VERBAL AND VISUAL COMMUNICATION

UTRECHT STUDIES IN MEDIEVAL LITERACY

37
UTRECHT STUDIES IN MEDIEVAL LITERACY

General Editor

Marco Mostert (Universiteit Utrecht)

Editorial Board

Gerd Althoff (Westfälische-Wilhelms-Universität Münster)

Michael Clanchy (University of London)
Erik Kwakkel (Universiteit Leiden)
Mayke de Jong (Universiteit Utrecht)
Rosamond McKitterick (University of Cambridge)
Arpád Orbán (Universiteit Utrecht)
Armando Petrucci (Scuola Normale Superiore di Pisa)
Richard H. Rouse (UCLA)
 VERBAL AND VISUAL
COMMUNICATION
IN EARLY ENGLISH TEXTS

Edited by

Matti Peikola, Aleksi Mäkilähde, Hanna Salmi,

Mari-Liisa Varila, and Janne Skaffari

F
H
 British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

The image on the front cover of this book, British Library Additional
37049, f. 9v, is made available under a Creative Commons CC0 1.0
Universal Public Domain Dedication, and can be found online at:
https://www.bl.uk/catalogues/illuminatedmanuscripts/ILLUMIN.ASP?
Size=mid&IIIID=13370

© 2017 – Brepols Publishers n.v., Turnhout, Belgium

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

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

D/2017/0095/201

ISBN 978-2-503-57464-6

e-ISBN 978-2-503-57465-3

DOI: 10.1484/M.USML-EB.5.112805

Printed on acid-free paper

 Contents

Abbreviations vii
Preface ix

Part I: Discourse Linguistics Meets Book History

Disciplinary Decoding: Towards Understanding the Language of

Visual and Material Features
MARI-LIISA VARILA, HANNA SALMI, ALEKSI MÄKILÄHDE,
JANNE SKAFFARI, and MATTI PEIKOLA 1

Part II: Communicating through Layout

Discourse Variation, Mise-en-page, and Textual Organisation in

Middle English Saints’ Lives
COLETTE MOORE 23
How the Page Functions: Reading Pitscottie’s Cronicles in
Manuscript and Print
FRANCESCA L. MACKAY 41
Verbal and Visual Communication in Title Pages of Early Modern
English Specialised Medical Texts
MAURA RATIA and CARLA SUHR 67
Quantifying Contrasts: A Method of Computational Analysis of
Visual Features on the Early Printed Page
JUKKA TYRKKÖ 95
vi Contents

Part III: Communicating through Script and Typography

Stating the Obvious in Runes

YIN LIU 125
Labours Lost: William Caxton’s “Otiose” Sorts, c. 1472-1482
ANYA ADAIR 141
Code-Switching, Script-Switching, and Typeface-Switching in Early
Modern English Manuscript Letters and Printed Tracts
SAMULI KAISLANIEMI 165
Seeing is Reading: Typography in Some Early Modern Dictionaries
R.W. MCCONCHIE 201
Whose Letters Are They Anyway? Addressing the Issue of Scribal
Writing in Bess of Hardwick’s Early Modern English Letters
I.J. MARCUS 219

Bibliography 251
Index 277
 Abbreviations

CEECS Corpus of Early English Correspondence Sampler (Helsinki, 1998), at:

<http://www.helsinki.fi/varieng/CoRD/corpora/CEEC/ceecs.html>.
DIMEV Digital Index of Medieval English Verse, at: <http://www.dimev.net/>.
ECCO Eighteenth Century Collections Online, at: <http://www.gales.com/
primary-sources/eighteenth-century-collections-online/>.
EEBO Early English Books Online, at: <http://eebo.chadwyck.com/home>.
EMEMT Early Modern English Medical Texts, see: <http://www.helsinki.fi/
varieng/CoRD/corpora/CEEM/EMEMTindex.html>.
ESTC English Short Title Catalogue, at: <http://estc.bl.uk/F/?func=file&file_
name=login-bl-estc>.
IMEV C. BROWN and R.H. ROBBINS, Index of Middle English Verse (New York,
1943), and R.H. ROBBINS and J.L. CUTLER, Supplement to the Index of
Middle English Verse (Lexington, KY, 1965).
LALME A. MCINTOSH, M.L. SAMUELS and M. BENSKIN, A Linguistic Atlas of
Late Mediaeval English, 4 vols. (Aberdeen, 1986).
LMEMT Late Medieval English Medical Texts, see: <http://www.helsinki.fi/
varieng/CoRD/corpora/CEEM/LMEMTindex.html>.
OED Oxford English Dictionary [started as: A New English Dictionary on
Historical Principles, Founded Mainly on the Materials Collected by The
Philological Society], 2nd and 3rd edn. (Oxford, 1989-). <www.oed.
com>.
STC POLLARD, A.W., and G.R. REDGRAVE, A Short-Title Catalogue of Books
Printed in England, Scotland, and Ireland, and of English Books Printed
Abroad 1475-1640, 2nd edn. W.A. JACKSON, F.S. FERGUSON, and K.F.
PANTZER, 3 vols. (London, 1976-1991).
TEAMS TEAMS Middle English Texts Series, at: <http://d.lib.rochester.edu/teams>.
WING WING, D., Short-Title Catalogue of Books Printed in England, Scotland
and Ireland, Wales and British America and of English Books Printed in
Other Countries 1641-1700, 2nd. edn., 3 vols. (1972-1988).
 Preface

THE EDITORS

he foundations for the present volume were laid at the international sym-

T posium Linguistics Meets Book History: Seeking New Approaches, or-

ganised by the ‘Pragmatics on the Page’ research team in Turku, Finland,
24-25 October, 2014. This event brought together thirty researchers from Fin-
land, the UK, the USA, and Canada to explore the interplay between the verbal
/ linguistic and the visual / material in medieval and early modern texts, with a
focus on both handwritten and printed media. The symposium helped to facili-
tate the exchange of ideas, methods and best practices between two broad fields
which – despite their mutual interest in this area – are not often brought into
dialogue with each other: linguistic scholarship (especially discourse linguis-
tics, pragmatics, stylistics) and book history (including textual scholarship and
material philology).
In our work towards this book we have incurred many debts of gratitude.
First of all, heartfelt thanks are due to our ten contributors for their high-quality
chapters and their unfailing commitment and collaboration during the editorial
process. We are grateful to Marco Mostert for providing us with an excellent
publication forum in Utrecht Studies in Medieval Literacy and for his erudition
and experience as the General Editor of the series. We wish to thank Ms.
Annimari Ollila for her invaluable assistance in copyediting and compiling the
general bibliography, and Ms. Liina Repo for her indispensable help with com-
piling the index. This book would not have materialised without the stimulating
intellectual environment of the ‘Pragmatics on the Page’ research team (2010-)
and the contributions made to it by our colleagues and team members Risto
Hiltunen, Ruth Carroll, Carla Suhr, and Ellen Valle. During the preparation of
x Preface

the volume, the members of the editorial team have been supported by research
funding from the Academy of Finland (decision numbers 136404, 257059 and
258434), the Alfred Kordelin Foundation, the Ella and Georg Ehrnrooth Foun-
dation and the Finnish Cultural Foundation. We also wish to express our grati-
tude to the Turku University Foundation, Turku Centre for Medieval and Early
Modern Studies (TUCEMEMS), the Department of English at the University of
Turku, and the City of Turku for their financial support of the 2014 sympo-
sium.
 Part I

Discourse Linguistics Meets Book History

 Disciplinary Decoding: Towards Understanding
the Language of Visual and Material Features*

MARI-LIISA VARILA, HANNA SALMI,

ALEKSI MÄKILÄHDE, JANNE SKAFFARI,
and MATTI PEIKOLA

Introduction

hen reading a text, our understanding of its meaning is influenced by

W the visual form and material features of the page on which it appears.
The use of Comic Sans MS as a body text font in a scholarly publica-
tion, for example, might raise suspicions about the reliability of the research
reported in it.1 Recent years have witnessed a growing interest in the visual and
material elements of early written communication in its various forms. The
*
The authors wish to express their gratitude to the Academy of Finland (decision numbers
136404, 257059 and 258434), the Alfred Kordelin Foundation, the Ella and Georg Ehrnrooth
Foundation, and the Finnish Cultural Foundation for funding the research projects during which
this volume was edited. We gratefully acknowledge the assistance of Ms. Annimari Ollila in the
copyediting of the manuscript.
1
See also S. KAISLANIEMI, “Code-switching, script-switching and typeface-switching in
Early Modern English manuscript letters and printed tracts”, in this volume, p. 199, n. 62.

......................................................................................................................................
Verbal and Visual Communication in Early English Texts, ed. Matti PEIKOLA, Aleksi
MÄKILÄHDE, Hanna SALMI, Mari-Liisa VARILA, and Janne SKAFFARI, Utrecht Studies in Medieval
Literacy, 37 (Turnhout: Brepols, 2017), pp. 1-20.

DOI 10.1484/M.USML-EB.5.114128
2 MARI-LIISA VARILA et al.

study of features such as layout and decoration has been brought to bear on
analyses of language and text. However, the terminology used to describe vi-
sual, material and physical features of early written communication varies
across different disciplines and even within the same field.
This introductory chapter explores terminology used in different fields to
refer to the material and visual features of texts in order to set the stage for the
contributions in the present volume. The vocabulary for describing visual and
material features of written communication is varied, and scholars focus on
different physical aspects of texts based on their individual and disciplinary
perspectives. First, fields concerned mainly with text and / or language include,
for example, discourse linguistics and textual scholarship, while in manuscript
studies and bibliography there is more emphasis on the visual and material
context in which that text is embedded. Secondly, the medium of written com-
munication studied – for example manuscript or print – may lead to different
terminology being used for similar concepts in different fields. While some of
this terminology is necessarily dictated by the specific technology of text pro-
duction employed, differences in terminology may hide the fact that many
underlying concepts might be usefully shared among researchers of different
media. Thirdly, research design is often influenced by traditional chronological
boundaries established to define historical periods (medieval / early modern)
or stages of language (Old English / Middle English). This may lead to gaps in
researchers’ awareness of other fields.
We hope that by crossing disciplinary and chronological boundaries and
increasing dialogue between different fields, it is possible to enhance and re-
shape theories and methods developed to describe and analyse these features.
While it is not necessary for every linguistic or textual study to contain an
analysis of material and visual features, or vice versa, an understanding of how
these features work with or against the linguistic and textual elements on the
page has the potential to benefit everyone working on early written communi-
cation. Moreover, it is important to be informed of research addressing visual
elements in modern texts (e.g. printed or digital), some of which we also re-
view in this chapter.
Our survey of terms and ideas is broad but can hardly be exhaustive. As the
focus of this chapter is on concepts applicable to research combining language
/ text with the visual / material, and not on the range of technical terms used in
fields such as manuscript studies, palaeography and bibliography, we do not
consistently define such terms as, for example, ‘fount’, ‘gathering’ and ‘ru-
Disciplinary Decoding 3

bric’, which nevertheless appear in this and the following chapters. As in many
disciplines focusing on features of everyday objects, some terms are problem-
atic because of their diverse uses: ‘emphasis’, ‘highlighting’, ‘conspicuity’,
‘foregrounding’ and ‘salience’, for example, are variously used as either clear-
ly defined technical terms or more general words. We return to these items in
the section ‘Communicating through script and typography’, but before that
elaborate on the importance of the non-verbal in written communication (in the
section ‘Visual and material features of written communication’) and on the
impact of page layout (in the section ‘Communicating through layout’); con-
cluding remarks follow in the section ‘Towards managing visual and material
features in theory and practice’. The chapters of the present volume are intro-
duced in the final section. Our contributors explore a variety of historical Eng-
lish textual situations and propose novel methods for analysing visual and
material features, including new digital applications. They map the use of such
features (for instance layout design or choice of script / typeface) against lin-
guistic features (for instance code-switching and spelling variation) to consider
how these choices reflect the communicative purposes of the text, for example
guiding readers to navigate the text in a certain way or persuading them to
arrive at a certain interpretation of it.

Visual and Material Features of Written Communication

The relevance of visual and material features of written communication has

recently been emphasised in both literary and discourse linguistic scholarship.
In the introductory chapter to their edited volume Communicating Early Eng-
lish Manuscripts, Andreas H. Jucker and Päivi Pahta note that manuscripts
simultaneously display material evidence of communication in the past and
communicate to the modern, scholarly reader.2 They point out, however, that
alongside actors such as the ‘author’ and the ‘reader’, other kinds of producers
and audiences could also play a role in delivering and receiving the messages
within manuscripts.3 As a result, they maintain that a

2
A.H. JUCKER and P. PAHTA, “Communicating manuscripts: Authors, scribes, readers,
listeners and communicating characters”, in: Communicating Early English Manuscripts, ed. P.
PAHTA and A.H. JUCKER (Cambridge, 2011), pp. 3-10, at p. 3.
3
JUCKER and PAHTA, “Communicating manuscripts”, p. 3.
4 MARI-LIISA VARILA et al.

pragmatic analysis that wants to investigate the communicative history of a manu-

script or a range of manuscripts must spell out in detail the roles of all the partici-
pants in the communicative situation created by the manuscript or, where socio-
historical facts are lacking, at least consider the possible alternative scenarios.4

Such detailed contextualisation is obviously very difficult to achieve much of

the time, as little may be known of any or all of the participants. Instead of
participant roles, James Daybell and Peter Hinds call attention to the immedi-
ate material and spatial contexts of written communication. They note that

physical books have their own what might be termed bibliographic rhetoric; letters
and other manuscripts contain social signals that are textually embedded within
material forms, such as handwriting and layout.5

They call for an integration of material readings with other approaches, such as
historical, textual and theoretical analyses.6
The challenge of establishing best practices of description and analysis
partly arises from the multiplicity of forms of early written communication. For
example, while there is currently a broad consensus on the importance of
palaeographical and codicological evidence for the literary analysis of early
texts, the best practices of this interdisciplinary field are yet to be established,
“particularly when we begin thinking about whole manuscripts rather than
individual texts”.7 The visual features in a given physical object can be dis-
cussed on various levels, from individual elements to a ‘programme’ covering
the whole production unit (for example book or gathering). We may decide to
focus on a physical object defined by its boundaries, and thus describe and
analyse the visual and material features of a ‘book’ or a ‘letter’. The bound-
aries may also be smaller, for example when analysing a ‘page’, or defined by
another type of physical object containing text, such as a runestone. The object
of study may also be defined by textual boundaries, for instance when focusing
on a certain ‘novel’ or ‘text’. The physical and the textual may coincide (for
example, in a letter written on a single sheet of paper), but this is often not the
4
JUCKER and PAHTA, “Communicating manuscripts”, pp. 3-4.
5
J. DAYBELL and P. HINDS, “Introduction: Material matters”, in: Material Readings of
Early Modern Culture: Texts and Social Practices, 1580-1730, ed. J. DAYBELL and P. HINDS
(Basingstoke, 2010), pp. 1-20, at p. 3.
6
DAYBELL and HINDS, “Introduction”, p. 15.
7
A.W. BAHR, “Reading codicological form in John Gower’s Trentham manuscript”,
Studies in the Age of Chaucer 33 (2011), pp. 219-262, at p. 219.
Disciplinary Decoding 5

case with early writing. For example, one physical volume may contain several
texts, and one text may be bound as several physical volumes. Furthermore,
these different units may have been produced over an extended period of time
and by several people. Text and book producers seldom explain their visual
and material choices to the reader.
A division may be drawn between studying what might be called higher-
level elements (for instance mise-en-page or layout) and lower-level elements
(for instance scripts and typography). While these two levels are intricately
intertwined and sometimes discussed by the same scholars (as is seen below),
a focus on one or the other may bring to the fore different aspects of written
communication. This basic division into communication through layout and
communication through scripts / typography provides the basis for the structure
of the present volume, and it also acts as the starting point of the following
discussion.

Communicating through Layout

One of the most influential contributions in the field of manuscript studies

addressing the organisation of the page is Malcolm Parkes’s study of the inter-
action between mise-en-page and information structuring.8 Parkes discusses the
concepts of ordinatio (the organisation of material) and compilatio (the process
of combining and rearranging material from different sources) and the interac-
tion between these two textual and material practices. The changing reading
practices in the twelfth century led to major developments in how text was
presented to the reader on the manuscript page.9 While monastic reading prac-
tices were linear and meditative in nature, the scholastic reader benefited from
an apparatus of glosses and references – much like the modern academic
reader.10 This requirement for scholarly apparatus led to a stabilisation of the
presentation of various elements on the page, such as supplying the main text
in a larger and more formal script and the commentary in a smaller script,
giving the source references in the margin, and using red ink to highlight the

8
M.B. PARKES, “The influence of the concepts of ordinatio and compilatio on the
development of the book”, in: Medieval Learning and Literature: Essays Presented to Richard
William Hunt, ed. J.J.G. ALEXANDER and M.T. GIBSON (Oxford, 1976), pp. 115-141 and Plates
IX-XVI.
9
PARKES, “The influence”, pp. 115-116.
10
PARKES, “The influence”, p. 115.
6 MARI-LIISA VARILA et al.

names of auctores.11 Parkes notes that these developments initially appear

sporadically, showing “first that readers felt the need for more ostensible help
in finding their way about in a highly sophisticated and technical argument,
and secondly that the producers of books had not yet developed a recognized
procedure for coping with this problem”.12 By the late Middle Ages, however,
the presence of at least some elements of the originally scholarly apparatus
came to be expected by readers and could be supplied by them if needed.13
Indeed, as Bonnie Mak observes, “the rich variations of the medieval mise-en-
page indicate that the interface was understood and exploited as a field of
engagement that could be reconfigured as needed”. 14
The term mise-en-page, as used for example by Parkes and Mak, refers to
the overall arrangement of the verbal and visual elements on a page.15 Since the
term retains a somewhat technical air in English, many researchers prefer the
more transparent ‘layout’ instead.16 While layout may thus conveniently be
viewed as a synonym for mise-en-page, narrower and wider definitions are also
possible. In her study of the “communicative potential” of layout, Tjamke
Snijders, for example, prefers a highly comprehensive definition which sub-
sumes the ways in which the material text “has been subdivided [...], illumi-
nated, and written down”.17 While palaeographic ‘micro features’ of script and

11
PARKES, “The influence”, p. 116.
12
PARKES, “The influence”, p. 118.
13
PARKES, “The influence”, p. 135.
14
B. MAK, How the Page Matters (Toronto, 2011), p. 16.
15
D. MUZERELLE, Vocabulaire codicologique: Répertoire méthodique des termes français
relatifs aux manuscrits. (Paris, 1985), p. 109, defines the term as “Disposition générale des
différents éléments figurant sur une page” (“General arrangement of the various elements on a
page”).
16
As done, for example, by A. DEROLEZ, The Palaeography of Gothic Manuscript Books:
From the Twelfth to the Early Sixteenth Century (Cambridge, 2003), pp. 34-39, and J.P.
GUMBERT, “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 and New Castle, DE,
1999), 7-13. M.P. BROWN, Understanding Illuminated Manuscripts: A Guide to Technical Terms
(Malibu, CA, and London, 1995), p. 86, glosses mise-en-page as a “term [that] refers to the layout
of the page”. The multilingual Codicologia term bank maintained by l’Institut de Recherche et
d’Histoire des Textes (IRHT, available at http://codicologia.irht.cnrs.fr/) offers page design and
lay-out as the English equivalents of the French headword Mise en page. The Italian equivalents
listed include terms such as organizzazione della pagina and architettura della pagine. For the
use of the architecture metaphor in English discussion of layout, see the section ‘Towards
managing visual and material features in theory and practice’ below.
17
T. SNIJDERS, Manuscript Communication: Visual and Textual Mechanics of Communi-
cation in Hagiographical Texts from the Southern Low Countries, 900-1200 (Turnhout, 2015:
Disciplinary Decoding 7

individual letter forms are not included (cf. our division between layout and
script / typography), Snijders’s inventory of the elements which constitute
layout otherwise comprises a wide range of verbal, visual and material fea-
tures: the codicological structure of the book, the dimensions of the page, the
relative amount of text (vs. blank space) on the page, the use of colour, titles
and rubrics, types of initials and their dimensions and shape, punctuation, ta-
bles of contents (and potentially other such paratextual elements), illustrations
of various kind, and the nature and quality of the writing support.18 Almost all
of these elements of layout also feature in chapters of the present volume.
The interaction between information content and its presentation is also
central for Guyda Armstrong, who applies the term ‘information design’ to
early modern printed translations. She points out that while various fields –
including graphic design, multimodal studies, translation studies, and
“material-textual studies” – address the forms and functions of translated books
of the early modern period, these approaches have not been systematically
brought together.19 Attempting to bridge this gap, she employs terminology
borrowed from translation studies on visual and physical features of early
modern print.20 Armstrong focuses on late sixteenth-century English transla-
tions of material of Italian origin, noting that “it is striking how far their diver-
gent translational purposes or functions are expressed in their material forms,
codified in their layout, paratexts, format, and / or language(s) of production”.21
For example, a bilingual page utilising a parallel-column layout to distinguish
the texts in different languages may be described in terms of translational ‘equiv-
alence’; Armstrong notes that such a layout essentially serves as a table, show-
ing the correspondences between the two texts.22
A key term appearing above is ‘multimodality’. Multimodal analysis “ap-
proaches representation, communication and interaction as something more
than language”; writing is thus only one of the communicational and semiotic
meaning-making tools explored alongside image, gesture, speech and other

Utrecht Studies in Medieval Literacy 32), p. 429. Instead of ‘material text’, Snijders prefers the
term scriptum: “[t]he material presentation of an internally coherent narrative unit [...]
conceptualised as a physically unique entity that combines text, layout, illumination and other
material considerations” (ibid., p. 431).
18
SNIJDERS, Manuscript Communication, pp. 48-78.
19
G. ARMSTRONG, “Coding continental: Information design in sixteenth-century English
vernacular manuals and translations”, Renaissance Studies 29 (2015), pp. 78-102, at p. 78.
20
ARMSTRONG, “Coding continental”, pp. 79-81.
21
ARMSTRONG, “Coding continental”, p. 82.
22
ARMSTRONG, “Coding continental”, pp. 86-87.
8 MARI-LIISA VARILA et al.

modes.23 The multimodal approach could thus lend itself to a discussion of

both layout and typography / script. Although multimodal studies typically
focus on present-day material, they offer a useful perspective for studying early
written communication as well. The multimodal approach is utilised, for exam-
ple, by Theo van Leeuwen, who has worked on the semiotics of visual systems
such as colour and typography.24 His understanding of these systems is based
on Jakobson and Halle’s theory of distinctive features (as in describing the
phonemes of English), with the adjustment that the distinctive features
(whether of sounds, letter forms or colours) are considered to have some mean-
ing potential in themselves, through connotation and experiential metaphor.25
He argues that this allows ideational, interpersonal, and textual meanings to be
transmitted through colour.26 For instance, the colour white has in some cul-
tures long been a symbol of purity, and such ideational, symbolic meanings
were commonly used already by medieval artists. The use of red colour as an
attention-catching device would serve an interpersonal function, while using a
particular colour to mark headings would be an example of the textual func-
tion. Of course, any use of colour will function on all three levels simulta-
neously.27 Similarly, a typeface can be used to suggest ideas of modernity and
rebelliousness, in addition to the obvious textual functions.28 As individual
colours are subjective to a degree, Van Leeuwen advocates examining colour
schemes: the ways in which a combination of colours is used to convey mean-
ing.29 Indeed, this context-dependency applies to many visual features: accord-
ing to Alexander H. Wertheim, “one can never derive the conspicuity of an
object from its features. Instead one needs to focus on how the perceptual sys-
tem deals with the interrelation between object, environment and viewing con-

