, 319, 321, 323 Desiderius, abt. of Monte Cassino 150, 280
Chanson de Roland 317, 319–20, 322, 326 Diemut, inmate of Wessobrunn 72, 75–6,
Charlemagne 295–7 Fig. 4.1, 79
Chartres Cathedral 90–2, 155, 200, 256, 260, 268 Diessen, duke of 331
And see Bernard, Fulbert, Ivo, John of Dioscorides 263
Salisbury, Thierry Dioscorides alphabeticus Fig. 15.1, 282–3, 287
Chelles, Nunnery of 89 Disticha Catonis 241
Chester 13 Doeg the Edomite 288
Chrétien de Troyes 312, 322 Dominicans 181, 209
Christ in Majesty 21, 61, 285 Dominicus Gundissalinus 92, 264, 272
Cicero 41, 64, 88, 90–1, 98, 141, 145–6, 216, 219, Donatus 142, 155, 241–2
230, 241–2, 246–8, 251–2, Fig. 13.2, 254 Dover Priory 46
Cicero, Ps. 141, 242, 247–8, 252 Durham Cathedral Priory 21, 46, 51, 74, 79, 81,
Circa instans 287, 290 179, 266, 277, 283, 290, 317
Cirencester Abbey 71
Cistercians 13, 16, 54, 56, 58, 93, 106–8, 118–20, Eadwine, monk of Canterbury 70–1
180–2, 247, 291, 317, 322, 329 Ebersberg, Abbey of 329, Fig. 18.1
Città di Castello 96 And see Williram
Claudian 97, 242, 247 Echternach, Abbey of 260
Clemence of Barking 319–20 Egmond, Abbey of 100, Fig. 13.1, 249, 340
Cleopatra, Ps. 280, Fig. 15.1, 281 Eilhart von Oberge 332, 334
Cluny, Cluniac 9, 92, 105–6, 112, 120, 212 Einsiedeln, Abbey of 329, 340
Cologne 72, 95, 159 Eleazar ben Judah of Worms 171, 173
St Gereon 54 Elias, monk of Corbie 57
Commentarius Cantabrigiensis 93 Elie of Winchester 321
Conrad II, K. 296 Elijah, Ascension of 60
Conrad of Hirsau 89, 242–3 Ely, Abbey of 21, 119
Constance of Le Ronceray 89 Engelberg, Abbey of 98
Constantine the African 91, 150, 261–2, 268, 271, England, English 2, 4, 11, 19–21, 33–4, 39–42, 53,
274, 278–9, Fig. 15.1, 282–4, 55–60, 69, 71, 73–4, 79–82, 92, 95, 101,
287–8 105–6, 114, 118, 120, 159, 161, 165, Fig. 9.2,
Constantinople 53, 268 169–70, 172, 184, 248, 265, 268, 272–3,
Copho 287 284, 286, 291–2, 306, 312, 315, 324,
Corbie, Abbey of 57, 149 336, 343
monks of, see Elias, John, Ingelrannus, Nevelo Anglo-Saxons 51, 53, 56–7, 63, 70, 271, 294,
Cornish 314 312–13, 315
councils: Channel 70
Lateran IV 204, 324 Norman Conquest of 39, 51, 294, 312–13,
Paris 116 315, 324
Sens 96 Ennius 241
Coventry, Franciscans of 269 Ermengarde 76
Cremona 265 Escaladieu, Abbey of 107, Fig. 6.1
Santa Lucia 265 Etienne de Fougères 323
Crucifixion 61–2, 285 Euclid 91–2, 260, 262–3, 267–8
Crusader States 263, 288 Eusebius 105, 149
Crusades 53, 159 Eutropius 90
Cuthbert, St 179 Everwin 57, 74
 General Index 403
Evesham Abbey 115 Gerard of Cremona 92, 264–6, 272, 276, 288–9
And see Roger Norreis, Thomas of Gerbert of Aurillac 91, 260
Marlborough. Gerhoch of Reichersberg 41
Évrard de Béthune 155 Germanicus 264
Exeter Cathedral 74, 79, 84, 315 Germany, German 2, 4, 19, 21, 33–5, 39–41,
St-Nicolas Priory 271 Fig. 3.4, 53, 64, 66, 74–6, 80–2, 98–100,
schools 115 102, 105, 114, 117, 159–61, 165, 167, 171, 181,
Eynsham Abbey 114 284, 293–4, 305, 311, 314, 323, 327–9, 332,
334–44
Faricius, abt. of Abingdon 70 Gilbert of Auxerre 193
Fécamp, Abbey of 41, 128, 136–7 Gilbert fitz Baderon 313
Felix, a painter 49 Gilbert of Poitiers (de la Porrée) 98, 200–1,
Firmicus Maternus 267, 276 206–7, 211–12
Flanders 33, 58 Giles, St 341
Fleury, Abbey of 118, 283, 285, 291–2 Glossa ordinaria 15–16, 168, 193–6, Fig. 11.1, 197,
Floreffe, Abbey of 46 Fig. 11.2, 200–1, Fig. 11.3, 204–7, 209,
Florilegium Angelicum 246 212–13
Florilegium Gallicum 246 Gloucester Abbey 80, 84, 101
Fontfroide, Abbey of 108 Goswin 239
France, French 2, 4, 21, 33–5, 39–41, 53, 55, 57, 59, Gothic script 25–6, 28–41, 68, 81, 159–60, 169, 171,