23
C. JEWITT, “Introduction: Handbook rationale, scope and structure”, in: The Routledge
Handbook of Multimodal Analysis, ed. C. JEWITT (Abingdon, [2009] 2011), pp. 1-7, at p. 7.
24
G. KRESS and T. VAN LEEUWEN, “Colour as a semiotic mode: Notes for a grammar of
colour”, Visual Communication 1 (2002), pp. 343-368; T. VAN LEEUWEN, “Typographic mean-
ing”, Visual Communication 4 (2005), pp. 137-43; T. VAN LEEUWEN, “Towards a semiotics of
typography”, Information Design Journal 14 (2006), pp. 139-155; T. VAN LEEUWEN, The Lan-
guage of Colour: An Introduction (London, 2011).
25
VAN LEEUWEN, The Language of Colour, pp. 57-58.
26
VAN LEEUWEN, The Language of Colour, pp. 10-12.
27
VAN LEEUWEN, The Language of Colour, p. 12.
28
VAN LEEUWEN, “Typographic meaning”, p. 140; VAN LEEUWEN, “Towards a semiotics
of typography”, p. 143.
29
VAN LEEUWEN, The Language of Colour, pp. 65-67.
Disciplinary Decoding 9

ditions”.30 We return to the relationship between conspicuity and context in the

section ‘Communicating through script and typography’.
John Bateman’s Genre and Multimodality (GeM) model is one approach
which can be used for analysing the visual features of written communication
from a multimodal perspective.31 It is an attempt to bring analytical rigour to
the study of multimodal documents: while some early analyses were satisfied
with providing a possible explanation for visual features, Bateman aims to
develop an empirical system that produces falsifiable results. For instance, he
proposes reducing the resolution of an image as an objective method of locat-
ing central elements on the page, without conflating the visual data with the
eventual function of elements on the page.32 The GeM model can be used to
analyse elements on four layers. These are the base layer (1), layout layer (2),
rhetorical layer (3), and navigational layer (4). Regarding (1) the base layer,
Bateman lists base units, the smallest building blocks of the page, which in-
clude, for example, titles, running heads, sentences, and page numbers.33 (2)
The layout layer is seen as consisting of three parts: layout segmentation, reali-
sation information, and layout structure. First, layout segmentation involves
identifying perceptually salient elements on the page, such as paragraphs or
images. Secondly, each of these units has a verbal or graphical realisation
which can be described further. Thirdly, the layout units are grouped into
larger chunks which form the page structure (presented as a tree structure with
the layout units as nodes, and complemented with an area model representing
the spatial divisions on the page).34 (3) The rhetorical layer is related to the
relationship between different sections of the document (verbal and visual) and
how these sections work as a whole.35 Finally, (4) the navigational layer, which
Bateman does not discuss thoroughly, can be said to comprise elements direct-
ing the reader within the document, for example pagination, indices and refer-
ences to other locations in the document. Tuomo Hiippala points out that due
to their subtle and often implicit nature, navigation structures may easily go

30
A.H. WERTHEIM, “Visual conspicuity: A new simple standard, its reliability, validity and
applicability”, Ergonomics 53 (2010), pp. 421-442, at p. 422.
31
J.A. BATEMAN, Multimodality and Genre: A Foundation for the Systematic Analysis of
Multimodal Documents (London, 2008). The GeM model has been applied by, for example, T.
HIIPPALA, The Structure of Multimodal Documents: An Empirical Approach (New York, 2015).
32
BATEMAN, Multimodality and Genre, pp. 67-69.
33
BATEMAN, Multimodality and Genre, p. 211.
34
BATEMAN, Multimodality and Genre, pp. 115-126.
35
BATEMAN, Multimodality and Genre, p. 144.
10 MARI-LIISA VARILA et al.

unnoticed – despite their importance to the reader interacting with the multi-
modal artefact.36
The ‘page’ is brought into focus as an interface for communication by Bon-
nie Mak in her study of the fifteenth-century treatise Controversia de nobilita-
te. Mak notes that while the page has long remained a basic unit of “graphic
communication of ideas” despite the changing methods of text production, its
role in the transmission of knowledge has not been adequately appreciated.37
She maintains that the “page is an expressive space for text, space, and image;
it is a cultural artefact; it is a technological device. But it is also all of these at
once”.38 Similarly, Gregg De Young sees the combination of blank space, text,
and visual aids (such as diagrams) on the page as a kind of ‘architecture’ of
manuscript and early print material.39 Mak further points out that in addition to
the linguistic message on the page, the letterforms and their structuring (includ-
ing the use of blank space), images, and decoration also carry communicative
value.40 Even in cases where the textual message itself remains relatively sta-
ble, changes in its material and visual presentation may considerably influence
the interpretation of that message by different readers and audiences, as shown
by Mak.41
The connection between layout and information structure has important
implications. The linguistic and textual measures taken by text producers to
ensure that their readers understand their message can be complemented – or
contradicted – by visual and material factors. For example, the visual pro-
gramme (layout, decoration and illustration) in a given manuscript may work
as a cohesive and structural device alongside textual signals such as chapter
headings or meta-textual commentary. There is naturally overlap between the
categories of linguistic and visual signalling of text structure. A chapter title
supplied in a larger script or font than the body text is an example of a combi-
nation of linguistic and visual marking of a textual boundary. Furthermore, in
addition to page design or layout, the lower-level visual elements on the page
– which have already appeared in some of the research cited above – also influ-
ence the reading of a given text.

36
HIIPPALA, The Structure of Multimodal Documents, pp. 52-53.
37
MAK, How the Page Matters, p. 8.
38
MAK, How the Page Matters, p. 18.
39
G. DE YOUNG, “Mathematical diagrams from manuscript to print: Examples from the
Arabic Euclidean transmission”, Synthese 186 (2012), pp. 21-54, at p. 21.
40
MAK, How the Page Matters, p. 17.
41
MAK, How the Page Matters, p. 20.
Disciplinary Decoding 11

Communicating through Script and Typography

Similarly to layout or mise-en-page, individual visual elements on the page

may also carry communicative value simultaneously with, in addition to, or
even in contrast to the verbal elements. The second part of this volume consists
of contributions exploring the relationship between language and script or
typography. Both manuscript and print material are examined in several of the
chapters of this volume, considering the influence of technology on communi-
cating through the letterforms on the page (see further the final section of this
chapter).
Guyda Armstrong suggests that the intended function and reading practices
related to an early modern book are signalled through format and layout, while
what she calls “linguistic difference” is indicated by typography – for example
when a roman font is used for Latin text and black letter for English.42 The
scribal and typographic distinctions related to language may also be made on
a level lower than that of whole texts. Armstrong views the use of certain type-
faces for certain languages as a “translational” visual practice.43 Tim William
Machan employs the term ‘visual pragmatics’ in his analysis of such scribal
choices in late medieval manuscripts.44 He seems to use this term in reference
to the practical considerations of designing a manuscript containing multilin-
gual literature, not as a label for a specific field of inquiry. Machan sees the
text producers’ visual decisions as communicative acts, constrained by contex-
tual factors such as the cost of production.45 He identifies only two means of
marking code-switching (the use of two or more languages in the same piece of
communication) in modern novels and newspapers - the use of italics or quota-
tion marks (if code-switches are visually marked at all) - whereas in manu-
scripts of late Middle English literature, there is a spectrum of scribal treatment
of code-switching from “non-recognition” to “a consistent graphic design that
visually emphasizes moments where a text changes languages”.46
Visual emphasis may be achieved by using a different ink or script, by
decoration, or by modifying text size or placement on the page, in the material

42
ARMSTRONG, “Coding continental”, p. 102.
43
ARMSTRONG, “Coding continental”, pp. 87-89.
44
T.W. MACHAN, “The visual pragmatics of code-switching in late Middle English
literature”, in: Code-Switching in Early English, ed. H. SCHENDL and L. WRIGHT (Berlin, 2011),
pp. 303-333.
45
MACHAN, “The visual pragmatics”, pp. 325-326.
46
MACHAN, “The visual pragmatics”, pp. 304, 310.
12 MARI-LIISA VARILA et al.

examined by Machan.47 Ruth Carroll et al. similarly recognise four variables

which may be altered to provide visual highlighting: size, style, colour and
position.48 The means listed are almost identical, but the term used of the end
result of applying these means varies. Ink changes and other modifications can
be described, for example, as emphasis or highlighting. The usage varies from
scholar to scholar, as does the precision with which the relevant terms are
defined. Lou Burnard notes that in the Text Encoding Initiative (TEI), the term
‘highlighted’ is employed to describe “any form of visual salience”, including,
for instance, the use of a typeface that differs from its surroundings.49 Bur-
nard’s formulation draws attention to the importance of the ‘context’ of the
textual utterance (“its surroundings”) for judging what is highlighted. Explicat-
ing the concept of ‘emphasis’, which is used for example in rhetoric and stylist-
ics to describe the addition of some implied, extra meaning to what is stated
explicitly, Gerda Eva Lauerbach points out that emphasis essentially “works as
contrast against a local norm and against a background of unstated assumptions
regarding linguistic, social, and cultural norms, beliefs, opinions, and values”.50
For emphasis to work, producers and recipients of texts need to be aware of
what such norms, beliefs, opinions, and values are and how they operate on
both micro and macro levels of context. An early example of instructions for
typographic emphasis, acknowledging the importance of context, is found in a
manual on “print letters” by Joseph Moxon. He advises one designing a title
for an inscription to “consider the words of emphasis, and make those words to
vary from the Letter your Discourse is in, as either Roman, Italick, or English
[black letter], according as the words may properly require”.51
The concept of ‘foregrounding’, as used in literary stylistics, overlaps to
some extent with highlighting and emphasis, although its usual application is
more constrained. The concept was developed by Russian Formalists and mem-

47
MACHAN, “The visual pragmatics”, pp. 310-311.
48
R. CARROLL, M. PEIKOLA, H. SALMI, M.-L. VARILA, J. SKAFFARI and R. HILTUNEN,
“Pragmatics on the page: Visual text in late medieval English books”, European Journal of
English Studies 17 (2013), pp. 54-71, at p. 57.
49
L. BURNARD, What is the Text Encoding Initiative? How to Add Intelligent Markup to
Digital Resources, new edn. (online) (Marseille, 2014) <http://books.openedition.org/oep/426>.
50
G.E. LAUERBACH, s.v. “Emphasis” in: Handbook of Pragmatics Online, ed. J.-O. ÖSTMAN
and J. VERSCHUEREN (Amsterdam, 2002-2015), DOI: 10.1075/hop.8.emp1.
51
J. MOXON, Regulae trium ordinum literarum typographicarum, or, The rules of the three
orders of print letters viz. the Roman, Italick, English capitals and small, printed for J. MOXON
(London, 1676), p. 10-11, WING M3019 (accessed through Early English Books Online).
Disciplinary Decoding 13

bers of the Prague School in particular for the analysis of literary language.52
Mick Short describes foregrounding as the psychological effect of a part of a
text (or a work of art in general) becoming “especially noticeable, or perceptu-
ally prominent”, adding that the main methods of foregrounding include devia-
tion, parallelism and repetition.53 Geoffrey Leech and Mick Short provide a
more fine-grained distinction by discussing three separate notions of saliency:
‘deviance’, ‘prominence’, and ‘relevance’ / ‘foregrounding’.54 ‘Qualitative de-
viances’ are forms which go against a particular rule of a language (for in-
stance grammatically incorrect forms, nonsensical forms) while ‘quantitative
deviances’ are statistically unexpected forms. ‘Prominence’ refers to psycho-
logical saliency of deviant forms, and ‘foregrounding’ itself consists in adapt-
ing prominence to literary ends. In other words, not all deviances (or parallel-
isms or repetitions) create prominence, and not all prominent features have any
literary function. Of special relevance for the present discussion is ‘grapho-
logical foregrounding’ in its various forms, such as layout, capitalisation,
italicisation, deviant spelling and punctuation.55 Although these forms are
analysed within stylistics as creating literary effects, it is possible to adopt a
broader understanding of the concept of foregrounding to cover instances of
forms which are not necessarily deviant in the strict sense of this term, al-
though they are salient, and which serve a particular discourse function.
As regards the motivation for using visually salient forms, Burnard remarks
that if the meaning of visual highlighting is extended to cover passages marked
as quotations,

the list of motivations for highlighting expands to include passages of direct

speech, material cited or quoted from elsewhere, or which the writer wishes to
indicate as being in some sense non-authorial, words which are being talked about
rather than used.56

52
See, for example, the papers by Havránek and Mukaøovský in: A Prague School Reader
on Esthetics, Literary Structure, and Style, ed. and trans. P. GARVIN (Washington, DC, 1964).
53
M. SHORT, Exploring the Language of Poems, Plays and Prose (Harlow, 1996), pp. 11,
12-15.
54
G. LEECH and M. SHORT, Style in Fiction: A Linguistic Introduction to English Fictional
Prose (Harlow, 2007), pp. 39-41.
55
E.g. M. SHORT, “Graphological deviation, style variation and point of view in Marabou
Stork Nightmares by Irvine Welsh”, Journal of Literary Studies 15,.3-4 (1999), pp. 305-323. Cf.
A. GIBBONS and S. WHITELEY, Style and Cognition: An Introduction (Edinburgh, forthcoming),
chapter 1: Foregrounding (pre-publication draft shared by authors).
56
BURNARD, What is the Text Encoding Initiative?
14 MARI-LIISA VARILA et al.

Machan also acknowledges that visual marking of quotations in another lan-

guage is likely to have a “rhetorical” – or discourse-organising – motivation:
such quotations in late medieval manuscripts may have been given visual em-
phasis because they represent a different type of passage or level of discourse,
not primarily because they are in another language.57

Towards Managing Visual and Material Features in Theory and Practice

The discussion above covers a range of approaches to how the simulta-

neous effect of text and its visual and material context can be examined. A key
notion shared by them is that choices made in layout or script / typography
render particular verbal parts of the whole more visible (salient, conspicuous,
foregrounded, emphasised, highlighted), which prompts the reader to under-
stand the text in the intended manner. The approaches differ in the choice of
terms, often used metaphorically in the designations. When visual and material
features are analysed from a textual or linguistic viewpoint using language-
oriented metaphors such as (visual) ‘grammar’, ‘prosody’, or ‘rhetoric’, the
emphasis is on the communicative function of the elements – the message on
the page. The use of visually oriented metaphors – ‘architecture’, for instance
– seems to place more weight on the visual appearance or design of the page.
The metaphors may also suggest a scope for the object of study: ‘architecture’
and ‘rhetoric’ imply perhaps a wider coverage or a higher-level perspective
than ‘prosody’. Researchers drawing from different fields and asking different
questions about their primary material may thus be inclined to employ different
types of metaphors when constructing models for visual and material features.
The metaphor of rhetoric, for example, can be applied successfully on
multiple levels. It is possible to use ‘visual rhetoric’ as a metaphor in order to
explain on a general level what is being communicated by the visual presenta-
tion of a manuscript or printed book, but the metaphor could also be extended
and used more specifically.58 Although the individual figures of speech distin-
guished in classical rhetoric may be strictly verbal, the three modes of persua-
sion described by Aristotle can perhaps be seen in action also in the visual

57
MACHAN, “The visual pragmatics”, p. 327.
58
Cf. R. BARTHES, “Rhetoric of the image”, in: ID., Image–Music–Text, selected and trans.
S. HEATH (New York, 1977), pp. 32-51.
Disciplinary Decoding 15

presentation of a book.59 The first (and most important) mode of persuasion,

according to Aristotle, was ethos: persuasion based on the character of the
speaker. A book proudly flourishing the arms of a well-known university or a
famous printer’s device on its cover (or another prominent location) probably
exemplifies ethos; the likely aim is to remind the reader that the producers
know their business, the implication being that the contents will be of equally
high quality. While the first mode of persuasion focuses on the speaker, the
second mode of persuasion, pathos, is based on stirring the emotions of the
hearer. Images are perhaps the most obvious visual means of achieving this.
For instance, the pictures depicting events from the life of Christ in a late medi-
eval Book of Hours could stir devotional emotions in readers, regardless of
whether they would understand its Latin text.60 The third mode of persuasion
is argumentation itself, or logos, which is perhaps more a question of contents
than of presentation. However, there are visual means of foregrounding the
argumentation: tables are a good example of this, since they are intended to
present the most important information in condensed form.
While visual elements of text can be seen as “configuring the reading expe-
rience in fundamental ways”, they may not be available to those reading the
text as a scholarly edition or accessing it through an electronic corpus.61 The
extent to which such elements are considered worth transmitting to the readers
of modern scholarly editions varies. In his chapter on transcription for elec-
tronic editing, Matthew J. Driscoll distinguishes between the structure of the
‘work’ being edited (for instance chapters, sections, and paragraphs) and the
structure of the ‘document’ (layout; “the arrangement of the text on the
page”).62 He notes that diplomatic transcriptions tend to follow the layout,
while less diplomatic transcriptions favour the structure of the work. As Dris-
coll argues, in early writing there is perhaps more often a discrepancy between
these two structures than in present-day writing.
Even when editions do not reproduce visual and material features of a
given work, these elements may convey important information concerning the

59
Aristotle, Art of Rhetoric I.2., 1356a, trans. J.H. FREESE (Cambridge, MA, 1926: Loeb
Classical Library 193), pp. 16-17.
60
See R.S. WIECK, Painted Prayers: The Book of Hours in Medieval and Renaissance Art
(New York, 2004), p. 22.
61
MAK, How the Page Matters, pp. 17-18.
62
M.J. DRISCOLL, “Levels of transcription”, in: Electronic Textual Editing, ed. L. BUR-
NARD, K. O’KEEFFE and J. UNSWORTH, (Modern Language Association of America, 2007),
<http://www.tei-c.org/About/Archive_new/ETE/Preview/driscoll.xml>.
16 MARI-LIISA VARILA et al.

language or text of these works. Such information will easily escape the reader
if the editor has not commented on the non-verbal elements in their introduc-
tion or description of editorial practices. For example, the size of the manu-
script page and the amount of available writing space therein may have con-
strained the copying practice of a scribe by making them increase the number
of abbreviations, use elliptical constructions or shorten the text by other
means.63 Carrie Griffin finds that the transmission of text in English instructive
and informative literature cannot be separated from its material context, as the
material form contributes to the meaning and reception of the message.64 The
same challenge also pertains to corpora. To counter it, Anneli Meurman-Solin
approaches the analysis of visual features of epistolary writing from the point
of view of annotation theory and corpus construction.65 She notes that the term
‘visual prosody’, modelled after phonetic prosody, captures the characteristics
of epistolary prose as interactive / dialogic, potentially idiosyncratic, and tak-
ing into account factors such as the writer’s style of producing utterances
(length and sequencing) and information structuring (including syntax).66
Meurman-Solin regards the following visual elements as linguistically relevant:
physical condition; number of leaves; line-break; position of text on the page;
change of scribal hand; script; idiosyncratic scribal features; insertions, correc-
tions and cancellations; punctuation and spacing; letterforms that are markedly
different (for instance considerably larger); and paragraph structure.67 She
argues that these features are especially important for studying syntactic and
discourse structure.68 Meurman-Solin maintains that to achieve a working tax-
onomy, we should, first, compile corpora of digital images of visual variants,

63
See M.-L. VARILA, In Search of Textual Boundaries: A Case Study on the Transmission
of Scientific Writing in 16th-Century England (Turku, 2016), p. 215.
64
C. GRIFFIN, “Instruction and information from manuscript to print: Some English
literature, 1400-1650”, Literature Compass 10 (2013), pp. 667-676, at pp. 669, 673. See also
VARILA, In Search of Textual Boundaries, pp. 330-331.
65
A. MEURMAN-SOLIN, “Visual prosody in manuscripts of letters in the study of syntax and
discourse”, in: Principles and Practices for the Digital Editing and Annotation of Diachronic
Data, ed. A. MEURMAN-SOLIN and J. TYRKKÖ (Helsinki, 2013) <http://www.helsinki.fi/varieng/
journal/volumes/14/meurman-solin_a/>; A. MEURMAN-SOLIN, “Taxonomisation of features of
visual prosody”, in: Principles and Practices for the Digital Editing and Annotation of
Diachronic Data, ed. A. MEURMAN-SOLIN and J. TYRKKÖ (Helsinki, 2013) <http://www.helsinki.
fi/varieng/series/volumes/14/meurman-solin_c/>.
66
MEURMAN-SOLIN, “Visual prosody”; cf. LEECH and SHORT, Style in Fiction, pp. 173-175
on the “rhythm of prose”.
67
MEURMAN-SOLIN, “Visual prosody”.
68
MEURMAN-SOLIN, “Taxonomisation”.
Disciplinary Decoding 17

and secondly, construct theoretically valid categorisation models which are

able to cope with the high level of variation instead of directly classifying
visual features as ‘marked’ or ‘unmarked’ – already an interpretation. An ele-
ment can in her model be described on different levels, for example in terms of
its shape, size and position. When these descriptions follow a (database-spe-
cific) systematic annotation scheme, they work as retrieval tools for further
research on the material. The dangers of the inevitable subjectivity of categori-
sation are lessened by transparent annotation practices.69 The visual aspect of
text is thus also a concern of corpus linguists.
Regardless of how one chooses to label them, visual and material ap-
proaches to early writing have become increasingly popular in the last several
years, and one reason for this must be the unprecedented access to older texts
through digital reproductions of primary materials. These resources help to
overcome the problem of missing visual information in editions and corpora,
and they are particularly valuable for researchers working far from major re-
positories of early writing. However, using a digital reproduction instead of the
original object is not without risk; digitisation cannot fully mediate the materi-
ality of the object – the material context is difficult to convey in 2D images.
With manuscripts, the codicological context is typically only given in a sepa-
rate manuscript description; and with printed books, most repositories only
provide images of a single copy of a given edition. The images available may
also be misleading in some way, for example due to poor resolution, distorted
dimensions or mistakes in image sequence. While the research opportunities
provided by the increasing number of digital resources should definitely be
utilised, there still remains a need for direct consultation and description of
primary materials as well. Although the two-dimensional space of the page is
explored by many, other types of material evidence, such as the writing support
and codicological evidence, are still relatively rarely brought to bear on dis-
course linguistic analyses. Ideally, then, ‘reading the page’ does not only in-
volve looking at the visual presentation of language and text, but considers the
material context as well. There are good grounds for continuing research in this
field.

69
MEURMAN-SOLIN, “Taxonomisation”.
18 MARI-LIISA VARILA et al.

The Chapters in This Volume

This volume has an inherently cross-disciplinary perspective. It is not only

situated against the theoretical backdrop provided by philology, discourse
linguistics and book studies, but also challenges traditional research bound-
aries. As noted at the beginning of this chapter, we cross technological bound-
aries by including research on both manuscript and printed materials. In fact,
several contributors explore both media and trace connections between them.
This approach is helpful for assessing the role played by technological con-
straints and economical aspects of book production in the interplay between
verbal and visual / material features. Moreover, we cross the division between
the medieval and the (early) modern. Importantly, the chapters investigating
early modern materials include several that focus primarily on manuscript
texts, which highlights continuity and challenges stereotypical period-medium
associations. The contributors also cross the boundary between public and
private communication, as the textual situations explored in the volume range
from mass-produced books to personal letters and even to inscriptions on ob-
jects. This rich array of materials helps us to consider the possible role played
by changing reading practices and commercial factors in the selection of visual
and material strategies. In what follows, we will briefly introduce the chapters
in the order in which they appear in this volume.
Colette Moore’s chapter evaluates the role of mise-en-page in discourse
organisation, linking manuscript studies with historical pragmatics and stylist-
ics. Her two case studies of Middle English saints’ lives show how scribal
variation in the linguistic and visual organisation between different manuscript
copies is critically linked to what she calls “hermeneutic or interpretational
shifts” in what these texts would convey to the reader. The case studies address
the representation of speech passages in the Life of St. Thaïs in the Northern
Homily Cycle and the use of paratextual elements (titles, headings, paraphs,
initials) in the life and miracle of St. Andrew in the South English Legendary.
Page design is also the focus of the chapter by Francesca L. Mackay, who
studies print and manuscript copies of Robert Lindesay of Pitscottie’s Cron-
icles of Scotland. She argues that an analysis of the mise-en-page sheds light
on the reading community for which a certain copy is intended, showing how
the copy is shaped to suit the literacy practices of that community. Her close
examination of the use of punctuation and paratextual features in three select
witnesses of the Cronicles, from the close of the sixteenth to the early eigh-
Disciplinary Decoding 19

teenth century, enables Mackay to place these copies at different points of the
continuum between intensive and extensive reading practices.
The chapter by Maura Ratia and Carla Suhr focuses on the design of the
title page in early modern printed medical texts. The authors argue that while
early modern title pages were intended to market the printer’s skills to the
audience, the title pages and titles simultaneously functioned as guides for
prospective readers, displaying the genre of the work and suggesting how to
interpret the work. This was achieved through the interplay of textual and
visual features. Ratia and Suhr explore how the various verbal elements on the
title page (labels, headlines) that describe the genre or the topic of the text are
given prominence by visual highlighting. They also discuss diachronic devel-
opments in these features during the period 1525-1700.
Jukka Tyrkkö’s contribution challenges the text-based approaches to early
writing, taking as its starting point the visual elements of early modern title
pages of surgical books. His quantitative analysis of the visual structure of the
title pages is conducted with the image processing tool ImageJ, originally cre-
ated for medical imaging. Tyrkkö’s case study introduces new empirical, data-
driven methods for approaching the visual and paratextual on the page based
on the computational analysis of large data sets of facsimile images. The image
analysis advocated by the author invites applications exploring cultural connec-
tions between visual, paratextual and linguistic features of early print produc-
tion.
In her chapter, Yin Liu challenges the conventional ideas that the primary
function of writing is always to convey information, and that this information
should by necessity be conveyed by the written text at its most abstract level.
She examines early English objects that ‘name themselves’ in runic script,
suggesting that the use of runes in objects such as swords or vessels may be
deliberately obscure. The otherness of the runes, limiting the readership of
these texts, may serve to increase the perceived value of these objects and
texts. The script itself may thus be more central in delivering the intended
message than the information codified in that script.
The chapter by Anya Adair explores the complex and dynamic relationship
between manuscript and print in the late fifteenth century by investigating the
founts used by William Caxton in the earliest stages of his career. She draws
attention to the considerable number of different designs for graphemes and
combinations of graphemes in Caxton’s Type 1 and Type 2 founts, asking what
function this variation served for Caxton and his audience. Adair argues that
20 MARI-LIISA VARILA et al.

while Caxton initially strived to employ founts with distinctively ‘English’

features and sorts that were intended to be used in certain orthographic con-
texts, the practicalities of print production eventually led to the use of a more
limited set of sort designs.
Samuli Kaislaniemi focuses on the visual marking of code-switching in
early modern manuscript letters and printed tracts. He studies both the linguis-
tic and visual dimensions of using script or typeface to differentiate between
languages, evaluating the implications of this script-switching or typeface-
switching and the extent to which it correlates with code-switching. Kaislanie-
mi’s first case study is based on a large collection of diplomatic letters written
by Richard Cocks, an English merchant operating in southern France in the
early 1600s. The second case study draws on coeval non-literary printed prose
texts from the Lampeter Corpus of Early English Tracts.
R.W. McConchie’s contribution similarly focuses on the relationship be-
tween typography and language. He inspects select dictionaries from the late
fifteenth century until the mid-eighteenth century, analysing the function and
use of different typefaces. McConchie finds that the growing extent of typo-
graphical sophistication during this period to differentiate visually between
specific textual elements of a dictionary entry tells of increasing readability
and accessibility from the perspective of the audience. He observes that the
solutions adopted by printers and lexicographers would vary to some extent
between learned and popular, classical and vernacular, English and continental,
as well as between poly- and monolingual dictionaries.
Epistolary writing is the focus of the chapter by Imogen Marcus. She notes
that while letters offer excellent evidence for studying linguistic change in the
early modern period, it is often difficult to establish whether the ‘author’ (the
signer) of the letter also wrote it in their own hand. Marcus draws attention to
the problem that early letters often have more than one producer, as it was
relatively common to employ a scribe to whom the contents would be dictated.
Based on her work on the personal correspondence of Elizabeth Talbot (Bess
of Hardwick), she proposes a method of scribal profiling that is sensitive to this
issue and may be used to differentiate between autograph and scribal writing
in letters.
 Part II

Communicating through Layout

 Discourse Variation, Mise-en-page, and Textual
Organisation in Middle English Saints’ Lives

COLETTE MOORE

ne of the primary pragmatic tasks of textual organisation is determining

O how we should understand the relation of some words to other words

and some groupings of words to others. Present-day conventions for
formal written English, for example, provide a series of more formulaic kinds
of written organisational marks (punctuation) and organisational words (dis-
course markers). How did these conventions emerge and develop? A critical
site for the development of strategies of written discourse organisation, I assert,
is the late medieval period, when written texts were proliferating and print had
not yet standardised these choices.
The conversation in historical English language studies about variation in
Middle English manuscripts has until recently been focused primarily on dia-
lectal differences: regional and social differences in orthography, syntax, and
phonology.1 These dialect studies have been critical to our understanding of

1
Research culminating in and building upon the scholarly atlas of late Middle English
texts: A. MCINTOSH, M.L. SAMUELS, and M. BENSKIN, A Linguistic Atlas of Late Mediaeval
English (Aberdeen, 1986).

......................................................................................................................................
Verbal and Visual Communication in Early English Texts, ed. Matti PEIKOLA, Aleksi
MÄKILÄHDE, Hanna SALMI, Mari-Liisa VARILA, and Janne SKAFFARI, Utrecht Studies in Medieval
Literacy, 37 (Turnhout: Brepols, 2017), pp. 23-40.

DOI 10.1484/M.USML-EB.5.114129
24 COLETTE MOORE

Middle English texts and manuscripts, their audiences and circulation, and
their interpretation. In more recent years, though, historical study of the Eng-
lish language has experienced ‘the pragmatic turn’ – increasing attention to
communicative contexts of language which includes the organisation of dis-
course: the way that speakers and communities structure their language and
their texts. Since most of the available linguistic evidence lies in written texts,
some part of the focus has necessarily been directed to aspects of organisation
connected with writing (even only to distinguish written methods of organisa-
tion from spoken ones). Just as the meanings of words are a collective agree-
ment between speech communities that is continually being renegotiated, so
are strategies of discourse organisation, and these fall under the purview of
historical pragmatics and stylistics. Historical pragmatics, moreover, has be-
come more engaged with the communicative importance of manuscripts: exam-
ining them as evidence for pragmatic structures and considering implications
of the written form for understanding genres and other communicative frame-
works.2
My methodology for this research, then, combines recent directions in
historical pragmatic and stylistic inquiry with manuscript study of the mise-en-
page (the layout on the page) to investigate variation in discourse organisation.
Although the vicissitudes of disciplinary divisions in the twentieth century
have split this conversation into different fields, I think that the combination of
these two perspectives creates a kind of binocular vision that provides depth to
our analyses. Recent scholarly conversations have begun to combine linguistic
perspectives with textual analysis of the written page in ways that are quite
promising.3 Combining these approaches also provides helpful perspectives for
2
A.H. JUCKER and P. PAHTA, “Communicating manuscripts: Authors, scribes, readers,
listeners and communicating characters”, in: Communicating Early English Manuscripts, ed. A.H.
JUCKER and P. PAHTA (Cambridge, 2011), pp. 2-10.
3
T.W. MACHAN, “The visual pragmatics of code-switching in Late Middle English litera-
ture”, in: Code-Switching in Early English, ed. H. SCHENDL and L. WRIGHT (Berlin, 2011), pp.
303-333; C. SUHR, Publishing for the Masses: Early Modern English Witchcraft Pamphlets
(Helsinki, 2011); S. PARTRIDGE, “Designing the page”, in: The Production of Books in England
1350-1500, ed. A. GILLESPIE and D. WAKELIN (Cambridge, 2011), pp. 79-103; R. CARROLL, M.
PEIKOLA, H. SALMI, M. VARILA, J. SKAFFARI, and R. HILTUNEN, “Pragmatics on the page: Visual
text in Late Medieval English books”, European Journal of English Studies 17.1 (2013), pp. 54-
71; M. PEIKOLA, “Guidelines for consumption: Scribal ruling patterns and designing the mise-en-
page in later medieval England”, in: Manuscripts and Printed Books in Europe 1350-1500:
Packaging, presentation and consumption, ed. E. CAYLEY and S. POWELL (Liverpool, 2013), pp.
914-931; H. SALMI, “Visual presentation of English Body and Soul Debates”, in: Manuscript
Studies and Codicology: Theory and Practice, ed. M. PEIKOLA and M. KYTÖ (= Studia
Discourse Variation 25

scholars approaching the text from literary and cultural angles. Variation in
scribal features of the mise-en-page, in fact, teaches us about hermeneutic or
interpretational shifts: how texts organise material for readers, and how aspects
of ordinatio (organisation on the manuscript page) are critical to pursuing
interpretational questions of the relation of passages within a text – whether
that is a smaller narrative switch from one speaker to another or a larger shift
from one subsection of a text to another. This chapter presents a frame for
considering changes in textual organisation and then examines two illustrative
case studies from Middle English saints’ lives: one that pursues manuscript
variation in the organising structure of speech representation in the life of St.
Thaïs and the other that considers the mise-en-page variation of paratextual
structure in the titles and headings of the life of St. Andrew. It is by consider-
ing variation in linguistic and visual methods of structuring together that we get
a fuller picture of manuscripts organising discourse.

Organisational Features

One complexity to investigating textual organisation lies in its multi-

valency: the words, marks, and layout combine to assist in interpretation. An
example of this can be found in a kind of organising feature that I have dis-
cussed elsewhere: the use of red ink in rubrication.4

(1) San Marino, Huntington Library, HM 149, f31r

lorde Jhc) amonge oþ went to John & preyed him þat he

wolde bapti•e hym wiþ oþ & John by holdynge hym & knowyng
hym in •irite was a dredde & wiþ grete reu ence •eyde ¶ a
lorde I shulde be bapti•ed of þe & þou come•t to me and
Jhc) an•wered •uffr nowe for þus yt falleþ & by •emeþ vs
to fulfille alle ry¥twy•nes ¶ as who •eye not þis nowe

Neophilologica: A Journal of Germanic and Romance Languages and Literature 86, Supplement
1 (2014), pp. 144-156; J. KOPACZYK, “In search of a framework for written code-switching:
Administrative multilingualism on the page in early modern Poland”, in preparation.
4
C. MOORE, “Visual pragmatics: Speech presentation and Middle English manuscripts”,
in: The Cambridge Handbook of English Historical Linguistics, ed. M. KYTÖ and P. PAHTA
(Cambridge, 2016), pp. 481-496, at p. 495.
26 COLETTE MOORE

[bold indicates red letters in the manuscript]

Lord Jesus Christ among others went to John and prayed to him that he
would baptise him with the others and John by holding him and knowing
him in spirit was filled with awe and with great reverence said, “Oh
Lord I should be baptised by you and yet you come to me”, and
Jesus Christ answered, “Do this now for thus it falls and beseems us
to fulfil all righteousness as who say not this now ...”.

This passage from Nicholas Love’s The Mirror of the Blessed Life of Jesus
Christ exemplifies how the red ink might develop double functionality. The red
is an auctoritas marking (it alerts us to words that derive from an authoritative
source: in this case, scripture), but it also corresponds to a shift to direct
speech: the ink colour changes at the onset of the words of John the Baptist and
changes back after the response from Jesus. At different points in the manu-
script (and in other manuscript witnesses for Love’s Mirror), then, the rubrica-
tion marks auctoritas and sometimes quoted direct speech and sometimes both,
and texts often provide muddled impressions as to whether the rubrication flags
quotation in the sense of direct discourse or quotation in the sense of scriptural
precedent. This is a not a unique functional overlap for speech representation,
I should say; compare, for example, the way that the verb say can mark direct
or indirect speech (as in, he said, “I’ll be there”) or quoted written discourse
(an expression like “Shakespeare says ‘brevity is the soul of wit’” – in which
“Shakespeare says” means “Shakespeare wrote”) or even become generalised
as a broader authoritative marker (“they say you should eat lots of leafy
greens” – where “they say” merely indicates generalised authority). Such a
functional shiftiness characterises the usage of rubrication. Carroll et al. re-
mind us that there is not always a one-to-one correspondence between form
and function; to examine pragmatics, therefore, is to investigate how this rela-
tionship between form and function is negotiated.5
When a word develops two or more meanings we call this polysemy. This
ambiguous functionality, then, serves as a ‘pragmatic polysemy’ of sorts: a
multifunctionality in the use of rubrication as a tool for ordinatio. I am positing
that such multifunctionality grows out of the cultural production of texts in the
late medieval period: an increase in manuscript production together with an
increasing sophistication in strategies of textual organisation that have not yet
been thinned out through the standardising of print conventions. What we see
5
CARROLL et al., “Pragmatics on the page”, pp. 11-12.
Discourse Variation 27

in late Middle English manuscripts, then, is a proliferation of strategies for

flagging different aspects of discourse.
The later process of narrowing – winnowing down the competing strategies
for methods of discourse organisation – might have similarities to the processes
of standardisation in language. Einar Haugen describes different subprocesses
for standardising language – two formal processes that he calls selection (the
choice of a particular variety of the language) and codification (the setting
down of usage norms and conventions in reference texts such as dictionaries
and grammars and the ensuing prescription involved in enforcing these as-
pects), and two accompanying functional processes that he calls acceptance
(how a culture or speech community digests and disseminates these practices),
and elaboration (how these forms are enriched through different registers of
language usage).6
The development of normative patterns, widespread conventions, and
finally standards in organisational practice might both replicate processes with-
in this model and contribute to processes of standardising in the language as a
whole. Late Middle English organising conventions are local products of scri-
bal practices and are employed inconsistently (even within the same manu-
script), but there is shared practice through lines of textual descent and shared
practice between scribes trained in similar locations. So, for example, Matti
Peikola examines practices of line ruling and shows that Wycliffite scribes
tended to share a particular pattern of ruling – what he characterises as a Type
A pattern of line ruling. He suggests that the high frequency of this pattern of
ruling could be evidence for an incipient production standard: a shared set of
practices in ruling that become convention and then standard.7 As these kinds
of shared organising practices are disseminated, therefore, in ways similar to
letter forms or dialectal forms, they become characteristic of different kinds of
local and supralocal scribal practice. Eventually, after the introduction of print,
we see codification and elaboration of organising practice – in features like
punctuation, for example. Perhaps one could speak of a preliminary phase to
Haugen’s model: a ‘proliferation’ that precedes selection. We might see a
contemporary analogous site in the new kinds of organisational practices that
are being created for texts in digital media: the structuring of different elec-

6
E. HAUGEN, “Dialect, language, nation”, in: The Ecology of Language: Essays by Einar
Haugen, ed. A.S. DIL (Stanford, 1966 [1972]), pp. 237-254, and ID., “National and international
languages”, in: The Ecology of Language, pp. 255-264.
7
PEIKOLA, “Guidelines for consumption”, p. 19.
28 COLETTE MOORE

tronic genres that are developing through a combination of community consen-

sus and formalising practice. Some of these are more regulated by institutional
constraints (such as number of characters permitted in a tweet), but some are
evolving through user practice (such as the development of the hashtag as a
discourse marker). The communities of writers in late medieval England partic-
ipated in a similar growth period for the medium of text production: writing in
the vernacular was growing by leaps and bounds, and writing expertise was
expanding in its domains and networks.

Case Study: The Life of St. Thaïs

As a case study of the kinds of organisational variation that we see in

saints’ lives (and the stylistic implications of them), this section examines a
brief passage from the life of St. Thaïs from the Northern Homily Cycle (NHC)
manuscripts. Saints’ lives are a fruitful genre for examination of variation in
organisation: there are many surviving manuscripts drawing from different
regions, so there is substantial evidence of variant textual practices. These
works are popular texts, then, that have roots both in Latin sources and in ver-
nacular tradition, meaning that their organisational frames draw upon both
written legendaries and orally-transmitted folk stories. The combination of
both provides the stylistic and generic options that a text or scribe could em-
ploy, and they are therefore an important place to examine how issues of varia-
tion overlap with questions of genre.
The life of St. Thaïs, for example, challenges models of the genre of the
female saint’s life. Breaking the mould of the virgin-martyr saints, Thaïs is a
prostitute-saint: exemplary not for adherence to sexual purity but for the vehe-
mence of her repentance after the fact. The narrative uses tropes that go along
with and others that disrupt the generic presentation of the events, and dis-
course organisation is important in creating this effect. Fig. 1 shows San Ma-
rino, Huntington Library MS HM 129 (ff. 114r-114v). We can see an interesting
interchange between Thaïs and a monk, which occurs near the beginning of the
tale, in which their dialogue shifts back and forth with differing levels of atten-
tion to the switches in speaker.
Discourse Variation 29

Fig. 1 MS San Marino, Huntington Library, HM 129, ff. 114r-v (detail).

(2) San Marino, Huntington Library MS HM 129 (ff. 114r-114v)

He come and gaff pennyis twelfe

And sayd I woll syn wt þe my selfe
And into chamyr scho hym led
And schowyd hym a fayr bed
And bade hym stey and do þi will
And he answerd and sayd hyr till
þis sted ys noght privey I noghe

He came and gave twelve pennies

and said, “I will sin with thee myself”.
And she led him into a chamber
and showed him a fair bed.
And she bade him, “Stay, and do thy will”.
And he answered and said to her,
“This room is not private enough”.

The turn-taking in this dialogue between the monk and Thaïs presents their
interchange through their own voices and marks these in the original manu-
script with the verbs sayd (in the second line) and bade (in the fifth line), and
with the shifts in pronouns that indicate a change in speaker (since quotation
marks are not yet employed in English). The exchange has a somewhat star-
tling effect, of course, because readers are obliged to wonder if a monk is real-
ly soliciting a sex worker in a narrative about the life of a saint – our assump-
tions about the genre of saints’ lives conflict with the content of the passage, in
30 COLETTE MOORE

other words. The conversation in HM 129 does not make clear (to either Thaïs
or to the reader) that the monk’s real intention is to convert Thaïs, and the
effect is that we sit up and take notice; the ambiguity draws us in.
The Vernon manuscript version of the life, by contrast, presents a variant
form of the passage with the dialogue represented indirectly.

(3) Oxford, Bodleian Library, MS Eng. Poet. a. 1

he com and ¥af hire penies twelue

And seide he wolde synge wt hir hi selve (1)
And in to a chambre heo hi ledde
And schewed hym a feir bedde
And bad him anon don his wille (2)
And he anon seide hire tille
þis stude is not priue I nouh (3)

He came and gave her twelve pennies

And said he would sin with her himself.
And she led him into a chamber
And showed him a fair bed
And bade him then to do his will.
And he then said to her,
“This room is not private enough”.