66, 69, 81–2, 88, 91, 95, 98–9, 106, 117–18, 174, 186, 288
143, 145–6, 159–61, 165, 167, 171, 181, 186, Gottfried von Strassburg 328, 332, 334–5, 337
191, 238, 245, 248, 254–5, 282–3, 287–9, Grande Chartreuse 136
291–2, 305, 312–17, 319–20, 322–4, 326, Gratian 23, 62, 293, 295, 304, 306–8
332, 334–7, 339, 343 Greek 1, 54, 58, 63–4, 91, 140, 148–9, 153, 168, 216,
Franciscans 181 259, 261–4, 268–72, 275–6, 280–2
Frankenthal, Abbey of 38, 72, 74, 79–80 Gregory I, Pope 62, 111, 119, 138, 194, 319, 322, 341
Frankfurt 100 Gregory IX, Pope 308
Frederick II, emperor 289 Gregory of Tours 149
Frederick, abp. of Cologne 46 Grillius 252
Freiburg 54 Guadalquivir R. 284
Freising 37, 40 Guda, a nun 57, 83
friars 115, 181 Guernes de Pont-Ste-Maxence 320
Friuli 341 Guido di Castello 95–6
Frowin, abt. of Engelberg 98 Guigo, prior of Grande Chartreuse 136
Fruttuaria 42 Guischart de Beaulieu 321
Fulbert of Chartres 145, 260 Gundulf, monk of Bec 132
Fulco, bp. of Beauvais 89, 100 Guy, bp. of Beauvais 88–9
Fustat 91
Hainault 49
Gaelic 314 Hamersleben 76
Gaius Fannius 251 Hartmann von Aue 312, 332, 335, Fig. 18.4, 338
Galen Fig. 15.1, 283 Hasidei Ashkenaz 160
Galen, Ps. 283 Hayyim ben Isaac 161, 164
Gariopontus of Salerno 150–1, 280, Fig. 15.1, ˙
Hebrew 95, 160, 163–5, 167–73, 211, 259, 264, 288
286–7 Heinrich, a count 331
Garlandus of Besançon 89, 96, 100, 226 Heinrich der Glichezaere 342
Garlandus ‘compotista’ 100 Heinrich von Veldeke 63, 331–2, 334–5, 337–8, 342
Gasapino Antegnati 265 Helmarshausen, Abbey of 46
Geffrei Gaimar 315, 320 Herimann, monk of 48
Gennadius 94 Henricus Aristippus 268, 275–6
Geoffroy de Breteuil 114 Henry II, k. of England 46, 268, 312–13, 320,
Geoffrey Moricii 97 323, 334
Gerald of Wales 313 Henry his son 320
Gerard, abp. of York 95 And see Matilda
404 General Index
Henry III, emperor 131 Isaac Israeli Fig. 15.1, 284, 288
Henry IV, emperor 46 Isaac ben Moshe of Vienna 173
Henry V, emperor 46 Isidore 64, 70, 105, 154, 194, 243, 250, 316–17
Henry VI, emperor 63 Isidore, Ps. 305
Henry VIII, k. of England 21 Islam, Islamic 98, 212, 283, 288–9
Henry the Lion 46, 331–2, 334–5 Israel 159
Henry of Northampton 21 Italy, Italian 2, 10, 21, 39, 54, 57, 60, 62, 80–1, 117,
herbal 63–4 146, 160, 163–4, 167, 169, 171–2, 251,
Herbert of Bosham 170 263–5, 273, 277, 280, 283, 285–8, 291, 293,
Herbort von Fritslar 332, 335 295–6, 305, 309, 314
Herebertus ‘medicus’ 266, 277, 290 Ivo, bp. of Chartres 92–4, 96, 305–7
Hereford Cathedral 11, 73, 80, 93
And see Robert of Melun Jacob ben Meir (Rabbenu Tam) 159, 163
Herman de Valenciennes 319–20 James of Venice 272
Hermann, landgrave of Thuringia 61, 331, 335, 343 Jativa 274
Hermann of Carinthia 92, 264, 267–8, 271–2 Jean Beleth 190
Heron 268 Jerome 13, Fig. 3.1, Fig 3.3, 90, 168, 194, 201, 242,
Herrad of Hohenbourg 62 246–7
Hilary, bp. of Chichester 102 Jerome, Ps. 168
Hildebert, painter and scribe 57, 72, 74 Jerusalem 207–8
Hildesheim 278 Jews 9, 92, 95, 159–61, 163, 167, 170–1, 272, 288
St Michael’s Abbey 61 And see Hebrew
Hippocrates Fig. 15.1, 285, 291 Johann Amerbach 214
Hirsau, Abbey of 64, 75–6, 105, 181, 190, 329, 341 Johann Sensenschmidt 201
And see Conrad Johannes notarius sacri palatii 297
Hohenstaufens 289 Johannes, owner of medical books 277, Fig. 15.1,
holster books 14, 16, 155, 245, 279 286, 288
Homer lat. 98, 241–2 Johannes Constantiensis 145
Honorius of Autun 64, 190 Johannes ibn Daud 272
Horace 88–9, 97, 141, 241–2, 245 Johannes Hispanus 272
Huesca 268 Johannitius: see Hunayn
Hugh, bp. of Lincoln 46 John, St 285