In the Vernon manuscript version, the dialogue does not switch voices; rather
than depicting the utterances directly (“And sayd I woll syn with thee myselfe”),
the passage retains the third person pronouns that represent the speech indi-
rectly (“And seide he wolde synge wt hir hi selve”). This kind of difference may
seem minor at first glance; but it affects the intensity and vividness of the pas-
sage. Isabelle Buchstaller, for example, discusses how direct speech can feel
more immediate to readers than indirect speech;8 it is often understood to make
the narrative feel more oral and more proximate, and it can be used to build the
tension of the account of Thaïs’s conversion.
The passage, in its varying incarnations, presents three possible locations
for speech representation (numbered 1, 2, and 3 in the excerpt above). Table 1

8
I. BUCHSTALLER, Quotatives: New Trends and Sociolinguistic Implications (Chichester,
2014).
Discourse Variation 31

Table 1. Representation of the speech passages in the life of St. Thaïs in the Northern
Homily Cycle (NHC)

Manuscript 1 2 3

DIMEV 56

Oxford, Bodleian Library, Selden Supra 52 DS (I) IS (hys) DS (þs / ys)

Oxford, Bodleian Library, Bodleian Eng. poet c.3 DS (I) IS (hys) ? unclear

Oxford, Bodleian Library Bodleian Eng. poet a.1(Vernon) IS (he) IS (his) DS (þis / is)

Cambridge, Cambridge University Library, Addit. 8335 DS (I) IS (his) DS (þis / es)

Cambridge, Cambridge University Library, Dd 1.1 IS (he) IS (his) DS (þis / is)

Cambridge, Cambridge University Library, Gg 5.31 IS (he) –– DS (þis / es)

London, British Library, Addit. 22283 (Simeon) IS (he) IS (his) DS (þis / is)

San Marino, Huntington Library, HM 129 DS (I) DS (þi) DS (þis / ys)

DIMEV 4975

London, British Library, Cotton Tiberius E VII IS (he) DS (we) DS (þis / es)

London, British Library, Harley 4196 IS (he) DS (we) DS (þis / es)

DIMEV 5598

Oxford, Bodleian Library, Ashmole 42 IS (he) IS (his) DS (þis / es)

London, British Library, Addit. 38010 IS (he) IS (his) DS (this / es)

London, British Library, Harley 2391 IS (he) IS (hys) DS (þis / is)

London, Lambeth Palace Library 260 DS (I) IS (hys) DS (þis / es)

shows how the passages of represented speech manifest as direct speech or

indirect speech in the manuscripts. The Digital Index of Middle English Verse
(DIMEV) recognises three different versions for the life, and classifies them by
their opening line; Table 1 groups the manuscripts by these DIMEV classifica-
tions. DIMEV 56 (or Index of Middle English Verse 40) presents one version,
titled after the first line: A good tale hereby lies / Of a woman that hight Thais.
It survives in nine manuscript witnesses (of which this study has examined
eight). DIMEV 4975 (or IMEV 3178) is the designation for the section that begins
Narracio, Some time befell so in this land in the two manuscripts that contain
32 COLETTE MOORE

the “expanded” version of the Northern Homily Cycle. And DIMEV 5598 (IMEV
3547) is the third version of the life, There was a woman of ill fame, which has
four manuscript witnesses.
In part (1), the speech of the monk can be represented either directly (“I
will sin with thee myself”) or indirectly (“he would sin with her himself”). The
manuscripts are divided as to the mode, with nine of them using direct speech
and five using indirect speech. The variation patterns are particularly notable,
too, since the most straightforward and likely way for manuscripts to vary like
this is along lines of recension: this is the result when a slight difference is
introduced into the text and it is reproduced by copyists. The DIMEV divisions
are often an excellent gauge of the lines of recension, since the larger points of
coherence are used to group the versions of the text. And yet in this passage the
divisions in the use of direct and indirect speech do not neatly follow the
DIMEV groupings of the manuscripts: although the expanded Northern Homily
Cycle in MS Harley 4196 and MS Cotton Tiberius E VII use the same modes of
representation for all of the variables examined, the other DIMEV classifications
do not show consistency in the mode of speech representation for the utterance.
The pattern of variation is therefore slightly unusual; the divergence between
direct speech and indirect speech indicates the stylistic possibilities that re-
sulted from transmission.
In (2), the second location for represented discourse, most of the manu-
scripts represent Thaïs’s words indirectly (“And bad him anon don his wille”);
HM 129 is an exception in this regard (“And bade hym stey and do þi will”).
The expanded NHC manuscripts also rephrase the sentence to shift the mode;
they orient the direct speech in the first person plural (“And said here may we
wirk oure will”). Finally, the representation of speech in (3), the third location
for represented discourse, is consistent with respect to the mode; all manu-
scripts employ the direct speech markers of deictic þis and present-tense is
(“þis sted ys noght privey I noghe”). But the stylistic effect of this direct speech
might be different if this is the first shift to direct speech, as it is for Ashmole
42, BL Add 38010, and BL Harley 2391, or if this is simply another turn marked
in a back-and-forth dialogue (as it is in HM 129). In the manuscripts for which
this is the first use of direct speech, it has the effect of highlighting the third
utterance with respect to the first two in the narrative account.
This kind of stylistic variation can align in varying ways to differences in
content – Harley 4196 and Cotton Tiberius E VII, for example, add an earlier
Discourse Variation 33

couplet that spells out the monk’s virtuous intent from the beginning, creating
a text that is safer but less absorbing.

(4) London, British Library, Harley 4196

Narracio de meretrice Story of a whore

Sum tyme bifell so in þis land Some time ago befell in this land
A comun woman was dwelland A common woman was dwelling
In a cete and used hir syn in a city, and she used her sin
And spared nowther more ne myn And spared no one without exception 5
hir fairehede foles to foly drogh Her fairness drew fools to folly
And many sawles with sin scho slogh And she slew many souls with sin.
So it bifell opon a tide So it befell upon a time
A haly hermit þar biside A holy hermit there nearby
Thrugh þe grace of god mighty Through the grace of mighty God 10
Gert hir forsake hir foly Got her to forsake her folly.
he went to þe cete on a day He went to the city on a day;
þis woman for he wald assay He wanted to try this woman.
he said and gaf hir penis twelue He said, and gave her twelve pennies,
þat he wald play with hir him sellue That he would play with her himself. 15
þarfore he said help now þat we Therefore he said, “Help now that we
war in astede of preuete were in a private room”.
þan to a chamber scho him led Then she led him to a chamber
whare he fand a burely bed where he found a pleasant bed
And said here may we wirk oure will And said, “Here may we work our will”. 20
he answerd þan and said hir till He answered then and said to her,
þis sted es noght preue inogh “This room is not private enough”.

Before the dialogue begins, the reader is told in lines 10-11 that the holy hermit
(through the grace of mighty God) got her to forsake her folly. The knowledge
that the monk ultimately converts Thaïs forecloses the ambiguity of the inter-
change between the monk and Thaïs, since the reader already knows that he is
only pretending to solicit her in order to convert her. This might reduce the
intensity of its effect somewhat; certainly it influences the reader’s experience
of the dialogue.
Other stylistic variations change the effect of the passage as well. First, we
have the Latin labelling of Thaïs as a meretrix (‘whore’) in the title / opening
description before the tale begins. Most manuscripts do not contain this, they
more often just mark the name of Thaïs in the margin. Some do contain English
34 COLETTE MOORE

phrases that cast similar aspersions at Thaïs; HM 129 begins the tale with,
“And a gode litill tale her lyis / off and hore þat heght tayis” (“And a good
little tale lies here; of a whore called Thaïs”). The Latin title meretrix / meretri-
ce is a word possibly known to readers without knowledge of Latin; the OED
entry for ‘meretrix’ regards it as a borrowed word in medieval English. Label-
ling Thaïs a meretrix from the outset is balanced, interestingly, by other varia-
tions in word choices that seem more euphemistic – many of the manuscripts
say that he would sin with her himself where Harley 4196 offers the more
discreet reading that he would play with her himself.
Another stylistic variation unique to the expanded NHC comes through the
double opening – two formulaic beginnings that provide different kinds of
information. William Labov and Joshua Waletzky in their analysis of the com-
ponents of oral storytelling narratives list two different narrative functions to
which these correspond: the orientation, which gives basic background about
the characters or the setting, and the abstract, which provides a summary de-
scription.9 Labov and Waletzky’s framework is designed for oral storytelling,
but is applicable here, since these tales about the lives of saints share structural
features with oral storytelling. The first opening to the life is an orientation; it
contains the formulaic opening clause “Sum tyme bifell so in þis land” together
with a brief clause describing how Thaïs uses her beauty to lead men astray.
After that, the narrative contains a second opening formula “So it bifell opon a
tide” that begins another opening frame, this one an abstract that sets out the
main thread of the tale: the holy hermit led her to forsake her sinful life. This
double opening, an initiating formula plus orientation and an initiating formula
plus abstract, provides a kind of narrative / discursive variation. Sometimes this
kind of doubled opening is the product of textual construction: appended be-
ginnings can characterise combined texts (consider, for example, the two ver-
sions of the creation of Adam and Eve that are presented sequentially in the
book of Genesis). Alternately, the doubled opening could point to rhetorical
copiousness in the expanded version – the very description of the manuscript
as “expanded” suggests a principle of elaboration and dilation at work in its

9
W. LABOV, “Some further steps in narrative analysis”, The Journal of Narrative and Life
History 7 (1997), pp. 395-415, at p. 402; W. LABOV and J. WALETZKY, “Narrative analysis: Oral
versions of personal experience”, in: Essays on the Verbal and Visual Arts, ed. J. HELM (Seattle,
1967), pp. 12-44, at p. 32. See also J. SMITH, “Narrative: Sociolinguistic research”, in: Ency-
clopedia of Language and Linguistics, ed. K. BROWN, R.E. ASHER, and J.M.Y. SIMPSON
(Amsterdam, 2006), pp. 473-476, and N.R. NORRICK, “Conversational storytelling”, in: The
Cambridge Companion to Narrative, ed. D. HERMAN (Cambridge, 2007), pp. 127-141.
Discourse Variation 35

construction.10 Either way, the discursive differences between the manuscripts

containing the expanded NHC and the other manuscripts of the NHC present us
with different narrative experiences for the tale, a variant kind of organisation.
The reader’s experience of the interchange and initial assumptions about
the monk’s intentions, therefore, can be quite different depending upon which
manuscript he or she happens to be reading. To compare the manuscript ver-
sions is to uncover some of the differences in pragmatic pressures that influ-
ence the compilatio or textual organisation – choices in arranging a text must
balance conflicting motivations: to make clear the speakers and the switches in
voice, to tell a good story, to avoid misunderstanding, to stay on the safe side
of religious censorship, to stay faithful to the source material, to organise the
presentation of material.
Most of the variation that I have been discussing in this passage is linguis-
tic: the difference between different modes of speech reporting and the stylistic
effects of narrative organisation. The mise-en-page feature of the red ink and
the Latin heading in Harley 4196, though, also plays a role in how we interpret
the narrative: the clause always marks the beginning of the tale and it serves as
a kind of orientation. Headings and paratextual structures are interpretative
constructions and they work together with the words of the text to create struc-
tural and hermeneutic hierarchies. Describing Thaïs’s life as “narratio de
meretrice” rather than merely marking it in the margin with her name, for ex-
ample, inflects our understanding of the placement of the tale within the collec-
tion: even before the onset of the tale, the Latin frame has offered a meta-
linguistic and metadiscursive evaluation of Thaïs. The next section considers
these kinds of structural frames in more depth.

Case Study: The Life of St. Andrew

For another look at the pragmatic and interpretative effect of paratextual

elements, we might consider how the hierarchical construction of textual units
through titles, headings, paraphs, and initials works in the structure of the
opening to the life of St. Andrew.11 St. Andrew’s life in the South English
10
S. NEVANLINNA, The Northern Homily Cycle: The Expanded Version in MSS Harley 4196
and Cotton Tiberius E VII, 3 vols. (Helsinki, 1972-1984: Mémoires de la Société néophilologique
38, 41, 43).
11
Paratextual features have been growing as objects of interest in the wake of G. GENETTE,
Paratexts: Thresholds of Interpretation, trans. J.E. LEWIN (Cambridge, 1997).
36 COLETTE MOORE

Legendary ends with his martyrdom. After the tale of his life ends, though,
most of the manuscripts contain an account of a miracle of St. Andrew (called
in the TEAMS edition “St. Andrew and the Three Questions”12) which does not
feature St. Andrew in person, but is a posthumous legend of a bishop devoted
to St. Andrew and the miracle that comes from this inspiration. Is the miracle
a part of the life of St. Andrew, therefore, or is it an entirely different entry as
a separate tale into the legendary?
The answer to this question is relevant for our understanding of the genre
of the work. The textual tradition of the South English Legendary grows out of
the Latin saints’ lives with their more particular generic designations of vita
(the story of the life), passio (the story of the suffering and martyrdom), and
miracula (the stories of attendant miracles). Paul Strohm describes these gen-
res coming together in these later vernacular collections like the South English
Legendary; they might contain brief versions of several genres, employing the
category terms lyf and vita, or sometimes more particular genre titles such as
passioun.13 Strohm describes different incarnations of these genres in late Mid-
dle English and connects them to the designating genre terms. He suggests that
knowing the associated genre terminology can give an audience a clearer sense
for what to expect from a narrative, though he acknowledges that the use of
such generic terms is only one of the ways by which a reader might recognise
the genre of a work.14 This research suggests that the paratextual organisation
can also create a visual clue to the assignation of genre categories. The ambigu-
ity of the manuscripts on this point, however, does not provide certain answers,
but rather competing structures to understand textual divisions in the legend-
ary.
This question of textual divisions is addressed, therefore, by the mise-en-
page markings of the beginning of sections within the legendary. In some
manuscripts, the beginning of the section describing the life of St. Andrew and
the beginning of the section describing the miracle are marked in the same
way, presenting them at the same level of textual hierarchy. In Oxford, Bod-
leian Library, Ashmole 43, for example, an enlarged initial (identically sized

12
“St. Andrew and the Three Questions in the Scottish Legendary”, in: Saints’ Lives in
Middle English Collections, ed. E.G. WHATLEY, A.B. THOMPSON, and R.K. UPCHURCH (Kala-
mazoo, 2004), web (4 February 2016), <http://d.lib.rochester.edu/teams/text/whatley-saints-lives-
in-middle-english-collections-st-andrew-and-the-three-questions-introduction>.
13
P. STROHM, “‘Passioun, lyf, miracle, legende’: Some generic terms in Middle English
hagiographical narrative”, The Chaucer Review 10.1-2 (1975), pp. 62-75, 154-171.
14
P. STROHM, “‘Passioun, lyf, miracle, legende’”, p. 165.
Discourse Variation 37

at three lines high) indicates the beginning of each. Other manuscripts that
mark the vita and the miraculum in the same way include MS Oxford, Bodleian
Library, Laud misc. 463; MS London, British Library, Harley 4196; MS Lon-
don, British Library, Harley MS 2277; and MS London, British Library, Egerton
1993.
In some other manuscripts of the South English Legendary, the textual
division at the beginning of the miracle is marked, but not marked as promi-
nently as the beginning of the life of St. Andrew. Manuscripts like this include
MS Oxford, Bodleian Library, Bodley 779; MS London, British Library, Stowe
949; MS Cambridge, Trinity College, R.3.25; and MS Cambridge, Magdalene
College, Pepys 2344. Bodley 779, for example, has a more substantive separa-
tion for the opening to the life of St. Andrew. A one-line-high bar decorated in
blue with a centred title in red lettering (“Seint Andreu þe postil”) separates the
beginning of the life of St. Andrew from the end of the previous tale. The life
of St. Andrew follows this marked separator, and it begins with an enlarged red
initial (four lines high). The beginning of the section with the tale about the
miracle also has an enlarged initial, but it is three lines high instead of four,
and there is no decorated separating bar. In manuscripts like the one with two
different degrees of marking, the beginning of the life would seem to be a more
significant textual boundary than the beginning of the section with the tale of
the miracle, and the latter is structurally subordinate to the former.
Finally, some manuscripts mark the beginning of the life and do not distin-
guish the beginning of the miracle at all; these include MS Oxford, Trinity
College, 57; MS London, British Library, Cotton Julius D. IX; and MS Cam-
bridge, Cambridge University Library, Addit. 3039. In MS Oxford, Trinity
College, 57, the beginning of the life of St. Andrew is visibly set off on the
page: the text contains a three-line-high decorated initial and some red Latin
lettering in the right hand margin indicating the life (“vita scî andree apli”), but
the beginning of the miracle blends into the text and is not set apart at all. It is
not that the text contains no other kinds of textual division, either; indeed, there
is a red paraph mark two lines below the onset of the tale of the miracle, show-
ing that other structural units are acknowledged with paraph marks. With no
textual separating feature in the mise-en-page, however, a reader who does not
come with preconceived expectations that the miracle is another section of the
text would not find anything in this manuscript to suggest that it is set apart in
any way.
38 COLETTE MOORE

What, then, are the organising levels of a text like the South English Leg-
endary? The differences in the mise-en-page when we compare manuscripts
like Ashmole 43, Bodley 779, and Trinity College 57 present a significant
distinction in the classification of textual sections. A miracle is one of the
requisite aspects for sainthood, and creates a kind of posthumous authority for
a saint. Marking such a tale of posthumous influence in the same way as a
saints’ life suggests narrative equivalence: that it is a kind of ‘life’ – or at least
that the transition between St. Andrew’s vita and his miraculum is on par with
the transitions between the lives of different saints. In texts like Ashmole 43,
these are visually flagged as two separate tales would be. Marking the first part
more prominently, on the other hand, embeds the tale of the miracle within the
life of St. Andrew. This presents the posthumous tale of influence as a second-
ary section under the heading of the life of a saint. This is interesting because
it subsumes under a saint’s life an account of a tale that occurs after his death.
Subsuming the miracle as part of the life of a martyr, in particular, disrupts the
narrative arc of the legend, since a narrative of martyrdom ought to end with
the saint’s death. Appending an additional section after St. Andrew’s death
changes the effect of the vita; it presents an ‘afterlife’ narrative of the saint.
The sectional divisions in the legendary are accomplished through the
mise-en-page, and the narrative relationship between the account of the life of
St. Andrew and the account of his miracle is partly constructed through scribal
ordinatio. The variation that exists in the layout for the legendary determines
whether these are to be perceived as two sections or as one section in the leg-
endary, and assumes, therefore, a hermeneutic position of the relation of the
account of a posthumous miracle to the narrative ‘life’ of a saint.
Another kind of organisational frame in the life of St. Andrew can be
found in the use of Latin titles; this is common to saints’ lives, which often use
Latin for metadiscursive elements. The Latin formulae combine with different
ink colours or scripts to mark transitions in texts. MS London, British Library,
Harley 4196 shows the following text that introduces and then begins the life
of St. Andrew:

(5) London, British Library, Harley MS 4196

Incipit hic tractatus de legenda sc) orum Here begins the saints’ legends

here may me) luke who likes to lere here may men look who like to read
of liues and dedis of saintes fere of lives and deeds of fair saints
Discourse Variation 39

And in olde times how it bifell And in old times how it happened
als men in inglis tung mai tell that men may tell in English
Out of latyn þus er þai draune Translated from the Latin of these tales.
omong laud men forto be knaune To be known by respected men
And first es ordand forto shewe And first it is ordained to show
þe solempne fest of Saint Andrew The solemn feast of St. Andrew.
de sc) o Andrea aplo) historia Of St. Andrew Apostle

Saint Andrew cristis apostil dere St. Andrew was Christ’s dear apostle
Whils he went in þis werld here While he was alive in the world
fful mekill folk in fere cuntre Many folks in fair countries
to cristen trouth convertid he He converted to Christian truth
And at þe last so it byfell And at last it happened
in a cete where he gun swell In a city, where this began
A domesman in þat cete was A judge was in that city
and his name was cald Ægeas and he was called Aegeas.

The Latin lines are in red ink, so they are set apart doubly by colour and lan-
guage. Moreover, they are not integrated into the rhyme scheme, so they are
not part of the verse structure of the tale. The first line begins the legendary
and the tenth line begins the life of St. Andrew. It is interesting, however, that
the enlarged initials (in gold, with red and blue backgrounds) do not mark the
onset of the Latin line; instead they flag the beginning of the English text that
follows. The fact that the enlarged red initial is attached to the onset of the
English text suggests that, hermeneutically, the Latin text is a metadiscursive
line – that it is not the ‘first’ line of the tale but is prior to the tale. Rather, the
English text, beginning with the enlarged initial, is the true onset of the tale.
Creating textual hierarchies that indicate the relations of sections of text to
one another, therefore, depends upon paratextual elements to designate these
units of discourse. The life and miracle of St. Andrew depend upon different
markers: initials, ink colour, language choice, poetics, headings. The organisa-
tion variation has narrative significance, namely, what gets prioritised in the
interpretation of a text, but it also has cultural and ideological significance
through the theological interpretation it presents for saints’ lives and their
genres.15

15
For a discussion of cultural and ideological significance of variation in the organisation
of Nicholas Love’s Mirror of the Blessed Life of Jesus Christ, see MOORE, “Visual pragmatics”.
40 COLETTE MOORE

Conclusion

The differences in the narrative structure and the mise-en-page for these
manuscripts, then, serve to mark the words of the text in varying ways that
differently construct the relationships of levels of discourse in written material.
Examining the variation in the organisational marks of the mise-en-page and
the manuscript methods of rubrication reveals the kinds of discourse levels that
the medieval scribes found important to designate, and the ways that these
overlap.
The devotional texts show examples of many different kinds of boundaries
in texts; this study examines direct speech, narrative hierarchies in paratextual
headings, and cultural authority. These boundaries get flagged with a combina-
tion of linguistic and scribal kinds of organisation: rubrication, section bound-
aries, modes of speech representation. Looking at a collection of manuscript
witnesses shows the interpretational significance of the variations in organisa-
tion, gives texts a wider range of possibilities for the hermeneutics of textual
ordering, and reveals to us the proliferation of organisational strategies that
occurred in the late Middle Ages. In considering the different (and sometimes
conflicting) pragmatic pressures upon scribes in organisation – the pragmatic
impulses to clearly delineate parts of a text, to make clear the speakers and the
switches in voice, to tell a good story, to avoid misunderstanding, to stay on the
safe side of religious censorship, to stay faithful to the source material, to man-
age systematically the presentation of material – we get a fuller vision of the
combination of manuscript ordinatio with linguistic structuring and the reasons
why some marking conventions become ossified and standardised and others
wither.
 How the Page Functions: Reading
Pitscottie’s Cronicles in Manuscript and Print

FRANCESCA L. MACKAY

Introduction

uth Carroll et al.’s article on the “Pragmatics on the page” highlights the

R potential of combining two strands of study (“book history and material-

ist philology on the one hand and historical pragmatics and historical
discourse linguistics on the other”).1 The approach presented by the “Prag-
matics on the page” article, in combination with Paul Zumthor’s theory of mou-
vance, which perceives the multiple witnesses of an individual work as “les
textes concrets qui la réalisent présentant, par le jeu des variants et remanie-
ments, comme une incessante vibration et une instabilité fondamentale”, form
the theoretical background of this chapter’s examination of the historical evo-
lution of a text’s form and function through close textual analysis of the mate-

1
R. CARROLL, M. PEIKOLA, H. SALMI, M. VARILA, J. SKAFFARI, and R. HILTUNEN, “Prag-
matics on the page: Visual text in late medieval English books”, European Journal of English
Studies 17 (2013), pp. 54-71, at p. 54.

......................................................................................................................................
Verbal and Visual Communication in Early English Texts, ed. Matti PEIKOLA, Aleksi
MÄKILÄHDE, Hanna SALMI, Mari-Liisa VARILA, and Janne SKAFFARI, Utrecht Studies in Medieval
Literacy, 37 (Turnhout: Brepols, 2017), pp. 41-65.

DOI 10.1484/M.USML-EB.5.114130
42 FRANCESCA L. MACKAY

rial page.2 Three witnesses of Robert Lindesay’s Cronicles of Scotland (Wod-

row MSS. Folio XLVIII (referred to as: the Wodrow Folio), Acc. 9769 84/1/1
1/2 (referred to as: Crawford MS I), and Robert Freebairn’s 1728 first printed
edition (University of Glasgow Library, Glasgow: Sp. Coll. BD5-b.4)) will be
examined in order to exemplify how features of mise-en-page (specifically
punctuation practices found within a selected extract from the Cronicles and
the paratextual features identified within the chronicle-text as a whole) can be
interpreted as indicators of the literacy practices of the anticipated readers for
whom the various witnesses were produced.
Composed c. 1542-1575, Robert Lindesay of Pitscottie’s Historie and
Cronicles of Scotland is the earliest prose chronicle of Scotland to be written
in the Scots vernacular. It documents the period of 1436-1575, covering the
reigns of the Scottish monarchs James II, James III, James IV, James V, and
Mary Queen of Scots. The work is often reproduced with an “Addition” by an
unknown author, which continues the chronicle into the early seventeenth
century and documents the reign of James VI of Scotland and I of England;
however, the date to which the “Addition” runs varies from witness to witness.
The earliest annal entries of Pitscottie’s Cronicles (1436-1460) are a transla-
tion of Hector Boece’s 1527 Latin chronicle of Scotland, Historia Gentis
Scotorum, and the subsequent section (1460-1542) is a compilation of various
earlier Scottish historians’ works.3 Only the entries for the contemporary pe-
riod in which Pitscottie was writing (c. 1542-1575) are his original composi-
tion, therefore Pitscottie was writing both during and about the events of the
Scottish Reformation. Pitscottie was a contemporary of the whole Reformation
movement in Scotland (approximately 1533-1567), and has since been labelled
a Protestant sympathiser, though he never openly displayed his political affilia-
tions and his name appears in no public records from this period.4
The reproduction history of Pitscottie’s Cronicles is notable due to its
extended period of manuscript circulation and the significant delay before its
publication in print. The work circulated in manuscript form throughout the

2
P. ZUMTHOR, Introduction à la poésie orale (Paris, 1983), p. 507. Translated as “those
concrete texts which constitute the work’s real existence present through the play of variants and
re-workings something like ceaseless vibration of fundamental instability” in P. ZUMTHOR and
J. MCGARRY, “The impossible closure of the oral text”, Yale French Studies 67 (1984), pp. 25-42,
at p. 33.
3
Æ.J.G. MACKAY, The Historie and Cronicles of Scotland: Robert Lindesay of Pitscottie
(Edinburgh and London, 1899), pp. XLII-XLIII.
4
MACKAY, The Historie and Cronicles of Scotland, p. XXXV.
How the Page Functions 43

late sixteenth and seventeenth centuries, yet was not printed until the eigh-
teenth century, when it was printed by Robert Freebairn in 1728, Robert Urie
in 1749, Thomas Cadell in 1778, and Sir Graham Dalyell in 1814.5 The sub-
stantial delay between the composition of the Cronicles and its first appearance
in print, despite printing having been gradually established in Scotland
throughout the sixteenth and seventeenth centuries, was due to a number of
socio-political reasons. First, Pitscottie comments negatively on the Earl of
Morton who was exiled during the reign of Mary Queen of Scots – the period
in which Pitscottie was writing – but who was instated as regent during the
early years of James VI and I’s reign (1572-1581). The period in which Morton
was in power, therefore, would have been the prime time (immediately post-
completion of the Cronicles) for the work to be printed, but it would not have
been deemed suitable (or safe) to print while Morton was in authority.6 Once
Morton’s control over the censorship of the Cronicles ended there were further
problems for the work: Pitscottie died in approximately 1578 followed by his
patrons, John Stuart, the fourth Earl of Athole, and Robert Stuart, the Bishop
of Caithness, in 1579 and 1586 respectively; therefore, after the death of the
Bishop of Caithness there was potentially no one with a vested interest in the
Cronicles to negotiate and pay for its publication in print.7 Also problematic
for wide circulation of the work in print were its Presbyterian tendencies and
Pitscottie’s outspoken comments on Mary Queen of Scots, neither of which
James VI and I would have approved of, meaning the Cronicles were not print-
ed until the early eighteenth century when the Stewart dynasty had ended and
the Hanoverian succession had begun.8

5
R. LINDSAY, The History of Scotland, ed. R. FREEBAIRN (Edinburgh, 1728) [University
of Glasgow Library, Glasgow: Sp. Coll. BD5-b.4]; R. LINDSAY, The History of Scotland, ed. R.
URIE (Glasgow, 1749) [University of Glasgow Library, Glasgow: Sp. Coll. Bo3-m.12]; R. LIND-
SAY , The History of Scotland, ed. T. CADELL (London, 1778) [British Library, London, re-
produced on Eighteenth-Century Collections Online: ESTC No. T083320]; and R. LINDSAY, The
History of Scotland, ed. J.G. DALYELL (Edinburgh, 1814) [University of Glasgow Library,
Glasgow: Sp. Coll. BD13-i.23].
6
MACKAY, The Historie and Cronicles of Scotland: Robert Lindesay of Pitscottie, p. LV.
7
MACKAY, The Historie and Cronicles of Scotland: Robert Lindesay of Pitscottie, p. LVI.
8
MACKAY, The Historie and Cronicles of Scotland: Robert Lindesay of Pitscottie, p. LVII.
44 FRANCESCA L. MACKAY

Theoretical Approach

The position of philology at the end of the twentieth century paved the way
for the conception of “Pragmatics of the page” as outlined by the previously
mentioned article (the combination of book history, materialist philology, his-
torical pragmatics, and historical discourse linguistics).9 Late twentieth-century
‘new philology’ was centred on manuscript culture and broadened the features
analysed by philologists to include the material context. ‘New philologists’
were often editors and as such recognised the multiplicity of texts and the
various forms by which medieval material culture has reached us today.10
‘New philology’s’ attention to textual variation corresponds with Bella
Millett’s conception of Zumthor’s mouvance – and, similarly, the late
twentieth-century emergence of historical pragmatics – which suggests that
textual variation is the result of differing functions, and that a work can be
“pragmatically adapted [...] for changing audiences and changing purposes”.11
It argues that the form of a text and its function are intrinsically linked.12 The
suggestion that there is variation between individual texts of a work due to
their discrete pragmatic functions is the premise of this chapter’s approach to
mise-en-page: that a work is repeatedly ‘re-formed’ – or continuously ‘appropri-
ated’ – to aid various specific socio-cultural practices.13 Analysis of multiple
witnesses of the same work – in accordance with David Pearson’s suggestion
that the late twentieth century saw copy-specific information become a grow-
ing approach to book history – will suggest how and why the mise-en-page of
a text (in manuscript or printed form) changes depending on the readership for
which the witness was produced.14
In support of the argument that mise-en-page can be interpreted as an indi-
cator of reading practices, Roger Chartier asserts that “the act of reading sim-

9
CARROLL et al., “Pragmatics on the page”.
10
S.G. NICHOLS, “Introduction: Philology in a manuscript culture”, Speculum 65.1 (1990),
pp. 1-10, at pp. 8-9.
11
B. MILLETT, “What is mouvance?”, Wessex Parallel Web Texts (24 Augustus 2011),
<http://www.southampton.ac.uk/~wpwt/mouvance/mouvance> [Accessed 27 February 2013].
12
A. FETZER, “Challenges in contrast: A function-to-form approach”, in: Contrastive
Pragmatics, ed. K. AIJMER (Amsterdam and Philadelphia, 2011), pp. 73-96, at p. 73.
13
R. CHARTIER, “Texts, printing, readings”, in: The New Cultural History, ed. L. HUNT
(Berkeley, CA, 1989), pp. 154-175, at p. 171.
14
D. PEARSON, “What can we learn by tracking multiple copies of books”, in: Books on the
Move: Tracking Copies through Collections and the Book Trade, ed. R. MYERS et al. (New
Castle, DE, and London, 2007), pp. 17-38, at p. 34.
How the Page Functions 45

ply cannot be divorced from the text itself”.15 Intensive readers read a small
number of texts but read them frequently, therefore predominantly committing
the texts to memory.16 While reciting aurally, intensive readers used the physi-
cal text as an aide-memoire to prompt their memory when needed, rather than
reading aloud from the material text directly word-for-word.17 During the medi-
eval and early modern periods, a readership of silent, extensive readers began
to emerge.18 Throughout this period, literacy spread, and as more people be-
came literate, more people began to read individually, in private, silently. Si-
multaneously, more works entered circulation (partially due to the introduction
of printing), and people began reading increasingly more extensively, reading
a wider range of works, and encountering the same text less frequently. During
the late medieval and early modern periods, therefore, the emerging extensive
readership was increasingly more likely to both encounter a material text di-
rectly themselves (rather than reciting from memory), and encounter a work
that was unfamiliar to them. Subsequently, as silent, extensive, private reading
practices became more widely used, the material pages of manuscripts and
books needed to provide increasingly more guidance for the reader in order to
aid the reading process, as the readers could no longer rely on their prior
knowledge of the text in order to both physically read and comprehend text.19

15
R. CHARTIER, The Cultural Uses of Print in Early Modern France, trans. L.G. COCHRANE
(Princeton, 1987), p. 7.
16
The terminology of ‘intensive’ and ‘extensive’ reading practices is discussed by J.J.
SMITH, “Punctuating Mirk’s Festial: A Scottish text and its implications”, in: Preaching the Word
in Manuscript and Print in Late Medieval England, ed. M.W. DRIVER and V. O’MARA
(Turnhout, 2013), pp. 161-192.
17
J. COLEMAN, Public Reading and the Reading Public in Late Medieval England and
France (Cambridge, 1996), p. 28, suggests an oral-aural-silent trichotomy of reading practices.
She describes oral performance as the verbal composition of bards and minstrels whereby the
utterance is not formed in writing, whereas aurality depends on a written text as the source for
reading aloud (usually in a public environment).
18
P. SAENGER, Space Between Words: The Origins of Silent Reading (Stanford, 1997), pp.
6, 258-259, acknowledges that the ability to read silently and rapidly began with the evolution
of word separation in the seventh century but became increasingly pervasive in the thirteenth,
fourteenth, and fifteenth centuries. E. JAJDELSKA, Silent Reading and the Birth of the Narrator
(Toronto, 2007), p. 3, suggests that this silent readership reached a “critical mass” in the late-
seventeenth and eighteenth centuries.
19
For further discussion of the use of memory during the reading process see: W. ONG,
Orality and Literacy, 2nd edn. (Abingdon, 2002), p. 117; A. MANGUEL, A History of Reading
(London, 1997), p. 28; M.T. CLANCHY, From Memory to Written Record: England 1066-1307
(London, 1979); M.J. CARRUTHERS, The Book of Memory: A Study of Memory in Medieval
Culture (Cambridge, 1992); and M. CARRUTHERS and J.M. ZIOLKOWSKI, “General introduction”,
46 FRANCESCA L. MACKAY

The concept of guidance and the analysis of the guiding features (punctuation
practices and paratextual features) provided by the scribe or compositor to aid
the specific readership they are catering for is therefore central to this analysis:
a text’s mise-en-page functions as a device to aid specific literacy practices.20
Crucially, and in accordance with current perspectives on the history of
reading by scholars such as Elspeth Jajdelska, Andrew Pettegree, Adam Fox
and Joyce Coleman, this chapter interprets reading practices as occurring with-
in a spectrum, a continuum wherein the two opposing ends indicate (largely
unattainable) fully intensive and extensive reading practices, with more nuan-
ced practices situated along it.21 It presents reading practices as being in a com-
plex, non-linear relationship of coexistence and mixedness: the boundaries
between specific – and often seemingly contradictory – practices were “thor-
oughly permeable and constantly shifting so that the dichotomy is difficult to
identify and impossible to sustain”.22 As Adam Fox states, these practices “are
rarely discrete entities or inversely related [...] instead they form a dynamic
continuum, each feeding in and out of the other in the development and nour-
ishment of both”.23 Supporting Fox’s “dynamic continuum” this study will
discuss individual texts within a spectrum which acknowledges the highly
nuanced practices of the readerships for which they were catering. Early mod-
ern readers were real, living entities, therefore they do not correspond with
abstract labels: “rather than imposing universal, self-validating categories of
‘oral’ and ‘literate’ style on texts, we should work outwards from given texts

in: The Medieval Craft of Memory: An Anthology of Texts and Pictures, ed. M. CARRUTHERS and
J.M. ZIOLKOWSKI (Philadelphia, PA, 2004), pp. 1-31.
20
The theory that features of a material page provide guidance for specific anticipated
readerships evolves from discussions regarding punctuation practices as guiding aids by M.B.
PARKES, Pause and Effect: An Introduction to the History of Punctuation in the West (Aldershot,
1992; Berkeley and Los Angeles, 1993); JAJDELSKA, Silent Reading; and SMITH, “Punctuating
Mirk’s Festial”.
21
For further discussion of reading practices as a spectrum, see: F.N. AKINNASO, “The con-
sequences of literacy in pragmatic and theoretical perspectives”, Anthropology and Education
Quarterly 12.3 (1981), pp. 163-200; CHARTIER, The Cultural Uses of Print; R. DARNTON,
“Towards a history of reading”, The Wilson Quarterly 13.4 (1989), pp. 86-102; COLEMAN, Pub-
lic Reading; A. FOX, Oral and Literate Culture in England 1500-1700 (Oxford, 2002); ONG,
Orality and Literacy; E.L. EISENSTEIN, The Printing Revolution in Early Modern Europe, 2nd
edn. (Cambridge, 2005); JAJDELSKA, Silent Reading; and A. PETTEGREE, The Book in the
Renaissance (New Haven and London, 2010).
22
FOX, Oral and Literate Culture, p. 39.
23
FOX, Oral and Literate Culture, p. 50.
How the Page Functions 47

and literary environments to develop culture-specific descriptive systems”.24

Therefore the spectrum used to describe the readerships of Pitscottie’s Cron-
icles will differentiate between specific – yet coexisting – reading practices
being used ‘more or less’.25 As such, antithetical terminology will be avoided;
describing reading practices as mutually exclusive entities leads to reductionist
options rather than the lively variations that actually exist.26
Elspeth Jajdelska’s discussion of the history of reading and reading prac-
tices – in relation to late seventeenth-century and early eighteenth-century
English prose – has influenced the construction of this chapter’s argument that
the features of a material text can be interpreted as indicators of reading prac-
tices.27 While agreeing that different reading practices coexist within a single
society, Jajdelska presents two opposing models of reader and emphasises that
a reader cannot, at a single moment in the reading process, be both a “reader as
speaker” (reading aloud) and “reader as hearer” (a silent reader subvocalising
the text and becoming a “hearer of an internal voice”).28 Subsequently, Jajdel-
ska’s argument can be expanded to suggest that when producing a text only one
model of reader can be actively catered for at a time. A scribe / compositor,
though, may cater for various reading practices within a single text; for exam-
ple, the scribe / compositor may fluctuate in regard to the model of reader he
is catering for due to uncertainty regarding the primary practices used by the
anticipated readership, or he may fail to consistently cater to newer reading
practices and therefore (accidentally) revert back to the traditional practices he
is more familiar with. Therefore, features of mise-en-page – which provide
guidance for the reader – can be interpreted as indicators of the contemporary
literacy practices. By examining features of mise-en-page within the context of
each witness as a whole, we can deduce the overall reading environment of the
anticipated readership for which the witness was produced.

24
COLEMAN, Public Reading, p. XLI.
25
AKINNASO, “The consequences of literacy”, p. 166.
26
COLEMAN, Public Reading, p. 15.
27
JAJDELSKA, Silent Reading.
28
JAJDELSKA, Silent Reading, pp. 7, 43-48. Further, JAJDELSKA, Silent Reading, p. 8, refers
to modern society’s acceptance of the metaphor that reading is hearing and the studies by M.L.
SLOWIACZEK and C. CLIFTON, “Subvocalization and reading for meaning”, Journal of Verbal
Learning and Verbal Behaviour, 19.5 (1980), pp. 573-582, and K. RAYNER and A. POLLATSEK,
The Psychology of Reading (New Jersey, 1989), as evidence for subvocalisation.
48 FRANCESCA L. MACKAY

Methodology

This chapter examines features of mise-en-page within the context of the

witness as a whole; see Ferdinand de Saussure’s structuralist concept of lan-
guage as a system “où tout se tient”.29 The punctuation practices and paratext-
ual features (for example: prefatory, supplementary, and additional material
(not directly related to the chronicle-text); features of sub-division; scribal
formal marginalia) found within each witness of the work function as part of an
overall supporting system within the text and therefore must be examined in
relation to other features within their own system of either punctuation or para-
text and in relation to each other, and within the context of both the material
text as a whole and the wider socio-cultural context for which the text was
produced. There are various potential reasons for the inclusion of an individual
feature: it could have been included to aid reading practices, but also, for ex-
ample, for technological reasons such as the limitations of the printing forme
– yet when it is examined alongside the other paratextual features and punctua-
tion practices in use, the specific practices of the literacy environment that is
being catered for become clearer. It is the relationship between individual
features and the overall guiding system they construct which indicates the
specific readership that the scribe or printer is anticipating.
The analysis of features of mise-en-page to produce hypotheses on larger
socio-cultural practices allows for a combination of a series of complementary
methodologies. Quantitative-to-qualitative and micro-to-macro methodologies
are combined through the quantitative description of the occurrence of the
specified micro-linguistic features of mise-en-page in each of the witnesses of
Pitscottie’s Cronicles under analysis (how often each punctuation mark occurs
within the selected extract, and how often each item of paratext occurs within
the overall witness), followed by qualitative discussion of these features in
relation to the socio-cultural practices for which they cater (the macro-linguis-
tic context). As Jucker and Taavitsainen suggest, the contextualisation of tex-
tual features (linguistic, visual, or material) is key to historical pragmatic re-
search: “contextualise your findings by considering larger issues, and by relat-
ing your findings to the multilayered context from the microlevel to the macro-
level of culture”.30

29
J.J. SMITH, A Historical Study of English: Function, Form and Change (London, 1996),
p. 5.
30
JUCKER and TAAVITSAINEN, English Historical Pragmatics (Edinburgh, 2013), p. 50.
How the Page Functions 49

A consideration of the methodology of this study is the issue raised by

Jucker and Taavitsainen regarding generalisations in historical pragmatic re-
search.31 Problematically, generalisations frequently occur in qualitative,
macrolinguistic research projects in which hypotheses are being made regard-
ing wider socio-cultural contexts on the basis of specific features. However,
this study emphasises that while it may offer hypotheses regarding the broader
literacy practices in Scotland during the early modern period, its findings are
firmly rooted in the data under analysis: the focus is on the reading practices
catered for within the selected witnesses of Pitscottie’s Cronicles, 1575-1814.
In order to analyse the punctuation practices used by the scribes / composi-
tors of Pitscottie’s Cronicles a close study of the text is required. For this pur-
pose a parallel section of text has been selected from each of the witnesses of
the work under analysis. The extract which has been chosen – the documenta-
tion of the siege of St Andrews castle (1546-1547) – was selected as it occurs
in all witnesses of the work and is from the section of the work which is
Pitscottie’s original composition. The content of the selected extract varies
slightly across the three witnesses of the Cronicles under analysis.32 The Wod-
row Folio and Crawford MS I both include slightly longer accounts of the siege
than Freebairn’s 1728 printed edition, but it was decided that it was essential
to include the account of the siege in its full form from all witnesses so as to
enable comparison of how the same content and textual environments were
punctuated and presented on the page (rather than, for example, extracting a
specified number of lines from each witness, which would have varied in con-
tent due to scribal layout and the different forms of the account).
In contrast, the whole bound volumes of the three witnesses of the Cron-
icles have been examined in order to collate the paratextual features which are
in use. This was necessary as paratextual features are more widely dispersed
throughout the text, meaning that the study of a specific extract – as has been
conducted to collect data on the punctuation practices – would not have pro-
vided accurate results or results of a significant enough quantity for subsequent
analysis.

31
JUCKER and TAAVITSAINEN, English Historical Pragmatics, p. 25.
32
Across witnesses of the Cronicles three general versions of the extract have been
identified: an abbreviated form (found within seven witnesses, including Freebairn’s edition;
extracts of 17-27 lines), a long form (found within seven witnesses, including the Wodrow Folio;
extracts of 30-44 lines), and the long form with idiosyncratic differences (found within three
witnesses, including Crawford MS I; extracts of 59-81 lines).
50 FRANCESCA L. MACKAY

Case Studies

Three witnesses, taken from across the period of the Cronicles’ reproduc-
tion history, have been selected, showing clear differences in regard to the
paratextual and punctuation systems they employ, and – arguably – therefore
the readerships for which they cater. The Wodrow Folio is the only known
witness of the Cronicles to be found within a miscellany. The miscellany vol-
ume contains numerous texts composed by different scribes, but a single scribe
seemingly composes the first ten textual items (all of which are sermons) and
the whole chronicle-text (the eleventh item in the miscellany). The sermons
immediately following the Cronicles are composed in a different hand. The
Wodrow Folio is undated in its catalogue entry, but it is potentially a mid
seventeenth-century witness, as the sermons surrounding the Cronicles in this
volume are all dated to 1638-1643.33
Crawford MS I is also undated within its catalogue entry; Mackay, though,
suggests the manuscript can be dated to c. 1598.34 The chronicle-text has been
composed by one or two scribal hands; there is possibly a change in hand to-
wards the end of the section documenting Mary Queen of Scots’ reign. The
pre-chronicle materials (two prefatory items and an additional item) are also
scribal, as – seemingly – are the post-chronicle additional items (apart from
Johnne Strattoris’ sermon).
Robert Freebairn’s first edition of the Cronicles was printed in Edinburgh
in 1728 in a roman serif font. It contains the “Verses to the Bishop” and the
“Author’s Account to the Reader”, which are prefatory items that are fre-
quently found within witnesses of Pitscottie’s Cronicles; however, as the first
printed witness of the work, Freebairn’s edition is the first witness to include
a “Printer’s Preface” and a “List of Subscribers”.