Hugh Metel 101 John of Amiens, monk of Corbie 57
Hugh of Morville 332 John Cassian 136
Hugh of St-Cher 209, 214 John of Cornwall 94, 97, 314
Hugh of St-Victor 94–6, 123, 133, 138, 204, John of Fécamp 123, 131–3, 138
208, 213 John of Salisbury, bp. of Chartres 2, 89–91,
Hugo ‘magister’ 49 240, 246
Hugo ‘pictor’ Fig. 3.1, 48, 56–7, 74, 82 John of Seville 92, 264, 272
Hugo of Santalla 263–4, 266–7, 275 Jordan Fantosme 320
Hunayn ibn Ishaq (Johannitius) 275, Fig. 15.1, Joscelin of Soissons 225, 237
283, 290 Joseph, abt. of Reading 119
Judah he-Hasid 173
Ibn al-Haythām 263 ˙
Judah the Pious 160
Icelandic 325 Julius Caesar 244, 252, Fig. 13.2
Igny, Abbey of 247 Jumièges, Abbey of 41, 58, 128
Ilias Latina 241 And see Alexander
Indersdorf, Abbey of 329, 340 Justinian 293–4, 299–302, 307, 323
India 64 And see Law: roman
Ingelrannus, monk of Corbie 57 Juvenal 41, 88, 93, 97, 141, 241–2, 244
Innocent II, Pope 95
Investiture Contest 305 Kaiserchronik 334, 340–1
Iraq 159 Kalonymos family 171
Ireland, Irish 10, 22, 80, 84, 106, 313–14 Kent 39
Isaac ben Jacob 164 Kleve, Countess of 331
 General Index 405
Konrad von Würzburg 335 diurnal 180
Kremsmünster Abbey 329, 339–40 epistolary 108, 113–14, 127, 176
evangeliary 176
La Garde-Dieu, Abbey of 107–8 exultet roll 62
La Rochelle, Abbey of 161, Fig. 9.1, 164 gospel book 21, 46, 61, 108, 113–14, 118
Laelius 251 gradual 108, 111, 176–7, 179, 184–5, 189, 191
Lahn, R. 341 hymnal 177, 189
Lambach, Abbey of 329, 340 lectionary 114, 127, 176–8
Lambert of St-Omer 62 Liber comicus 125
Lamspringe, Nunnery of 76, 79–80 manual 183
See also Ermengarde, Odelgarde, nuns of matutinal 180
Lanfranc, abp. of Canterbury 112–13, 119 missal 14, 61–2, 79, 108, 113–14, 127, 175–9, 183,
Laon 193–5, 196, 200, 203 189–90
See also Anselm of, Ralph of nocturnal 180
Larino 280 ordinal 176, 178
Laudine 325 passional 14, 62, 111
law pontifical 183–4, Fig. 10.2, 190
Admonitio Generalis 296 processional 183
Anglo-Saxon 294 psalter 14, 60–1, 177, 180, 317, Fig. 17.1, 319, 322,
Canon 13, 62, 92–3, 293–5, 302–8 325–6
Collectio Anselmo dedicata 305 sacramentary 61, 108, 176, 190
Collectio Brittanica 306 troper 14, 23, 185, 191
Epitome Codicis 300–1, 304, 309 Liutprand, K. 295–6
Epitome Juliani 297 Llanbadarn Fawr 314
Liber Papiensis 295–6, 299, 301, 304, 309 Lodewijk I, count of Loon 338
Lombard 293–6, 299–300, 302, 304–5, 307–9 Loire, R. 102
Panormia 306, 310 Lombards 286, 295–6
Roman 91, 293–4, 299–302, Fig. 16.2, 304–5, London 2, 21
307–9 St Paul’s Cathedral 21
Rothari’s Edictum 295 Lothar, K. 297
Salic 293–4 Lotharingia 260
Tripartita 306 Louis the Pious 296
Walcausina 295, 297, Fig. 16.1, 299–300, Low Countries (Netherlands) 2, 21, 33, 39, 81, 142,
302, 309 144, 148
And see Anselm of Lucca, Burchard, Lucan 97–8, 141, 152, 241–2, 245, 248, Fig. 13.1,
Gratian, Ivo 250–1
Leicester, St Mary-in-the-Fields 248, 254 Lucca 171
Leiningen, Count of 335 Ludwig III, landgrave of Thuringia 331
Leis Willelme 319 Ludwig, monk of Wessobrunn 76
Leo, a priest 280 Luxeuil 88
Liber Catonianus 242
Liber Pancrisis 203 Macer, Ps. Fig. 15.1
Liège 21, 260 Macrobius 88, 98, 143
Liessies Abbey 82 Mahberet of Menahem ben Saruq 161, 169
Limoges 21 ˙ zor Vitry 161, 163,
Mah ˙ 169, 172
Lincolnshire 11 ˙ R. 328
Main,
Lincoln 11, 80 Mainz 159, 171
liturgical books: Malmesbury Abbey 78–80, 82
antiphoner 111, 177, 179, 184, 191 And see Robert, William of
benedictional 114, 183, 190 Malvern Priory 269
breviary 13–14, 62, 108, 127, 175, 177–80, 183 Manegold of Lautenbach 254
calendar 13, 60, 178, 180, 182, 190 Manerius 70