The Wodrow Folio

Overall, the witness of Pitscottie’s Cronicles within the Wodrow Folio (MS
Edinburgh, National Library of Scotland, Wodrow Folio XLVIII) provides
very little guidance for an unfamiliar reader of the material text; this witness is

33
MACKAY, The Historie and Cronicles of Scotland, p. LXXXVIII, broadly dates this witness
to post-1605.
34
MACKAY, The Historie and Cronicles of Scotland, p. LXXXIII.
How the Page Functions 51

almost completely void of punctuation marks.35 Within the selected extract

only litterae notabiliores are found and they are used relatively infrequently;
therefore the scribe does not seem to have attempted to compensate for the lack
of diversity within the punctuation system through more frequent use of litte-
rae notabiliores (“more noticeable letters from a display script [...] used to
indicate the beginnings of sententiae or periods”).36 For example, only twenty-
three litterae notabiliores are found in the selected extract from the Wodrow
Folio, whereas the extract within another manuscript (MS Edinburgh, Univer-
sity of Edinburgh Library, La.III.218) also lacks diversity within its punctua-
tion system (similarly only including litterae notabiliores) but it includes sixty-
six litterae notabiliores – a much higher quantity than is found in the extract
from the Wodrow Folio.
The use of a punctuation system which neither makes frequent use of punc-
tuation marks nor includes a diverse range of marks suggests that the scribe
anticipated a more intensive reader of the text. As discussed, intensive readers
encountered a limited number of texts and therefore read the same text fre-
quently; becoming extremely familiar with the selected texts with which they
engaged and to a certain extent committing them to memory, using the material
texts as aide-memoires. Therefore, as intensive readers would have had exem-
plary prior knowledge not only of the content of the text but also of how to
read it – for example, what it should sound like; where pauses should be placed
– less guidance was required on the material page. It can thus be argued that a
sparse punctuation system, such as the system found in the Wodrow Folio,
would be sufficient for a more intensive reader.
Similarly, the Wodrow Folio lacks a clear system of textual division for the
conventional five monarchs’ sections and the “Addition”. Every witness of the
Cronicles examined divides the text into clear sections documenting the reigns
of James II, James III, James IV, James V, Mary Queen of Scots, and – if in-
cluded – James VI and I (labelled as the “Addition”), each of which are intro-
duced with an intertitle, along with various accompanying features of textual
division: for example, page divisions, white space, enlarged initials, enlarged
or emboldened text. The scribe of the Wodrow Folio, though, while still seg-

35
The following punctuation marks have been identified within witnesses of Pitscottie’s
Cronicles: puncti, commata, distinctio, semi-cola, double puncti, novel combinations of
punctuation marks, and litterae notabiliores (PARKES, Pause and Effect, p. 43 describes how
litterae notabiliores became more important as the role of puncti diminished, reinforcing this
chapter’s decision to treat litterae notabiliores as part of a text’s punctuation system).
36
PARKES, Pause and Effect, p. 305.
52 FRANCESCA L. MACKAY

menting the text into the above-mentioned monarchs’ sections, does not do so
as distinctively as other scribes (several of the intertitles do not even include
surrounding white space), nor does he introduce the monarchs’ sections in a
consistent style (i.e. there are varying degrees of emphasis given to the inter-
titles and incipits).
The lack of features marking textual division in this manuscript – for in-
stance the lack of consistently emphasised monarchs’ sections, enlarged ini-
tials, paragraphs, chapters – suggests that the scribe is anticipating a readership
that would already be familiar with the text and therefore require little guid-
ance (as was suggested by the punctuation system). Problematically, though,
the sparse markers of textual division which are included in this manuscript are
not visually distinctive – as noted above. Therefore the scribe has not provided
any paratextual features to enable the manuscript to be used as an aide-me-
moire, a form which would have aided more intensive reading practices. Hugh
of St. Victor (writing in the early twelfth century) explains that manuscript
page layout and decoration are of

great value for fixing a memory-image that when we read books, we study to im-
press on our memory [...] the colour, shape, position, and placement of the letters
[...] in what location (at the top, the middle, or bottom) we saw [something] posi-
tioned.37

Hugh of St. Victor’s comments suggest that while the anticipated readership
are expected to have prior familiarity with the Cronicles (from encountering
the text frequently), they are not expected to be using this familiarity to recite
the text from memory using the material text as an aide-memoire. Instead, the
reader is expected to use his prior knowledge of the text to navigate the content
correctly and comprehend the divisions appropriately without consistency in
layout.
Further, there are numerous pro-Protestant sermons and psalms included
alongside Pitscottie’s Cronicles in this miscellany. As analysis of the above
features suggested that the scribe anticipated that the reader would have prior
familiarity with the text, it could be suggested the scribe focussed their
paratextual provision on features which aid the reader to ‘correctly’ (reli-
giously, politically, or socially) interpret the chronicle-content, rather than to
functionally encounter the written text. For example there are a series of mid-

37
CARRUTHERS, The Book of Memory, p. 9.
How the Page Functions 53

section intertitles highlighting the martyrdom of George Wishart and scribal

formal marginalia are included – both of which are used to ensure that the read-
er is ‘correctly’ interpreting the work and is therefore receiving the pro-Re-
formist message it is aiming to transmit.38
Finally, it seems a notable feature of the Wodrow Folio that Pitscottie’s
Cronicles were selected for use in the miscellany despite differing so drasti-
cally in length to the other texts which are included. This suggests that the lack
of visually-notable textual division in this witness of the Cronicles could per-
haps be further explained in relation to the other forms / layouts of texts in the
miscellany. The Cronicles are considerably longer than the individual psalms
and sermons that are positioned both before and after it; perhaps, therefore,
while (some) features of textual division have been included within the Cron-
icles to aid the reading of such a lengthy text type (which a reader of the ac-
companying sermons and psalms may not have been accustomed to), the scribe
has avoided using visually distinct forms so as to prevent confusion for the
reader of the miscellany volume as a whole. There is the potential for a reader
of the miscellany who was unfamiliar with the content and structure of Pitscot-
tie’s Cronicles to misinterpret emphasised features of textual division as a
marker of a new textual item within the miscellany. For example, it is common
in other manuscript witnesses of the Cronicles for each monarch’s section to
begin on a new page with an emphasised intertitle positioned centrally at the
top of the page. However, within the Wodrow Folio this practice is often de-
ployed to indicate a new textual item; the presentation of the Cronicles in this
form would have had the potential to cause confusion and misinterpretation for
the reader, and the scribe of the Wodrow Folio has therefore adjusted the con-
ventional form of the Cronicles to correspond with the function of a miscel-
lany.
The resulting diversity of reading practices which are being highlighted
within the Wodrow Folio, therefore, reinforces the central hypothesis of this
study: that early modern reading practices were not dichotomous entities; read-
ers were extremely nuanced in terms of the practices they employed. The Wod-
row Folio includes scribal formal marginalia and a mid-section intertitle for the
section documenting George Wishart’s martyrdom – both of which would aid
an unfamiliar reader – yet the conventional textual division is inconsistent and

38
W.W.E. SLIGHTS, Managing Readers: Printed Marginalia in English Renaissance Books
(Ann Arbor, MI, 2001), p. 11, suggests that “the margins were conceived of as a space in which
readers’ responses to a text could be influenced”.
54 FRANCESCA L. MACKAY

the punctuation system provides little guidance, therefore suggesting that the
scribe was anticipating a readership at a medial position on this spectrum:
features variously suited to more intensive and more extensive reading prac-
tices are included. The Wodrow Folio therefore indicates the presence of a
transitional readership which incorporates elements of both more traditional
intensive and more innovative extensive reading practices.

Crawford MS I

The punctuation system of Crawford MS I (MS Edinburgh, National Library

of Scotland, Acc. 9769 84/1/1 1/2), while still relatively sparse, is more com-
prehensive than that of the Wodrow Folio. While the Wodrow Folio only in-
cluded twenty-three litterae notabiliores within the selected extract, the paral-
lel extract from Crawford MS I includes thirty-one litterae notabiliores (which
does not greatly differ from the frequency of litterae notabiliores found in the
Wodrow Folio, as Crawford MS I contains twenty-two more lines), along with
twelve puncti and one distinctio (the distinctio belongs to a system of punctua-
tion labelled the distinctiones – developed in Antiquity – in which puncti are
placed at different heights in ascending order of importance; the distinctio was
“a high point used to indicate a final pause”).39 Further, outside of the extract,
the scribe of Crawford MS I also employs double puncti and puncti positioned
at various heights (distinctiones), indicating not only a higher quantity of punc-
tuation within this witness but more diversity of punctuation marks and greater
specificity of pause lengths.40
The difference between the paratextual provision of the two manuscripts
is more significant. Like the Wodrow Folio, Crawford MS I is divided into the
conventional five monarchs’ sections and the “Addition” found across the
majority of witnesses of Pitscottie’s Cronicles (although, unusually, this wit-
ness does not differentiate the “Addition”; it is labelled as a monarch’s section
for James VI and I). However, unlike the Wodrow Folio, Crawford MS I consis-
tently uses intertitles and enlarged initials to mark the beginning of each mon-

39
PARKES, Pause and Effect, pp. 303-304.
40
See SMITH, “Punctuating Mirk’s Festial”, p. 5: ‘punctuation’ derived from ‘pointing’, the
system of marks being employed to provide intonation guidance for priests during oral delivery.
P ARKES, Pause and Effect, p. 65, states that “according to the grammarians these pauses were
assigned arbitrary time values, the main feature of which is that they were graded in relation to
each other”.
How the Page Functions 55

arch’s section, along with varying quantities of enlarged / emboldened text and
an “Exclamation” (a short summary, in verses, of the life of the specified mon-
arch) to mark the end of the section. The separate sections of the Cronicles are
therefore more clearly identifiable for the reader of Crawford MS I, and the
degree of pause after each monarch’s section is strongly emphasised, aiding the
reader’s comprehension of the text. This textual layout would benefit an unfa-
miliar, more extensive reader in that it provides clear guidance as to the struc-
ture of the text and indicates when a significant change in content is about to
take place, so as to aid comprehension and prevent misunderstanding and to act
as a finding aid for non-linear readers.
The scribe further uses paratext to aid an anticipated more extensive reader
by including scribal formal marginalia and scribal (though since cropped)
running titles in order to continually guide the reader throughout the reading
process. The recurrent aid that features such as this provide is especially useful
in the period in which silent, extensive reading practices were emerging but
were not yet established as the primary reading practice; during this period,
readers were attempting to employ some extensive reading practices but may
not have yet fully acquired the skills to do so, making repeated guiding features
regarding the content or structure of an unfamiliar text particularly useful to an
emerging extensive readership.
The most prominent paratextual feature of this witness, though, is its struc-
ture within the conventional monarchs’ sections: this manuscript is the only
known manuscript witness of the Cronicles to divide the whole text into chap-
ters.41 This is especially interesting if the early date (1598) assigned to it by
Mackay is taken into account.42 The comprehensive system of textual division
provided by the chapters in this manuscript is a clear aid for a more extensive
reader of the text. It breaks the lengthy text of the monarchs’ sections into
much smaller, more manageable sections so readers can read shorter quantities
of material before a significant pause for reflection, while the sections remain
self-contained in regard to content, which minimises misinterpretation for an
unfamiliar reader. To further aid a more extensive reader, each chapter in this
manuscript begins with a short introductory paragraph which summarises the

41
Several manuscript witnesses of the work include chapters to sub-divide the first
monarch’s section (documenting the reign of James II): MS Edinburgh, National Library of
Scotland, Adv. 35.4.11; MS Edinburgh, University of Edinburgh Library, La.III.583; MS
Edinburgh, University of Edinburgh Library, La.III.198; and MS Edinburgh, National Library of
Scotland, Acc. 9769 84/1/1 2/2 (Crawford MS II).
42
MACKAY, The Historie and Cronicles of Scotland, p. LXX.
56 FRANCESCA L. MACKAY

content of the subsequent body of text. The provision of introductory sections

of text, coupled with the scribal formal marginalia provided in this manuscript,
which summarises the textual content in the margin alongside the relevant
section of text, indicates that the producer of this manuscript envisaged a read-
ership that had no prior familiarity with the text. The scribe is therefore em-
ploying several coordinating paratextual features to aid the extensive reading
process by providing consistent reminders of the content in order to aid com-
prehension. Further, each of the chapters within this manuscript is numerically
labelled, therefore providing a finding aid for a more extensive readership that
is reading the Cronicles in a non-linear fashion. This reading practice would be
further aided by the short introductions to each chapter which would ensure
that the reader had recalled the correct chapter number for the desired content
before they commence reading the chapter itself.
Despite Crawford MS I including several items of paratext which cater for
a more extensive readership, this manuscript contrastingly also includes en-
larged initials to begin each page (even if the page begins mid-sentence) which
– due to their traditional function in this position as aide-memoires – are strong
indicators of a more intensive readership.43 While the scribe of the Wodrow
Folio did not provide consistent markers of textual division to aid a more ex-
tensive reader effectively, it was noted that they also did not provide distinctive
visual markers for the page to act as an aide-memoire for a more intensive
reader. Crawford MS I, however, provides both. Each page begins with an en-
larged initial which could act as a mnemonic device for an aural, more inten-
sive reader to prompt their recollection of the content of the page of text due to
an association between the visual appearance of the page and the content
through frequent reading of the text. However, the inclusion of this feature
could potentially be detrimental to the reading process of a more extensive
reader, as it could cause confusion within the system of enlarged initials as
textual division, because enlarged initials are additionally employed to empha-
sise the first word of every page visually – words which are not significantly
distinct from the last words of the previous page in regard to content. This
could lead a reader who was unfamiliar with the content of the text (a more
extensive reader) to place pauses or interpret a shift in topic in inappropriate
positions, resulting in misinterpretation. The use of enlarged initials at the
beginning of every page is therefore usually fairly indicative of a scribe cater-
ing for a more intensive readership.

43
CARRUTHERS, The Book of Memory, p. 226.
How the Page Functions 57

As mentioned, in addition to using enlarged initials at the beginning of

each page, the scribe of Crawford MS I also consistently employs enlarged
initials as a feature of textual division. This manuscript uses enlarged initials
not only for the incipit of each monarch’s section but also for several other
features of textual division (for instance monarchs’ sections intertitles, chap-
ters’ introductory paragraphs, exclamations), but the enlarged initials which are
employed as features of textual division are usually larger than those used to
begin individual pages.44 As the scribe of Crawford MS I displays such system-
atic use of enlarged initials (and as several other features of the paratextual
system of this manuscript suggest that the scribe was catering for a more exten-
sive readership), perhaps, while traditionally functioning to aid more intensive
reading practices, the inclusion of enlarged initials at the beginning of every
page of Crawford MS I could have been employed due to what the scribe per-
ceived as the conventional form of the Cronicles’ textual tradition rather than
being an active decision to aid more intensive reading practices.45 Therefore it
is possible that the scribe has not considered that the use of this feature in this
position may hinder the reading process of his anticipated more extensive read-
ership. Indeed, in the transitional period of early modern Scotland’s literacy
practices, the reading public may not yet have been reading extensively enough
for the dual purposes of this feature to be a hindrance. Instead, the use of the
feature in two positions which can each be traditionally associated with inten-
sive and extensive reading practices, is perhaps further evidence of the success-
ful coexistence of traditional and emerging practices in transitional periods of
history: the anticipated readership still has sufficient understanding of the
function of the feature in both positions so as not to misunderstand its meaning
(i.e. not to interpret each new page as a significant division in the text).
A further contrast within Crawford MS I is the presence of a complex and
thorough paratextual system while the punctuation system remains relatively
sparse. This is additional evidence of the transitional literacy environment of
early modern Scotland, and indicates how, by acknowledging the nuances of
the punctuation practices and paratextual systems used to cater for the multi-
plicity of practices in use, the precise literacy environment of the anticipated

44
The initials used within the intertitles of the monarchs’ sections are 21-29 mm and those
used to begin the content of monarchs’ sections are 11-86 mm, whereas the initials which are used
at the beginning of pages (mid-sentence) are approximately 7-20 mm.
45
For example, this usage also occurs in three other manuscripts of the Cronicles (MS
Edinburgh, National Library of Scotland, Acc. 9769 84/1/1 2/2; MS Edinburgh, National Library
of Scotland, 185; MS Edinburgh, National Library of Scotland, 3147).
58 FRANCESCA L. MACKAY

readership can be identified. It could be suggested, therefore, that, while the

scribe of Crawford MS I was catering for a more extensive readership than that
of the Wodrow Folio, it was still relatively early in the development of these
practices. Readers were therefore still learning how to employ these emerging
practices in relation to traditional reading practices, and, simultaneously,
scribes were learning how to cater textually for the emerging practices through
the employment of new features of mise-en-page and the adaptation of tradi-
tional forms. Therefore, while the scribe of Crawford MS I provides some guid-
ing features for the emerging extensive reading practices (for example, a con-
sistent system of division for the monarchs’ sections, chapters, formal margina-
lia, running titles, and more comprehensive punctuation than is found in the
Wodrow Folio); the (overall comparatively sparse) punctuation system and use
of enlarged initials as aide-memoires do not seem to be catering for the same
degree of extensive reading.
The contrasting features within Crawford MS I suggest that punctuation and
paratextual systems function slightly differently. Qualitative analysis has indi-
cated that there is not always a direct correspondence between the quantity of
provision and the degree of guidance provided; therefore the contrast between
the comprehensive paratextual system and sparse punctuation system may not
actually suggest that the scribe misunderstood the degree of more extensive
reading practices for which he was catering when he produced this witness’s
punctuation system. During the editing process the insertion of punctuation is
a substantial intervention, therefore scribes catering for a different readership
to that of the copy-text may have been reluctant to interfere too heavily with
the text by adding a large quantity of punctuation. This consideration is partic-
ularly relevant to Crawford MS I due to the hypothesis that an emerging exten-
sive readership is being catered for by the scribe. The reader is therefore likely
to be unfamiliar with the Cronicles and, as such, though some guidance is
necessary, too much intervention could potentially lead to misinterpretation of
the text rather than providing assistance as a reading aid. Crawford MS I in-
cludes litterae notabiliores and punctuation marks indicating two different
lengths of pause, therefore there has been some attempt to balance this issue
through the provision of some features of guidance, but the punctuation system
provided is neither as qualitatively comprehensive nor as experimental as
found in several other witnesses of Pitscottie’s Cronicles.46

46
For example, MS Edinburgh, National Library of Scotland, Adv. 35.4.11 and MS
Edinburgh, National Library of Scotland, 185 each include litterae notabiliores, punctuation
How the Page Functions 59

The suggestion, therefore, is that Crawford MS I is catering for a distinctly

different – but coexisting – readership to that catered for by the Wodrow Folio.
The development of literacy practices was not a process of linear chronological
change from a homogeneous society of aural, intensive readers to one of silent,
extensive readers; instead, readers engaging in different reading practices coex-
isted in contemporary society, and reading practices were positioned on a spec-
trum rather than existing successively.

Freebairn’s 1728 Edition

The punctuation system of the extract from Freebairn’s 1728 edition is

comprehensive – containing five different punctuation marks: litterae notabi-
liores (sixty-nine), puncti (seven), double puncti (one), semi-cola (six), and
commata (twenty-four) – but, notably, not as comprehensive as the systems
found in some manuscript witnesses of the work (for example, MS Edinburgh,
National Library of Scotland, 185, as outlined above). It is, however, slightly
more detailed than the punctuation systems of the subsequent printed editions
of 1749, 1778, and 1814, which each include four different punctuation marks
within the selected extract. The punctuation system of Freebairn’s 1728 edition
therefore indicates the next crucial stage in the use of extensive, silent reading
practices. As discussed throughout this chapter – and argued by many scholars
–, as more extensive reading practices became increasingly more widely used
throughout the late medieval and early modern periods, scribes attempted to
cater for this practice by providing increasingly more explicit and thorough
guidance on the material page.47 However, as suggested by Jajdelska, by the
early eighteenth century more readers than ever before were employing exten-
sive reading practices (i.e. frequently reading silently in solitary reading envi-
ronments, regularly encountering unfamiliar texts, and reading directly from
the material page with no prior knowledge of the text).48 As a result readers
potentially became ‘skilled’ in the practice of reading extensively and no lon-
ger required such explicit guidance on the page in order to successfully en-
counter an unfamiliar text. The comprehensive punctuation system of the first

marks indicating four different lengths of pause, and novel combinations of marks.
47
See PARKES, Pause and Effect; JAJDELSKA, Silent Reading; and SMITH, “Punctuating
Mirk’s Festial”.
48
JAJDELSKA, Silent Reading.
60 FRANCESCA L. MACKAY

printed edition of the Cronicles (a high quantity but low variation of punctua-
tion marks) suggests that Freebairn is catering for this hypothesised ‘skilled’
extensive readership: a readership which, though still requiring guidance as to
where to pause in order to correctly interpret the content of an unfamiliar text,
did not require such detailed guidance as did earlier (less experienced) exten-
sive readers, as they had the literacy skills to comprehend the text correctly
when explicit guidance was not provided. For example, readers themselves
could potentially now distinguish between the meaning of similar lengths of
pause in different positions without the need for a wide range of different
marks to do so. This suggested pattern (the gradual increase in a ‘skilled’ ex-
tensive reading public and concomitant decrease in the complexity of guidance
required) seemingly continued throughout the early modern period. It explains
why the punctuation systems of Urie’s 1749 edition, Cadell’s 1778 edition, and
Dalyell’s 1814 edition are subsequently slightly less comprehensive than Free-
bairn’s first edition.49
Freebairn’s 1728 edition employs a system of textual division which re-
mains consistent throughout, and consistency is a trait which continues
throughout the printed editions of the work in the eighteenth and nineteenth
centuries. Intertitles are used to introduce every monarch’s section within the
text and they are presented identically on the page every time, an enlarged
initial is used to begin every monarch’s section, and marginal items such as
running titles and pagination occur on every page. Further, each feature of
textual division employed within Freebairn’s 1728 edition is used within a
single environment, rather than the same feature being employed in multiple
environments with various (and often quite distinct) functions, as is found in
several manuscript witnesses of the text. For example, whereas in Crawford MS
I enlarged initials were used both at the beginning of each page of the manu-
script and for the incipit of each of the monarchs’ sections, the witnesses of
Freebairn’s 1728 edition only employ enlarged initials at the beginning of a
new monarch’s section. This enables the enlarged initials within Freebairn’s
edition to have a clear representational value which makes it easier for the
reader to understand what the feature is indicating both as an individual marker
of textual division within the system of textual division and within the overall

49
JAJDELSKA, Silent Reading discusses how silent reading practices reached a critical mass
in the eighteenth century, after which such practices were actively catered for by the producers
of texts.
How the Page Functions 61

paratextual system in relation to other features of the page.50 For example, it

allows enlarged initials to function as an indicator of a significant change in the
content of the text without confusing the significance of the feature by also
employing the feature in other positions with a different purpose. This clarity
of function would be a primary concern when catering for a more extensive
readership unfamiliar with the text. As discussed, by this stage in the develop-
ment of literacy practices, Freebairn anticipated a readership that used prac-
tices positioned significantly towards the extensive end of the spectrum of
reading practices. Therefore, not only have the producers of texts learnt that
more extensive, silent reading practices need to be catered for by guidance on
the material page, but also that inconsistency is detrimental to successful exten-
sive reading.
As discussed throughout, the literacy practices of late medieval and early
modern Scottish readers existed within a spectrum of more intensive and more
extensive reading practices, with the nuances of the specific practices em-
ployed by the various readerships being indicated by factors such as the quan-
tity, variety, and consistency of the guiding features (and a balance between
these three factors) provided by the texts’ producers. While several of the
manuscript witnesses of the Cronicles contain a high quantity and variety of
paratextual features, Freebairn’s printed edition presents a simplified system by
slightly reducing the quantity and variety of features but increasing the consis-
tency of their use. By providing a balanced quantity, variety, and consistency
of paratextual features, Freebairn provides a significantly clear form of guid-
ance for an unfamiliar reader of the text and therefore caters for a more exten-
sive readership than is indicated by the manuscript witnesses.
Further, Freebairn’s 1728 edition is the first version of the Cronicles to
divide the text consistently into paragraphs throughout, which aids a more
extensive reader of the text by providing frequent, appropriate positions to
pause for interpretation and clear segregation of content to aid understanding.
The majority of manuscript witnesses of the Cronicles do not include para-
graphs, and those that do only make use of them sporadically, therefore dis-
playing the long, gradual period of transition during which extensive reading
practices became increasingly more widely used and more recognised by the
producers of texts.51 As can be seen from the different uses of paragraphs

50
See Anttila’s discussion of ‘isomorphism’ in R. ANTTILA , Historical and Comparative
Linguistics (Amsterdam and Philadelphia, PA, 1989), p. 407.
51
According to PARKES, Pause and Effect, p. 10, “scribes deployed features of layout to
62 FRANCESCA L. MACKAY

across the various witnesses of the work (c. 1575-1814), versions of the Cron-
icles were variously created for readers across the spectrum of intensive and
extensive reading practices. Further reinforcing the relationship between the
consistency of features and more extensive reading practices, as discussed
above, a pattern across the manuscripts and printed witnesses of the Cronicles
can be interpreted: the lack of paragraphs indicates a more intensive reader-
ship; the inconsistent use of paragraphs indicates some awareness of a more
extensive readership; and the consistent use of paragraphs indicates an antici-
pated readership significantly towards the extensive end of the reading spec-
trum.
The paratextual system of Freebairn’s 1728 edition is relatively simple and
unobtrusive. In accordance with the needs of the ‘skilled’, extensive readership
for which Freebairn was catering, it includes the lowest quantity of features
necessary to ensure an easy reading process and correct understanding of the
text. Further, the paratextual features within this witness do not interfere too
abruptly with the text itself. For example, there are no formal marginalia in this
edition, a feature which, if present, would frequently require the reader to
transfer their eye line between the main text and the marginal position of the
marginalia, causing a disruption to the reading process. Further, aside from a
standardised system to introduce each of the conventional monarchs’ sections
with a centralised, enlarged intertitle atop a new page, there are few markers of
textual division; for example, there are no intertitles positioned mid-section. By
the early eighteenth century, the anticipated readership – as suggested above –
is likely to have been skilled enough in the processes of reading extensively to
read successfully in silent, private environments. Freebairn therefore caters for
this anticipated readership, that would be reading directly from the material
page, by causing as little distraction as possible for both the eye and the mind
by only including paratextual features in positions / functions that are entirely
necessary to ensure a successful reading of the text, and making them as clear
and unobtrusive as possible when they are included.

indicate major divisions or sections of a text, such as chapters and paragraphs, a practice which
seems to go back as far as the second century B.C.”. While the rubrication marks capitulum and
the pilcrow were the conventional symbols of rubrication by which to mark paragraphs during the
Middle Ages, spacing techniques such as blank space and indentation began to be used alongside
– and as an alternative to – this rubrication during the latter part of this period. Gradually
throughout the early modern / modern periods spacing became the primary method of marking
paragraph structures. The witnesses of the Cronicles only mark paragraphs – when employed –
with spacing techniques.
How the Page Functions 63

The content of the first printed edition of Pitscottie’s Cronicles further

reinforces the suggestion that the printer / editor was creating a text for an
established more extensive readership. It includes a considerable quantity of
prefatory material, including the “Author’s Account to the Reader” and the
“Verses to the Bishop” (as are frequently found in manuscript witnesses of
Pitscottie’s Cronicles), and it also includes items original to the printed form
such as Freebairn’s “Printer’s Preface” and the “Subscription List”.52 At the
most basic level, the inclusion of prefatory, supplementary, and / or additional
material suggests a more extensive reader encountering the text in a private
reading environment, as non-essential surrounding material is possibly less
likely to have been orated in a public reading environment or to have been
deemed worthy of memorisation. Instead, this ‘extra’ material is only likely to
have been read by a reader who was encountering the material text directly
himself, within a private, silent reading environment in which it was suitable
to take the time to read this non-essential material. Further, this material is
likely to have been of more interest to an extensive reader; more intensive
readers read the same small number of texts frequently, whereas more exten-
sive readers are likely have been more receptive to encountering new surround-
ing texts such as those supplied in selected manuscripts and printed witnesses
of the Cronicles.