cantatorium 14, 23, 176, 185 Manfred, k. of Sicily 289
collectar 177 Marbach, Abbey of 48
customary 176 Sintram, canon of 48
406 General Index
Marbod 317 Notker the German 334, 340, 344
Marburg 327 Nottingham 322
Marie de Champagne 322 Noyon 291
Marie de France 313 Nuremberg 201, 214
Marius Victorinus 216, 230, 252
Martianus Capella 141–3, 152, 235, 255–7, 267 Oberaltaich, Abbey of 341
Martin of Braga 247 Occitan 316–17, 323, 326, 332
Marvels of the East 64 Odelgarde 76
Māshā’allāh 272 Odo, student of Abelard 95, 170
masora 172–3 Oigny, Abbey of 120
Matilda, wife of Henry the Lion 331, 334 Old Norse 325
Matilda, countess of Tuscany 132 Orderic Vitalis 38, 42, 71
Mauger, bp. of Worcester 115 Origen 90, 168
Maurice of Kirkham 95 Orléans 260
Maximianus 97, 241 Orosius 251
Meaux Abbey 110, 119 Orval, Abbey of 244
Menahem ben Saruq 161, 169 Otloh of St Emmeram 334
Menah˙ em ben Yehosadaq 164 Otte 332
Meran,˙ duke of 331 ˙ Otto V, count of Scheyern 343
Merovingians 73 Otto of Freising 63, 98, 344
Metrodora 282 Ottonians 46, 53–4, 57, 70, 295
Meuse, region of 21, Fig. 3.2, 57–8, 60, 291, 337–8 Ourscamp, Abbey of 291
Michael of Ireland 97 Ovid 88–90, 98, 100, 141, 241–2, 245
Millstatt, Abbey of 329 Oxford 2, 265, 288, 317
Minnesang 328 schools 115, 243, 265
Moggio, Abbey of 341 St Frideswide, Augustinian Canonry 321
Mont-St-Michel, Abbey of 58, 71, 73, 79, 92,
128, 149 Palermo 268, 289
Monte Cassino, Abbey of 91, 96, 150, 261–2, Palinurus, Cape 268
279–80, 282–3, 285–6, 288 Pamphilus 97–8, 102
And see Constantine, Desiderius paper 9, 22, 264–5, 274
Montfaucon, Bernard de 164 Papias 88
Morigny, Abbey of 96 parchment 9–11, 18, 22, 267
Mozarabic 61, 283 parchmenters 11
Munich 72 Paris 20, 71, 90, 92, 95–8, 107, 116, 134, 170, 192,
Muscio 280–2 198, 200, 209, 224, 260, 287, 288
Muslim: see Islam booktrade 15, 59, 62, 70, 81, 192
schools 90, 93, 95, 116, 134, 192, 215, 217, 243
Namur 46 Cathedral of Notre-Dame 96, 192, 195,
Nantes 243 201, 204
Nemesius 91, 261, 271 Ste-Geneviève, 192
Netherlands: see Low Countries And see St-Victor
Nevelo, monk of Corbie 57 Parzival 335
Nevelo of Compiègne 88 Passau, St Nicholas 105–6
Newhouse Abbey 11 Paucapalea 307
Nicholas of Salerno 287 Paul, St 246–7, 251
Nigel Witeker 36 Paul, abt. of St Albans 70–1
Nimrod 285 Pavia 295, 297
Noah’s Ark 209 Persius 90, 141, 152, 241–2, 244
Norbert of Xanten 46 Peter Abelard 2, 88, 93–8, 101, 134, 138, 215, 217,
Normandy, Normans 35, 38–9, 41–2, Fig. 3.1, 46, 221, 226, 228–9, 237–8
48, 56–7, 71, 74, 78–9, 81, 91–2, 100, 105, Peter of Blois 243
114, 128, Fig. 9.2, 170, 312, 322, 324 Peter the Chanter 204, 213
Northumberland 313 Peter Comestor 23, 203–4, 207–8, 213
Northungus 278 Peter of Eboli 63
 General Index 407
Peter Lombard 13, 94–8, 101–2, 195, 201, Fig. 11.3, Remigius 145
204–6, 213, 307 Rhine, R. 328
Peter of Poitiers 207–8, 213 Rhineland 58, 60, 72, 159, 171
Peter the Venerable 9, 92, 212 Richard I, k. of England 332
Peterborough 319 Richard Poore, bp. of Winchester 324
Petrus Alfonsi 101, 268, 321 Richard of St-Victor 207–8
Petrus Helias 93 Rievaulx Abbey 110, 119
Petrus Pictor 88 Robert, monk of Malmesbury 104
Pfaffe Konrad 332, Fig. 18.3, 334, 337 Robert Amiclas 93
Pfaffe Lambrecht 328, 332 Robert of Cricklade 268
Philaretus Fig. 15.1 Robert de Galone 100
Philip of Harcourt, bp. of Bayeux 91, 99 Robert Grosseteste 276
Philippe de Thaon 317, 320–1 Robert of Ketton 92, 212, 264, 268, 271
Pictor in Carmine 54 Robert of Melun, bp. of Hereford 93, 213
Pippin, K. 296 Robert of Selby 268
Pistoia 87–8 Robert of Torigny 92
Plato 58, 64, 141, 143, 217, 241, 262, 275 Roche Abbey Fig. 6.2
Plato of Tivoli 264, 271–2 Rochester Cathedral Priory 39, 42
Plautus 241 And see Ralph
Pliny the Elder 244, 268 Roger, abp. of York 101
Plutarch 268 Roger the hermit 108, 118
Poitiers 200, 252 Roger de Mowbray 313
Porphyry 89, 215–17, 219–20, 222, 230 Roger Norreis, abt. of Evesham 115
Portuguese 314 Roman Empire 2, 53
Prémontré, Premonstratensians 11, 46, 106, 112, Romance 311, 316–17, 320, 322, 324, 328, 335,
120, 181, 329, 341 337–9
Priscian 88–9, 93, 97–8, 141, 146, 148, 151, 155, 238, ‘Romanesque’ decoration 43, 51, 53, 55, 61, 65
241–2 Rome 54, 96, 103, 105, 115, 208, 251
Prudentius 63, 242 Roscelin 88–9, 99–100, 237
Prüfening Fig. 3.4, 64, 70, 82, 114 Rouen 95, 159
See Swicher St-Ouen 41
Ptolemy 92, 262, 268 Rudolph of Bruges 271–2
Publilius Syrus 247 Rufillus 48
Rufford Abbey 322
Qairouan 261, 268 Rule of St Augustine 114
Qur’ān, al 92, 212 Rule of St Benedict 48, 113–14, 124, 132, 135, 181, 329
Qustā ibn Lūqā 271 Rule of Chrodegang 78
˙ Rumanian 325
Rabanus Maurus 194, 206, 237 Ruotger, monk of Bamberg 105
Ralph, prior of Rochester 39
Ralph of Laon 15, 193 Sa‘adya Gaon 167
Rashi (Solomon ben Isaac) 95, 159, 161, St Albans Abbey 49, 63, 71, 79, 108, 272, 317,
169–70, 173 Fig. 17.1, 319
Raulinus of Fremington 71, 82 And see Paul, Simon
Reading Abbey 106, 108, 119 St-Amand, Abbey of 262
And see Joseph St-André-de-Rosans, Abbey of 100
Red Sea, Crossing of 60, 62 St-Bertin, Abbey of 262
Regensburg 63–4, 70 St-Denis, Abbey of 134
St Emmeram 46 St-Évroult, Abbey of 38, 71, 105, 128
Reginfridus 89 St-Victor, Abbey of 70, 97–8, 107, 116, 133, 192,
Regino of Prüm 305 208, 246
Reichenau 260 Liber Ordinis of 19, 70, 107, 111–13, 120
Reims 247 And see Andrew of, Hugh of, Richard of
St-Thierry 242 St-Wandrille, Abbey of 108
Reinhardsbrunn, Abbey of 329 S. Paulo fuori le Mura 60
408 General Index
Saints’ Lives 14, 49, 51, 62, 65, 179, 341 Stephen, k. of England 324
Salerno, Salernitan 91, 261, 268, 278, 280, 286–8, Stephen Langton, abp. of Canterbury 116
290, 292 Stephen of Antioch/of Pisa 263, 274, 279, 288–9
And see Alfanus Strassburg 200
Salians 295 Suetonius 105, 250
Salisbury Cathedral 78–80, 82, 84 Swabia 64, 75
Osbern, bp. of 78 Swicher, monk of Prüfening 70
Sallust 91, 241–2, 244, 252 Switzerland 33, 117, 340
Salomon, monk of St Augustine’s Canterbury Sylvester II: see Gerbert of Aurillac
267, 277–8, 290
Salzburg, St Peter’s 262 Talmud 9, 159–61, 163, 165, 168–9, 171, 174
Samuel ben Meir (Rashbam) 159 Targum 165, Fig. 9.2, 167–8, 173
San Benedetto, Abbey of 280 Tegernsee, Abbey of 90
San Millán de Cogolla, Abbey of 325 Terence 63, 141, 145, 152, 241–2, 251
Sankt Blasien, Abbey of 329 Tertullian 145
Sankt Gallen, Abbey of 282, 308, 334 Themistius 268
Sankt Georgen, Abbey of 329 Theobald ‘parcamenarius’ 11
St. Trudperter Hohes Lied 329 Theodolus 89, 93, 97
Sanson de Nantueil 321 Theodorus Priscianus 284
Saxony 46, 76, 79, 332 Theophilus Fig. 15.1
Scaevola 251 Thierry of Chartres 90–2, 99–100, 145, 219, 254–6
Scandinavia 117 Heptateuchon 90–1, 219, 245–6, 267–8
schools 3, 90, 98, 116, 122–4, 126–27, 133, 135, 150, Thomas of Brittany 320
192, 194–5, 212, 215, 217, 236, 255 Thomas, abt. of Hyde 119
And see Oxford, Paris Thomas Becket, abp. of Canterbury 201,
Schwarzenthann, Nunnery of 48 Fig. 11.3, 317
Guta, nun of 48 Thomas of Malborough 115
Scotland 10, 313–14 Thuringia 329, 335, 337–8
scriptoria 4, 76–80, 83–4 Tironian notes 35, 40, 153–4, 300
Sefer Hasidim 160 Toledo 9, 92, 264, 272, 274, 288–9
Segovia 92 Torah 163–4