Conclusion

This chapter exemplifies how features of mise-en-page can be interpreted

as valid indicators of the contemporary reading practices for which a scribe or
compositor was catering, and suggests that diachronic analysis of such features
can be used to pose broader hypotheses on the literacy practices of early mod-
ern Scotland. It also presents a methodology which, in its basic structure, is
transferable to other studies which examine the relationship between the mate-
rial form and socio-cultural uses of written texts within their historical context.
Through the analysis of one specific example from the history of Scot-
land’s material culture (manuscripts and early printed editions of Robert Linde-
say of Pitscottie’s Historie and Cronicles of Scotland), this chapter has aimed
to provide plausible insights into the reading practices – and, further, the range

52
Subscription lists were a method of pre-modern book publication whereby members of
the public would subscribe to a forthcoming work in order to ensure its publication.
64 FRANCESCA L. MACKAY

of these said practices – which were employed by early modern Scottish read-
ers. This chapter argues for the recognition of the extremely nuanced reading
practices of historical readers. The combined analysis of the punctuation and
paratextual features within an individual witness of the work indicates that,
rather than discussing literacy practices in terms of universals, specific readers
are positioned within a spectrum of literacy practices and employ more or less
intensive and extensive, oral, aural, and silent practices, rather than explicitly
one or another of these practices. These terms are not mutually exclusive, and
the findings of this study reinforce the current school of thought which seeks
to banish the presentation of the practices as dichotomous.53 For example,
despite medieval society having been labelled as predominantly oral in prac-
tice, readers are known to have employed oral, aural, and silent reading prac-
tices in various contexts: Saenger suggests that “while private, silent reading
became increasingly pervasive in the thirteenth, fourteenth and fifteenth centu-
ries, public lectures continued”.54 Similarly, it has been suggested that ‘silent’
reading environments were not necessarily wholly silent; medieval ‘silent’,
solitary readers were known on occasion to murmur the words of the text
aloud, or to remain silent but shape the spoken words with their lips.55 As
Jajdelska has persuasively argued, silent readers hear an internal voice
subvocalising the words of the text, therefore there is still an element of orality
within silent reading – it has merely become internalised.56
Though not specifically dated, the two potentially near-contemporary
manuscript witnesses of the Cronicles discussed here display quite significant
differences in regard to the readerships for which they cater. The results show
that the transition in literacy practices in early modern Scotland was extremely
gradual, and that multiple coexisting but disparate readerships were active
within single periods of time. Therefore, while overall early modern Scotland
gradually became increasingly more extensive in practice (as evidenced by the
hypothesised ‘skilled’ extensive readership of Freebairn’s 1728 edition), the
history of literacy cannot be interpreted as a clear uni-directional evolution.
The reality is a much more complex literacy environment in which coexisting
but distinct readerships functioned and interacted. This chapter has sought to

53
AKINNASO, “The consequences of literacy”, p. 166; CHARTIER, The Cultural Uses of
Print, pp. 5-7; DARNTON, “Towards a history of reading”, p. 92; COLEMAN, Public reading, pp.
xii, 15-16; FOX, Oral and Literate Culture, pp. 5, 39, 50; JAJDELSKA, Silent Reading, pp. 7, 21.
54
SAENGER, Space Between Words, pp. 258-259.
55
S.R. FISCHER, A History of Reading (London, 2003), p. 91.
56
JAJDELSKA, Silent Reading, pp. 45-46.
How the Page Functions 65

recognise and display this over-arching multiplicity of reading practices in

early modern Scotland:

if we accept the multiple forms in which our artifacts have been transmitted, we
may recognise that medieval culture did not simply live with diversity, it cultivated
it [...] we need to embrace the consequences of that diversity, not simply live with
it, but to situate it squarely within our methodology.57

57
NICHOLS, “Introduction: Philology in manuscript culture”, pp. 8-9.
 Verbal and Visual Communication in Title Pages
of Early Modern English Specialised Medical Texts

MAURA RATIA and CARLA SUHR

Introduction

o the eyes of the modern reader, early modern title pages look perplex-

T ing. They seem to contain either too much or too little information, often
both. They might have edifying yet overly general mottos, such as
“Reade without preiudice: Iudge without partialitie”.1 They may contain illus-
trations that do not have anything to do with the contents. Sometimes title
pages even fail to mention the author. Visually they are a mishmash of every-
thing from black letter type to unadorned and simple roman type side by side
with elaborate swash capitals. Title pages were the means for the printers to
show the repertoire of their skills. More importantly, they were advertisements

1
F. HERING, A modest defence of the caueat giuen to the wearers of impoisoned amulets,
as preseruatiues from the plague wherein that point it somewhat more largely reasoned and
debated with an ancient physician, who hath mainteined them by publicke writing: as likewise
that vnlearned and dangerous opinion, that the plague is not infectious, lately broched in
London, is briefly glansed at, and refuted by way of preface, by Fr. Hering D. in Physicke
(London, 1604), title-page.

......................................................................................................................................
Verbal and Visual Communication in Early English Texts, ed. Matti PEIKOLA, Aleksi
MÄKILÄHDE, Hanna SALMI, Mari-Liisa VARILA, and Janne SKAFFARI, Utrecht Studies in Medieval
Literacy, 37 (Turnhout: Brepols, 2017), pp. 67-93.

DOI 10.1484/M.USML-EB.5.114131
68 MAURA RATIA and CARLA SUHR

of the time, posted on street-corners to attract more readers, highlighting spe-

cific words or topics that were considered important or that printers saw fit to
advertise, as it was commonly the printer’s job to produce them.2
The present study is based on our previous work on title pages of plague
treatises of the Stuart period and of sixteenth- and seventeenth-century witch-
craft pamphlets and news pamphlets.3 In this paper we examine early modern
medical texts, but rather than attempting to cover the whole register of medical
writing, we limit the scope to medical texts with a specific focus instead of a
more generalist approach. Though still a small-scale study, in comparison to
these earlier linguistic studies of textual labels that focused on either a specific
topic or a specific (sub)genre, the material nonetheless covers a wider range of
topics and genres, and is thus the beginning for a more general view of the
connections between visual and verbal communication in the title pages of the
early modern period.
Our pragma-philological or pragma-linguistic study has the following aims.
First, we will describe the visual features of title pages to see if or when visual
highlighting coincides with textual elements that describe the genre or the
topic. Second, we aim to investigate what gets highlighted and to see whether
there are any diachronic developments. Methods used in this study combine
macro and micro approaches which are descriptive rather than quantitative
because of the still small sample of texts.

2
E.F. SHEVLIN, “‘To reconcile book and title, and make ’em kin to one another’: The
evolution of the title’s contractual functions”, Book History 2 (1999), pp. 42-77, at p. 52; C.
SULLIVAN, “Disposable elements? Indications of genre in early modern titles”, Modern Language
Review 102 (2007), pp. 641-653, at pp. 645-646; R.W. MCCONCHIE, “Some reflections on early
modern printed title-pages”, in: Principles and Practices for the Digital Editing of Diachronic
Data, ed. A. MEURMAN-SOLIN and J. TYRKKÖ (Helsinki, 2013: Studies in Variation, Contacts and
Change in English 14).
3
M. RATIA, “Investigating genre through title-pages: Plague treatises of the Stuart period
in focus”, in: Principles and Practices for the Digital Editing of Diachronic Data, ed. A. MEUR-
MAN-SOLIN and J. TYRKKÖ (Helsinki, 2013: Studies in Variation, Contacts and Change in English
14); C. SUHR, Publishing for the Masses: Early Modern English Witchcraft Pamphlets (Helsinki,
2011: Mémoires de la Société Néophilologique 83); C. SUHR, “News and relations: Textual labels
in the titles of early modern news pamphlets”, in: Sociocultural Dimensions of Lexis and Text in
the History of English, ed. P. PETRÉ, H. CUYCKENS and F. D’HOEDT (Amsterdam and Phila-
delphia, forthcoming).
Verbal and Visual Communication in Title-Pages 69

Background

Title pages were introduced to printed texts only in the early sixteenth
century, but at first their primary functions were to protect the unbound sheets
of a volume and to advertise unsold volumes to prospective buyers perusing a
bookseller’s shop.4 At first, the most important information on the title page
was the title and the name of the printer (who was usually also the publisher of
the text), but gradually the range of information displayed on the cover in-
creased to include the name of the publisher and / or bookseller, the author, in-
formation about previous editions, and descriptions of the contents of the book.
The layout of the page changed, too: sixteenth-century title pages often con-
tained elaborate borders, illustrations, flourishes and printer’s devices and the
text was arranged into patterns such as inverted triangles, but by the seven-
teenth century title pages had become less ornate.5 Seventeenth-century titles
continued the earlier tradition of being long and comprehensive enumerations
of the (purported) contents of the book, though a new characteristic was the
short and catchy headline.6
According to Gérard Genette, the title page of a book is a part of a text’s
paratext, the textual and non-textual elements that surround the text proper and
guide a reader’s approach to a text and his reception of the text.7 Specifically,
in Genette’s classification, the title page belongs to the ‘publisher’s peritext’
which includes all paratextual elements that are “the direct and principal (but
not exclusive) responsibility of the publisher (or perhaps, to be more abstract
but also more exact, of the publishing house)”, along with elements such as
indices and other referential devices. The publisher’s peritext also includes
features that “constitute the book’s material realization”, such as the selection
of type, typesetting, layout and the choice of paper; these are termed ‘physical
peritext’ in Genette’s framework.8 In the sixteenth- and seventeenth-century
printing house it was the compositor who decided on the layout of pages, both
of the text proper and the title page, according to his own judgement of the

4
M.M. SMITH, The Title-Page: Its Early Development 1460-1510 (London, 2000), p. 16;
M. BLAND, A Guide to Early Printed Books and Manuscripts (Hong Kong, 2010), p. 66.
5
BLAND, A Guide, p. 66; G. GENETTE, Paratexts: Thresholds of Interpretation, trans. J.E.
LEWIN (Cambridge, 1997), p. 33.
6
SHEVLIN, “%To reconcile’”, p. 55.
7
GENETTE, Paratexts, p. 33.
8
GENETTE, Paratexts, p. 33-34.
70 MAURA RATIA and CARLA SUHR

author’s meaning but also of the readers’ capabilities to process the informa-
tion. As Joseph Moxon wrote in his guide on the art of printing in 1683,

A good Compositor is ambitious as well to make the meaning of his Author intelli-
gent to the Reader, as to make his Work shew graceful to the Eye, and pleasant in
Reading: Therefore if his Copy be Written in a Language he understands, he reads
his Copy with consideration; that so he may get himself into the meaning of the
Author, and consequently considers how to order his Work the better both in Title
Page, and in the matter of the Book: As how to make his Indenting, Pointing,
Breaking, Italicking, &c. the better sympathize with the Authors Genius, and also
with the capacity of the Reader.9

However, the moderately-sized English printing community had certain

established conventions for layout,10 such as highlighting foreign quotations by
a switch in type, and individual printing houses also had their own conventions
within those set parameters, depending on what their resources were in terms
of, for instance, type, illustrations or symbols for “pointing”.11 Thus many of
the features mentioned by Moxon as being under the purview of the compositor
were in fact guided by convention as well as the compositor’s judgement. In
addition, there is some evidence that reprints, even when produced by different
printers, reproduced fairly closely the layout of the previous edition.12 It made
sense to copy the visual look of a book that was enough of a best-seller to re-
quire a new printing. It also made the compositor’s task easier when he could
simply follow the layout of the copy-text (assuming the format of the book did
not change).
Authors, then, did not decide on the layout of the title page. The extent to
which authors had control even over the titling of their texts is not clear. Elea-
nor Shevlin argues that authors had practically no control over what their texts
were called on the title page of the book, which led to titles being simple vehi-

9
J. MOXON, Mechanick exercises: Or, the doctrine of handy-works: Applied to the
compositors trade, 2 vols. (London, 1683), 2, p. 220.
10
BLAND, A Guide, p. 119.
11
J. TYRKKÖ, “Printing houses as communities of practice: Orthography in early modern
medical books”, in: Communities of Practice in the History of English, ed. J. KOPACZYK and A.H.
JUCKER (Amsterdam and Philadelphia, 2013), pp. 151-176, at p. 155.
12
J. TYRKKÖ, V. MARTTILA and C. SUHR, “The Culpeper Project: Digital editing of title-
pages”, in: Principles and Practices for the Digital Editing of Diachronic Data, ed. A. MEUR-
MAN-SOLIN and J. TYRKKÖ (Helsinki, 2013: Studies in Variation, Contacts and Change in English
14).
Verbal and Visual Communication in Title-Pages 71

cles for marketing the text rather than valid descriptors of its content, but Ceri
Sullivan takes a more moderate view and relegates complete lack of control
over titles only to pirated texts.13 For the purposes of this paper, it is not neces-
sary to know who decided on the title, but to keep in mind that the decision to
highlight certain words in the title was made by the printer as part of title page
design, and thus printing conventions and marketing considerations likely
influenced these decisions.
There is a connection between the general layout of the early modern title
page and the titling of texts, for it has been argued that at least until the 1640s,
the visual layout of the title page of certain kinds of texts was more important
as an indicator of the type of content of the book than the verbal elements of
the title. Illustrations and choice of typeface, for example, seem to be early
signals of popular texts,14 and it is only in the 1640s that highlighting of textual
labels such as ‘news’ or ‘account’ in popular news pamphlets became common.
This shift from visual to verbal cues has been suggested to have taken place
because the new semi-literate audiences had become familiar enough with
these texts to be able to identify textual labels and what they implied about the
contents of the texts.15 This is why visual cues focusing on the layout of the
title page as a whole could be replaced with highlighted verbal cues, or textual
labels. Maura Ratia examined textual labels and their accuracy as genre mark-
ers. Textual labels proved to be quite accurate in predicting the contents – with
the exception of religious argumentation, which was not necessarily advertised
on the title page.16
In this paper we conduct a preliminary investigation of texts that have a
wider range of readership, ranging from literate treatises to advertisements of
proprietary medicines, to see whether the earlier findings about textual labels
in new popular texts apply to more established texts with seasoned readers
familiar with all kinds of texts. As the highlighting of textual labels in these
earlier studies has been explicitly connected with a newly literate readership,

13
SHEVLIN, “%To reconcile’”, p. 52; SULLIVAN, “Disposable elements?”, pp. 645-666.
14
For romances, see R.S. LUBORSKY, “Connections and disconnections between images and
texts: The case of secular Tudor book illustrations”, Word & Image 3 (1987), pp. 74-85; for
Chaucer, see S. ORGEL, “Textual icons: Reading early modern illustrations”, in: The Renaissance
Computer: Knowledge Technology in the First Age of Print, ed. N. RHODES and J. SAWDAY
(London and New York, 2000), pp. 59-94; for witchcraft pamphlets, see C. SUHR, Publishing for
the Masses.
15
SUHR, Publishing for the Masses; SUHR, “News and relations”.
16
RATIA, “Investigating genre”.
72 MAURA RATIA and CARLA SUHR

we anticipate highlighted verbal cues to appear earlier in more established

genres.
The term ‘textual label’ is rather complex and wide, covering an array of
various types of discourse.17 Due to the ongoing development of genres in the
medicine of the seventeenth century,18 texts often consisted of a mixture of
different genre conventions, and some labels only later became genre terms.
Established genre labels at the time were, for instance, ‘regimens’ and ‘surgical
texts’, originating from the medieval period. The genre label ‘treatise’, albeit
common, was not yet fully established during this period and it could refer to
any formal discussion on a matter. Many of the common textual labels of the
time are very general and uninformative, for instance ‘discourse’, ‘passages’,
and ‘reflections’. A number of labels refer to interactive discourse forms, for
instance ‘dialogue’, ‘questions’, or ‘queries’.19 In addition to an interactive
function, some labels also carry with themselves a pragmatic function, namely
‘apology’, ‘caveat’, and ‘defence’. A few labels refer to text types according to
Werlich’s typology.20 For example, ‘advice’, ‘directions’, ‘instructions’,
‘method’, ‘rules’, and ‘advertisements’ all represent the text type of ‘instruc-
tion’, whereas ‘history’ or ‘historical account’ point to the text type of ‘narra-
tion’.
Topical labels refer to the topics that are dealt with, such as ‘plague’ or
other specific diseases or medicines, or they might refer to the intended audi-
ence. For instance, ‘midwifery’ texts were of particular interest to these medi-
cal practitioners more than other types of medical professionals or lay audi-
ences. The borders between topical and textual labels are fuzzy: ‘medicines’,
‘remedies’, and ‘antidotes’, for example, are topical labels referring to content,

17
C. CLARIDGE, “Pamphlets and early newspapers: Political interaction in news reporting”,
in: English media texts past and present: Language and textual structure, ed. F. UNGERER
(Amsterdam and Philadelphia, 2000: Pragmatics and Beyond, New Series 80), pp. 25-43; RATIA,
“Investigating genre”; SUHR, Publishing for the Masses, p. 140; SUHR, “News and relations”;
SULLIVAN, “Disposable elements?”.
18
I. TAAVITSAINEN, “Discourse and genre dynamics in early modern English medical
writing”, in: Early Modern English Medical Texts: Corpus Description and Studies, ed. I. TAA-
VITSAINEN and P. PAHTA (Amsterdam and Philadelphia, 2010), pp. 29-53.
19
CLARIDGE, “Pamphlets”, pp. 28-29; RATIA, “Investigating genre”; on dialogue: G. FRITZ,
“Topics in the history of dialogue forms”, in: Historical Pragmatics, ed. A.H. JUCKER (Amster-
dam and Philadelphia, 1995), pp. 469-498, at pp. 471-472.
20
E. WERLICH, A Text Grammar of English (Heidelberg, 1976), pp. 39-41.
Verbal and Visual Communication in Title-Pages 73

but they also indicate the presence of ‘recipes’ (the text type of ‘instruction’)
or ‘recipe paraphrases’ (the text type of ‘description’).21
The third type of label identified in this study is the headline, which Elea-
nor Shevlin characterises as “brief, suggestive, and sometimes racy wording to
attract readers”.22 In medical books, alternate or double titles – a subset of
headlines according to Shevlin – are much more common than suggestive
wording. Double titles often take the form of a Latin or Greek label followed
by an English translation, as in “Speculum Ægrotorum: | THE SICKE-|mens
Glasse”, or a catchy phrase with an explanatory postmodifying element, as in
“Little Venus Unmask’d: | Or, a perfect | DISCOVERY | of the | FRENCH POX”.
Headlines could also consist of a combination of the author’s name and a text-
ual label, as in “HARWARDS | Phlebotomy”.