Seneca the Younger 90–1, 96, 246–7 Tosafists 159
Seneca, Ps. 247, 250–1 Touraine 317, 322
sententiae 93, 203–4 Tours 56
sequence 184–5, 189 Tree of Jesse 60–1
Servius 142, 241 Trier 79, 260, 340–1
Shemayah 160 St Matthias 340
Shrewsbury 11 trope 184–5
Shropshire 95 Trota 287
Sicily 2, 53–4, 63, 80, 263–4, 268, 289, 317, 323 Troyes 49, 159, 164
siddur 161, 172 Tuscan 325
Sigebert of Gembloux 149
Silos, Santo Domingo de, Abbey of 125, 325 Ulrich von Zatzikhofen 332, 337
Silvanès, Abbey of 107 Ulrich of Zell 105, 112
Silvester, St 341
Simhah of Vitry 159 Valcroissant, Abbey of 119
˙ a priest 100
Simon, Valerius Maximus 90
Simon, abt. of St Albans 49, 108, 195 Varro 241
Socrates 217 Vegetius 90, 284, 291
Solomon 267, 290 Veneto 265
Sophie, daughter of Count Otto of Scheyern 343 Venice 54
Spain (or Iberian Peninsula) 2, 9, 61, 80, 92, 117, Victor, St 102
150, 160–1, 165, 170, 263–4, 271, 273–4, Victorines 106
283, 325 Virgil 88–9, 97, 141–3, 152, 155, 241–2
Statius 88, 97–8, 141, 241–2, 245 Virgil, Ps. 88
 General Index 409
Virgin and Child 61 William de Fonte Morini 97
Visigothic script 25, 40, 80, 84 William of Hirsau 105, 112, 120
Vitalis, abt. of Savigny 69 William of Lucca 226
Vivarium 212 William of Malmesbury 2, 78, 83, 95, 104–5
Vorau, Abbey of 329, 340 William of St-Calais 46
William of St-Thierry 58, 96, 247, 323
Wace 313, 320 Williram of Ebersberg 329, Fig. 18.1, 334, 340,
Walcausus/Gualcausus of Pavia 297 344
Walcher of Malvern 269, 275 Winchcombe Abbey 11, 80, 84
Wales, Welsh 36, 313–14 Winchester Cathedral Priory 20, 51, 109
Walter of Mortagne 88 And see Richard Poore
Waltham Abbey 262 Witham, Carthusian house of 46
Wearmouth-Jarrow 20 Wittelsbachs 338, 343
Wedric, abt. of Liessies 49 Wolfenbüttel 54
Wessobrunn, Abbey of 72, 75–6, Fig. 4.1, 79 Wolfram von Eschenbach 328–9, 332, 335
See also Diemut, Ludwig Wolfrat von Tengelingen 331
Wibald of Corvey 245 Worcester Cathedral Priory 11, 19, 101, 268–9,
Wido 296 288–9, 292, 315
Wigmore, Abbey of 95 And see Mauger
Wilgelmus 239 Worms 72, 159, 171, 305
William I, k. of Sicily 268 World War II 292
William, nephew of Peter of Blois 243 Wulfram, abt. of Bamberg 105
William, oblate of St-Évroult 71
William of Champeaux 225–6, 239, 254 York Minster 95
William de Chemillé, bp. of Angers 161 Yorkshire 110
William of Conches 90
William the Conqueror 132 Zwiefalten 64, 329
 CAMBRIDGE STUDIES IN MEDIEVAL LITERATURE

1 Robin Kirkpatrick Dante’s Inferno: Difficulty and Dead Poetry

2 Jeremy Tambling Dante and Difference: Writing in the “Commedia”
3 Simon Gaunt Troubadours and Irony
4 Wendy Scase “Piers Plowman” and the New Anticlericalism
5 Joseph J. Duggan The “Cantar de mio Cid”: Poetic Creation in its Economic
and Social Contexts
6 Roderick Beaton The Medieval Greek Romance
7 Kathryn Kerby-Fulton Reformist Apocalypticism and “Piers Plowman”
8 Alison Morgan Dante and the Medieval Other World
9 Eckehard Simon (ed.) The Theatre of Medieval Europe: New Research in Early
Drama
10 Mary Carruthers The Book of Memory: A Study of Memory in Medieval
Culture
11 Rita Copeland Rhetoric, Hermeneutics, and Translation in the Middle Ages:
Academic Traditions and Vernacular Texts
12 Donald Maddox The Arthurian Romances of Chrétien de Troyes: Once and
Future Fictions
13 Nicholas Watson Richard Rolle and the Invention of Authority
14 Steven F. Kruger Dreaming in the Middle Ages
15 Barbara Nolan Chaucer and the Tradition of the “Roman Antique”
16 Sylvia Huot The “Romance of the Rose” and its Medieval Readers:
Interpretation, Reception, Manuscript Transmission
17 Carol M. Meale (ed.) Women and Literature in Britain, 1150–1500
18 Henry Ansgar Kelly Ideas and Forms of Tragedy from Aristotle to the Middle
Ages
19 Martin Irvine The Making of Textual Culture: ‘Grammatica’ and Literary
Theory, 350–1100
20 Larry Scanlon Narrative, Authority, and Power: The Medieval Exemplum and
the Chaucerian Tradition
21 Erik Kooper (ed.) Medieval Dutch Literature in its European Context
22 Steven Botterill Dante and the Mystical Tradition: Bernard of Clairvaux in the
“Commedia”
23 Peter Biller and Anne Hudson (eds) Heresy and Literacy, 1000–1530
24 Christopher Baswell Virgil in Medieval England: Figuring the “Aeneid” from
the Twelfth Century to Chaucer
25 James Simpson Sciences and the Self in Medieval Poetry: Alan of Lille’s
‘Anticlaudianus’ and John Gower’s ‘Confessio Amantis’
26 Joyce Coleman Public Reading and the Reading Public in Late Medieval
England and France
27 Suzanne Reynolds Medieval Reading: Grammar, Rhetoric and the Classical
Text
28 Charlotte Brewer Editing ‘Piers Plowman’: The Evolution of the Text
29 Walter Haug Vernacular Literary Theory in the Middle Ages: The German
Tradition, 800–1300, in its European Context
30 Sarah Spence Texts and the Self in the Twelfth Century
31 Edwin D. Craun Lies, Slander and Obscenity in Medieval English Literature:
Pastoral Rhetoric and the Deviant Speaker
32 Patricia E. Grieve “Floire and Blancheflor” and the European Romance
33 Huw Pryce (ed.) Literacy in Medieval Celtic Societies
34 Mary Carruthers The Craft of Thought: Meditation, Rhetoric, and the Making
of Images, 400–1200
35 Beate Schmolke-Hasselmann The Evolution of Arthurian Romance: The Verse
Tradition from Chrétien to Froissart
36 Siân Echard Arthurian Narrative in the Latin Tradition
37 Fiona Somerset Clerical Discourse and Lay Audience in Late Medieval England
38 Florence Percival Chaucer’s Legendary Good Women
39 Christopher Cannon The Making of Chaucer’s English: A Study of Words
40 Rosalind Brown-Grant Christine de Pizan and the Moral Defence of Women:
Reading Beyond Gender
41 Richard Newhauser The Early History of Greed: The Sin of Avarice in Early
Medieval Thought and Literature
42 Margaret Clunies Ross (ed.) Old Icelandic Literature and Society
43 Donald Maddox Fictions of Identity in Medieval France
44 Rita Copeland Pedagogy, Intellectuals, and Dissent in the Later Middle Ages:
Lollardy and Ideas of Learning
45 Kantik Ghosh The Wycliffite Heresy: Authority and the Interpretation of Texts
46 Mary C. Erler Women, Reading, and Piety in Late Medieval England
47 D. H. Green The Beginnings of Medieval Romance: Fact and Fiction, 1150–1220
48 J. A. Burrow Gestures and Looks in Medieval Narrative
49 Ardis Butterfield Poetry and Music in Medieval France: From Jean Renart to
Guillaume de Machaut
50 Emily Steiner Documentary Culture and the Making of Medieval English
Literature
51 William E. Burgwinkle Sodomy, Masculinity, and Law in Medieval Literature:
France and England, 1050–1230
52 Nick Havely Dante and the Franciscans: Poverty and the Papacy in the
“Commedia”
53 Siegfried Wenzel Latin Sermon Collections from Later Medieval England:
Orthodox Preaching in the Age of Wyclif
54 Ananya Jahanara Kabir and Deanne Williams (eds.) Postcolonial Approaches
to the European Middle Ages: Translating Cultures
55 Mark Miller Philosophical Chaucer: Love, Sex, and Agency in the “Canterbury
Tales”
56 Simon A. Gilson Dante and Renaissance Florence
57 Ralph Hanna London Literature, 1300–1380
58 Maura Nolan John Lydgate and the Making of Public Culture
59 Nicolette Zeeman ‘Piers Plowman’ and the Medieval Discourse of Desire
60 Anthony Bale The Jew in the Medieval Book: English Antisemitisms, 1350–1500
61 Robert J. Meyer-lee Poets and Power from Chaucer to Wyatt
62 Isabel Davis Writing Masculinity in the Later Middle Ages
63 John M. Fyler Language and the Declining World in Chaucer, Dante, and Jean
de Meun
64 Matthew Giancarlo Parliament and Literature in Late Medieval England
65 D. H. Green Women Readers in the Middle Ages
66 Mary Dove The First English Bible: The Text and Context of the Wycliffite
Versions
67 Jenni Nuttall The Creation of Lancastrian Kingship: Literature, Language and
Politics in Late Medieval England
68 Laura Ashe Fiction and History in England, 1066–1200
69 J. A. Burrow The Poetry of Praise
70 Mary Carruthers The Book of Memory: A Study of Memory in Medieval
Culture (Second Edition)
71 Andrew Cole Literature and Heresy in the Age of Chaucer
72 Suzanne M. Yeager Jerusalem in Medieval Narrative
73 Nicole R. Rice Lay Piety and Religious Discipline in Middle English Literature
74 D. H. Green Women and Marriage in German Medieval Romance
75 Peter Godman Paradoxes of Conscience in the High Middle Ages: Abelard,
Heloise, and the Archpoet
76 Edwin D. Craun Ethics and Power in Medieval English Reformist Writing
77 David Matthews Writing to the King: Nation, Kingship, and Literature in
England, 1250–1350
78 Mary Carruthers (ed.) Rhetoric Beyond Words: Delight and Persuasion in the
Arts of the Middle Ages
79 Katharine Breen Imagining an English Reading Public, 1150–1400
80 Antony J. Hasler Court Poetry in Late Medieval England and Scotland:
Allegories of Authority
81 Shannon Gayk Image, Text, and Religious Reform in Fifteenth-Century
England
82 Lisa H. Cooper Artisans and Narrative Craft in Late Medieval England
83 Alison Cornish Vernacular Translation in Dante’s Italy: Illiterate Literature
84 Jane Gilbert Living Death in Medieval French and English Literature
85 Jessica Rosenfeld Ethics and Enjoyment in Late Medieval Poetry: Love after
Aristotle
86 Michael Van Dussen From England to Bohemia: Heresy and Communication
in the Later Middle Ages
87 Martin Eisner Boccaccio and the Invention of Italian Literature: Dante,
Petrarch, Cavalcanti, and the Authority of the Vernacular
88 Emily V. Thornbury Becoming a Poet in Anglo-Saxon England
89 Lawrence Warner The Myth of “Piers Plowman”: Constructing a Medieval
Literary Archive
90 Lee Manion Narrating the Crusades: Loss and Recovery in Medieval and Early
Modern English Literature
91 Daniel Wakelin Scribal Correction and Literary Craft: English Manuscripts
1375–1510
92 Jon Whitman (ed.) Romance and History: Imagining Time from the Medieval
to the Early Modern Period
93 Virginie Greene Logical Fictions in Medieval Literature and Philosophy
94 Michael Johnston and Michael Van Dussen (eds.) The Medieval Manuscript
Book: Cultural Approaches
95 Tim William Machan (ed.) Imagining Medieval English: Language Structures
and Theories, 500–1500
96 Eric Weiskott English Alliterative Verse: Poetic Tradition and Literary History
97 Sarah Elliott Novacich Shaping the Archive in Late Medieval England:
History, Poetry, and Performance
98 Geoffrey Russom The Evolution of Verse Structure in Old and Middle English
Poetry: From the Earliest Alliterative Poems to Iambic Pentameter
99 Ian Cornelius Reconstructing Alliterative Verse: The Pursuit of a Medieval
Meter
100 Sara Harris The Linguistic Past in Twelfth-Century Britain
101 Erik Kwakkel and Rodney Thomson (eds.) The European Book in the Twelfth
Century