Materials and Methods

This study investigates title pages of specialised medical texts, that is to

say, of texts concentrating on a specific topic rather than explaining theoretical
medical principles, giving general health advice, or collating recipes for var-
ieties of illnesses. The register of vernacular medical writing was a thriving one
in the early modern period: alongside continuing medieval genres such as regi-
mens, recipe collections, and surgical texts, medical writing in this period also
reflects the changing views of scientific inquiry, the discoveries of new thera-
peutic substances and new diseases, new medical fields such as military medi-
cine, as well as widening audiences of medical texts.23 In their analysis of the
kinds of books in the Elizabethan marketplace of print that were popular (in the
economic sense of the ratio of printed titles), Alan B. Farmer and Zachary
Lesser categorise medicine as an “innovative” genre that consisted mainly of

21
I. TAAVITSAINEN, “Changing conventions of writing: The dynamics of genres, text types,
and text traditions”, European Journal of English Studies 5 (2001), pp. 139-150; R. CARROLL,
“Middle English recipes”, in: Medical and Scientific Writing in Late Medieval English, ed. I.
TAAVITSAINEN and P. PAHTA (Cambridge, 2004), pp. 174-191; M. MÄKINEN, “Herbal recipes and
recipes in herbals – Intertextuality in early English medical writing”, in Medical and Scientific
Writing in Late Medieval English, pp. 144-173; M. RATIA, Texts “Con and Pro”: The Early
Modern Medical Controversy over Tobacco (Helsinki, 2011: Mémoires de la Société
Néophilologique 82), pp. 192-200; RATIA, “Investigating genre”.
22
SHEVLIN, “%To reconcile’”, p. 55.
23
H. MIKKELI and V. MARTTILA, “Change and continuity in early modern medicine (1500-
1700)”, in: Early Modern English Medical Texts, pp. 13-28.
74 MAURA RATIA and CARLA SUHR

new titles rather than reprints of established older titles, though the results for
medicine actually indicate that the ratio of completely new titles and reprints
is about even.24 Thus medical texts are a good choice for an in-depth analysis
of textual labels – they offer a good mixture of old and new titles, and repre-
sent evenly both tradition and novelty.
The macro-level analysis makes use of sixty-nine title pages of specialised
medical texts printed between 1525 and 1700.25 The texts are included in the
category of treatises on specific topics in the corpus Early Modern English
Medical Texts (EMEMT).26 This corpus category covers specific diseases (six-
teen title pages), specific methods (nineteen title pages), specific therapeutic
substances (fifteen title pages), midwifery and children’s diseases (ten title
pages), and plague texts (nine title pages).27
The EMEMT corpus was designed to be a representative sample of medical
writing in early modern England. The texts included in the corpus have been
carefully chosen to reflect the range of genres, topics, and traditions character-
istic of the period, and the sizes of the six corpus categories (and the subcate-
gories) are also indicative of the contemporary textual world.28 The category of
treatises on specific topics accounts for about 30% of the entire corpus, and it
contains a wide variety of texts written mostly but not solely by medical pro-
fessionals of different kinds of social and educational backgrounds, for audi-
ences ranging from specialists to the general population.29 The topics covered
in the texts continue old traditions (for instance phlebotomy) but also innova-
tions of the period (for instance advertisements for proprietary medicines).

24
A.B. Farmer and Z. Lesser, “What is print popularity? A map of the Elizabethan book
trade”, in: The Elizabethan Top Ten: Defining Print Popularity in Early Modern England, ed.
A. Kesson and E. Smith (London and New York, 2013), pp. 19-54, at pp. 48, 42.
25
The chronological distribution of the sixty-nine texts is as follows: five texts from 1525-
1550, fifteen from 1551-1600, twenty-two from 1601-1650, and twenty-seven from 1651-1700.
26
I. TAAVITSAINEN, P. PAHTA, M. MÄKINEN, T. HILTUNEN, V. MARTTILA, M. RATIA, C.
SUHR and J. TYRKKÖ, Early Modern English Medical Texts (Amsterdam and Philadelphia, PA,
2010); P. PAHTA and M. RATIA, “Category 2: Treatises on specific topics”, in: Early Modern
English Medical Texts, pp. 73-100.
27
A few texts have more than one sample in the corpus, but their title-pages are only
counted once in this study. When multiple text samples of a text have been placed in different
corpus subcategories, the title-page is counted in the subcategory mentioned first in the title. One
text, 1699_Colbatch, has been taken from a collection of essays; in this case, the title-page of the
collection is analysed here rather than the title-page of the essay included in the corpus.
28
P. PAHTA and I. TAAVITSAINEN, “Introducing early modern English medical texts”, in:
Early Modern English Medical Texts, pp. 1-8, at pp. 4-5.
29
PAHTA and RATIA, “Category 2”.
Verbal and Visual Communication in Title-Pages 75

While it must be kept in mind that the EMEMT corpus was constructed for lin-
guistic purposes, it nonetheless also gives a good indication of changes and
continuity in medical publishing of the period. The title pages of the texts in
this category were produced by many different printers, so they reflect broad
conventions rather than specific printing house practices. Therefore the title
pages of the texts selected for the corpus can be used to chart developments in
title page design in this field.
The macro-level analysis of larger trends in medical title pages is supple-
mented by a case study of 30 texts exemplifying medical plague writing from
the period 1603-1666. Stuart England was heavily afflicted with recurring
plague epidemics and the impact of plague on both the individual and the soci-
etal level was immense because of the appalling mortality rates. Plague trea-
tises were published sporadically, during or right after the visitations.30 Due to
the prevalence of religious discourse, plague treatises differ from other medical
texts of the early modern period.31 Nevertheless, the genre is not a coherent
whole: it consists of a variety of texts that are tied together by a common topic
and often nomenclature. All texts consist of multiple discourse forms – recipes,
for example, can be found in all plague treatises, even amidst religious dis-
course. The texts represent different styles of writing from learned discussions
to collections of practical advice. The question of audience is intriguing: some-
times references are made on the title page to the general public, for instance,
“Published for the Benefit of all FAMILIES” or “of the poore of this City”, but
there are only a couple of short mentions in the texts to prepare affordable
remedies. Latin and sometimes Greek citations in the long title might also
suggest that the intended audience was mainly other physicians.
The method adopted for both macro- and micro-analyses of texts started
with the qualitative assessment of visually prominent title page features such
as the use of illustrations, decorative borders, or a variety of typefaces, as well
as visual highlighting of specific words or expressions in relation to the whole
title. By highlighting we refer to larger type size, differences in typeface, and
the use of all capital letters, italics, or underlining. The focus is, therefore, on
typographical highlighting of specific elements as opposed to considering the
title page as a pleasing whole making use of illustrations or geometric / sym-

30
A. WEAR, Knowledge and Practice in English Medicine, 1550-1680 (Cambridge, 2000),
pp. 275, 277-278; see PAHTA and RATIA, “Category 2”, pp. 95-96.
31
P. SLACK, The Impact of Plague in Tudor and Stuart England (London, Boston, Mel-
bourne and Henley, 1985), p. 38.
76 MAURA RATIA and CARLA SUHR

metric patterning of text. We also aim to investigate what gets highlighted:

textual label, topical label or headline – and to see whether there are any dia-
chronic developments. This information was collected into a spreadsheet. All
highlighted elements were then categorised as textual labels, headlines or topi-
cal labels. When transcribing titles in this article, we retain the original use of
capital letters, italics, and underlining. Line divisions are indicated by a verti-
cal line “|”. We do not distinguish between all capitals or small capital letters,
or attempt to reproduce swash characters, black letter typeface, different type-
face sizes or empty space. However, these features have all been included in
the analysis, and will be described when relevant.
Highlighting more than one element of the title is frequent in the material,
but we made no distinction of whether an element appeared first in a title or
after some other element; in other words, we did not categorise labels as pri-
mary (i.e. the first element) or secondary (i.e. coming after the first element).32
In many cases the first element, though clearly highlighted with regard to the
bulk of the title, was nonetheless in a smaller-size typeface than the second or
even third highlighted elements, which downplays the importance of the first
element in relation to the ones after it. Alternately, several elements could be
highlighted in the same manner, as in Fig. 1 (in the section ‘Specialised medi-
cal texts 1525-1700’ below), which makes it difficult to evaluate which is the
most important highlighted element. Usually, the relative size of the typeface
was our primary means for identifying highlighted elements, but the practice is
not without problems, for it relies on our subjective assessments of the sizes of
lines rather than exact measurements.33 A significantly large typeface size was
taken as a clear indication of highlighting, but sometimes the differences in
size were not clear-cut. The size of the typeface has to be considered together
with differences in typeface, as a word or line in small roman capitals can seem
bigger than a line in regular italic, for example – and who is to say that an early
modern reader, or even a modern reader, would agree with our assessments. At
times it was hard to determine whether an element is highlighted or simply part
of an overall title page design, especially if a title page contains short lines in
many different typefaces or follows a design of incrementally decreasing sizes
for each subsequent line. In these cases we made use of cues such as splitting
words over two lines, which indicates the importance of visual design over

32
See SUHR, Publishing for the Masses, pp. 133-139.
33
For an example of a study aiming for a more objective analysis by relying on exact
measurements, see TYRKKÖ, MARTTILA and SUHR, “The Culpeper Project”.
Verbal and Visual Communication in Title-Pages 77

textual considerations. Conversely, the placement of articles, prepositions and

what we call ‘tags’34 on separate lines (often in a much smaller typeface), cre-
ates empty space that automatically gives prominence to the lines it surrounds
and in this way signals that the printer seeks to emphasise certain words rather
than create a visually pleasing block of text.
Once we had completed the identification and categorisation of all high-
lighted elements, we analysed the data for any emerging trends or patterns in
visual and verbal communication both on the macro level of specialised medi-
cal texts and the micro level of plague treatises of the Stuart period. For the
macro-analysis of texts, a simple division of highlighted versus not highlighted
was used instead of placing highlighted elements in any order of significance,
whereas for the micro-analysis a more detailed approach was adopted. We turn
next to our results.

Results and Analysis

Specialised Medical Texts 1525-1700

We start the macro-level analysis of specialised medical texts of the early

modern period by giving an overview of the developments in their title page
layout. The earliest title pages are all in black letter, as expected, and feature
the inverted triangle figure in arranging the text. The first changes appear in the
1550s: the first line or lines of the title begin to look slightly different from the
rest of the text. These lines are printed in slightly bigger size type, or using
roman or italic typeface rather than black letter. Nonetheless, the lines consist
of several words, and words are still regularly broken over two lines, indicating
that the variation in size and typeface are functions of title page design rather
than emphasis. After the 1580s black letter disappears from title pages, but the
variation in the lines still continues, with a new feature of using small capitals
in some lines. In addition, the text is no longer arranged in a tight block of text,
rather there are several groupings of text and more empty space on the page.
Intricately decorated borders are fairly common, as are arabesques and fleurons
at the beginnings of some lines or at the bottom tip of an inverted triangle,

34
Tags are words (or phrases in some cases) that indicate that what follows is an alternate
title or fuller description of the contents. Examples of tags are conjunctions such as ‘or’ and ‘and’,
adverbs like ‘also’ and ‘wherein’, and verbs like ‘showing’ or ‘containing’.
78 MAURA RATIA and CARLA SUHR

though they are not found consistently in all title pages. Only two of all the title
pages had illustrations; it is noteworthy that both of these texts were uroscopies
and portrayed a doctor “reading” a urine sample on the title page. They are also
very early texts, as one is from 1525 and the other from 1547. All in all, the
typographical developments of the 20 sixteenth-century title pages of our sam-
ple match up with the findings for title pages of literary texts in the same pe-
riod. These general developments have been linked to the fact that in the 1580s
there was a generational shift in master printers, and the new printers were
ready for typographical experimentation.35 The new typographical develop-
ments were picked up more slowly in popular texts such as Chaucer’s texts, in
romances and plays, in ballads, and in news pamphlets; primers were printed
in black letter until the 1660s.36 Clearly, then, sixteenth-century specialised
medical texts were aimed at a limited readership consisting of literate and
educated people, despite the occasional claims that the material covered in a
text are “for euery person to loke in”.37
It is only around the turn of the seventeenth century that individual words
or expressions begin to be systematically highlighted using size and different
typefaces, though there are isolated instances already in the 1550s. These early
instances are most likely cases where the short length of the lines gives the
impression of highlighting words, when in reality decreasing size or typeface
switches are simply part of the title page design. This suggestion is supported
by the fact that, in these early texts, words are still broken over two lines as in
the T. Phayer’s 1546 “THE KEGI|ment [sic!] of life where|unto is added...”, or
non-content words are emphasised, as in the 1556 book on physiognomy by
Cocles that has the first line in bigger typeface and thus seems to highlight the
35
M. BLAND, “The appearance of the text in early modern England”, Text 11 (1998), pp. 94-
107.
36
For Chaucer, see S. ORGEL, “Textual icons”; for romances and plays, see C. MISH, “Black
letter as a social discriminant in the seventeenth century”, Periodical of the Modern Language
Association 68 (1953), pp. 627-630; for ballads and primers, see A. MCSHANE, “Typography
matters: Branding ballads and gelding curates in Stuart England”, in: Book Trade Connections
from the Seventeenth to the Twentieth Centuries, ed. J. HINKS and C. ARMSTRONG (New Castle,
DE, and London, 2008), pp. 19-44); for news pamphlets, see SUHR, Publishing for the Masses,
and SUHR, “News and relations”.
37
T. MOULTON, This is the Myrrour or Glasse of helth necessary and nedefull for euery
person to loke in, that wil kepe their bodye from the syckenesse of the Pestilence, and it sheweth
how the planettes do raygne in euery houre of the daye and nyghte, with the natures and
exposicio~s of the XII sygnes, deuyded by the .XII. Monethes of the yeare, and shewed the remedies
for many dyuers infirmities and dyseases that hurteth the bodye of Manne, 5th edn. (London,
1540?), title-page.
Verbal and Visual Communication in Title-Pages 79

words “A brief and” rather than words that would help a reader identify what
kind of text he was looking at. The first title page that can be argued to high-
light elements in the title on purpose is Timothy Bright’s 1586 “A | TREATISE OF
| MELANCHOLY”, where the first three lines are not only capitalised (two of them
with swash capitals), but they are also significantly larger in size than the rest
of the lines of text in the title page. The placement of the first article on a line
of its own is likely a result of the large typeface and the length of the words:
the words could not have been fitted into even two lines without decreasing the
size of the typeface. The alternation of swash and regular roman capitals may
have had an aesthetic rather than a highlighting function, but the decision to
use a large typeface for the first three lines seems a conscious highlighting
choice.38

Morbus Anglicus :
OR,
THE ANATOMY
OF
CONSUMPTIONS.
CONTAINING
The Nature, Causes, Subject,
Progress, Change, Signes, Progno-
sticks, Preservatives ; and several
Methods of Curing all Consumptions
Coughs, and Spitting of Blood.
With Remarkable Observations touching the
same DISEASES:
To which are Added,
Some brief Discourses of Melancholy, Mad-
ness, and Distraction occasioned by Love.
Together with certain new Remarques
touching the Scurvy and Ulcers of the Lungs.

Fig. 1 The layout of the title of Gideon Harvey’s Morbus Anglicus (1666).

38
This particular imprint (STC (2nd edn.) 3748) was printed by John Windet, one of the
innovative new generation of printers who, along with his contemporary John Wolfe, has been
praised for the elegance of his title-pages (see M. BLAND, “The appearance of the text”, p. 103).
Early English Books Online contains an image of the imprint printed by Thomas Vautrollier in
the same year (STC (2nd edn.) 3747); the layout is very similar but not identical – for example the
swash letters appear only in Windet’s imprint.
80 MAURA RATIA and CARLA SUHR

The highlighting in seventeenth-century title pages is more easily identified

as intentional, as articles and conjunctions are often printed in smaller type
faces on separate lines above and / or below the most important words that are
highlighted not only by their bigger or different typeface but also by the empty
space generated by the mostly empty lines around them. An example of this is
reproduced crudely in Fig. 1 above.
The three lines that stand out the most are the same size but in different
typefaces, which are most likely simply chosen for aesthetic purposes. What
makes the lines stand out in addition to the size and variety of the typeface in
comparison to the bulk of the text is the empty space generated by the centred
tag words “or” and “containing” as well as the preposition “of”, especially
when the rest of the title is crammed rather tightly together. The example also
shows the convention of switching typeface for foreign terms, place names and
names of authors, diseases or specific medical substances. As the majority of
seventeenth-century title pages are filled with long titles that describe the con-
tents of the text, with information about the author’s credentials, epigrams as
well as the printer’s or stationer’s information, the contrast of the short high-
lighted lines in comparison to the blocks of smaller text is striking.
Clear-cut, established textual labels are not as common as we expected,
though they are the most common ones: ‘discourse’ was highlighted only twice
(though it appeared a further four times without highlighting), and ‘treatise’
was highlighted six times (and appeared another four times without highlight-
ing). All other textual labels are only found a few times in the sample. Other
highlighted textual labels are more ambiguous, and some of them also imply
the contents or purpose of the text. For example, ‘method’, ‘advertisement’,
‘directions’, and ‘directory’ indicate the instructive aim of the texts, so a reader
will expect to get advice on a medical topic. ‘Anatomy’ could be considered a
conventional medical genre label for texts that explains human anatomy in a
specific way, but we see the use of the label broadened for explaining illnesses
such as melancholy or consumption; the choice of label drew on the prestige of
the learned genre, but it also implied that the discussion of these illnesses is as
systematic and detailed as the descriptions of the human body. The common
contemporary tropes of discovering secrets and looking in mirrors is reflected
in labels such as ‘discovery’, ‘key’, ‘secret’, and ‘glasse’. All in all, there is a
wide range of textual labels found in the sample, reflecting both the variety of
genres and the width in scope of the texts from general to specific, and from
learned to practical. Textual labels were often accompanied by highlighted
Verbal and Visual Communication in Title-Pages 81

topical labels that elaborated on the content of the text, perhaps to offset either
a very general genre label or a textual label such as ‘method’ that was poten-
tially opaque to readers, as in the title of a 1685 text that gave instructions for
preventing and curing the small-pox: “A DIRECT | METHOD | Of Ordering and
Curing People of that Loathsome Disease, the | SMALL-POX” where the textual
label ‘method’ and the topical label ‘small-pox’ are printed in much bigger
typeface than other words in the title, thus letting the readers see at a glance the
most important information about the contents of the text.
Textual labels become less common as highlighted elements in the second
half of the seventeenth century, whereas headlines gain popularity. Especially
frequent are constructions such as “HARWARDS | Phlebotomy”, where the au-
thor’s name is coupled with a topical label (‘phlebotomy’) which is then fol-
lowed by another explanatory label – in the case of “HARWARDS | Phlebotomy”,
by a combination of a textual label and a specification of the topic: “A Treatise
of letting of Bloud”. Authors’ names were generally not given prominence in
title pages, so when it did happen, the author was presumably so well known
that using his name was a selling point for the book.39 Another popular kind of
headline was a Latin or Greek formulation of the topic, as in “Speculum Ægro-
torum”, followed by a highlighted translation: “THE SICKE-|mens Glasse”. The
foreign headlines were intended perhaps to convey authority to the text, or to
indicate the scholarly nature of the text. Headlines containing names of authors
or foreign elements required a certain level of sophistication from the readers:
the ability to recognise a medical authority or the foreign words. Headlines
could also be made up of more or less cryptic phrases that would catch a casual
observer’s eye, as in “Little Venus Unmask’d”, or the names of the medicines
(such as “Elixir Salutis”) hawked in the text. Headlines never appear alone, so
they seem to require another highlighted element, commonly one that describes
the topic of the text (in the two examples above, a topical label followed the
headline identifying the texts as dealing with the pox and a miraculous pill sold
by Anthony Daffy).
By far the most common type of highlighted element was a topical label
that indicated that the text deals with medicine, whether it be about a specific
disease such as the plague or the gout, or specific medical substances such as
the proprietary medicines that appear in the medical marketplace in the seven-
teenth century. Topical labels can also specify who the text is intended for:

39
For a case study of Nicholas Culpeper as a brand, see TYRKKÖ, MARTTILA and SUHR,
“The Culpeper Project”.
82 MAURA RATIA and CARLA SUHR

highlighting the topic as ‘midwifery’ makes it clear that the author has a spe-
cific target audience in mind. More often than not, topical labels, sometimes
several of them, appear in conjunction with either a textual label of some sort
or a headline; they are rarely completely on their own or even in clusters.
The most striking characteristic of highlighting elements of the title in title
pages that this macro-analysis has revealed is the emerging convention of
stacking or clustering different kinds of highlighted elements. Highlighted
textual labels were often accompanied by highlighted topical labels, and head-
lines especially seem to require an explanatory postmodifying element (or
elements) that often contain highlighted textual and / or topical labels. It seems
likely that the clustering is intended to help a reader skim through the title and
to identify relevant information about the topic and perhaps also the purpose of
the topic and the level of education needed to be able to follow the argumenta-
tion of the text. The findings of the macro-analysis are, however, necessarily
only descriptions of emerging trends, so we turn next to a more in-depth analy-
sis of more focused texts.

Stuart Plague Treatises

In our micro-analysis of texts, we examined specialised medical texts deal-

ing with the plague. All available material from Early English Books Online
(EEBO) was gathered from 1603–1666, that is, plague treatises (first or second
editions) published with their own title pages. The material consists of 30 texts.
These texts were popular and a number of them were reprinted; for example
Francis Herring’s Certaine rvles, directions, or advertisments from 1603 was
republished in 1625, 1636, 1641, 1665 (the last two under the name Preserva-
tives against the plague), each correlating with the beginning of another plague
epidemic.40 The authors were physicians or, in one case, even clergymen. How-
ever, during that time it was common for divines to act as medical practitio-
ners.41 Wear claims that plague treatises were commonly short, even under
5,000 words.42 However, in our dataset only half, i.e. fifteen texts, were under
5,000 words, eleven texts were between 5,000 and 12,000 words and six texts

40
See R. TOTARO, Suffering in Paradise: The Bubonic Plague in English Literature from
More to Milton (Pittsburgh, 2005), p. 2.
41
SLACK, The Impact of Plague, p. 38.
42
WEAR, Knowledge and Practice, p. 278.
Verbal and Visual Communication in Title-Pages 83

were quite long: 20,000-29,000 words. A multitude of topics having to do with

the plague are covered in the material, often representing the generic structure
of ‘practica’, which was typical of early modern treatises on diseases.43 In
addition to causes (“both naturall and diuine”), prognostication or ‘signs’, and
treatment, quarantine and fleeing were often discussed and some texts also
contained prayers.
Table 1 presents the main part of the long title which is visually separate
from the rest of the title and features the most highlighted elements. It com-
monly ends with punctuation (see Table 1) – full stop, comma, colon, or a
semicolon – which separates the core part of the title from the subtitle, descrip-
tion of contents, possible references to the audience, author’s credentials, and
perhaps an epigraph. The names of the authors are included in Table 1, but
they are not used in the analysis of title pages as printers, instead of authors,
were generally responsible for their design. For analysis, the years of the great
epidemics (1603-1604 and 1665-1666) form their own subgroups with eight
and fifteen texts respectively (separated by horizontal lines in Table 1). The
texts in the middle group are more scattered, consisting of seven texts from
three different decades (1620s, 1630s, and 1640s).

Table 1. Distribution of plague treatises featuring the highlighted main part of

the long title and ESTC numbers.

1603_Balmford, James_STC (2nd ed.) 1338

A SHORT DIALOGUE CONCERNING THE PLAGVES INGECTION.
1603_Herring, Francis_STC (2nd ed.) 13239.5
CERTAINE RVLES, DIRECTIONS, OR ADVERTIS MENTS FOR THIS TIME OF PESTILENTIAL
CONTAGION:
1603_Hobbes, Stephen_STC (2nd ed.) 12577
A NEVV TREATISE OF THE PESTILENCE, containing the Causes, Signes, Preseruatiues
and Cure thereof.
1603_ I. W._STC (2nd ed.) 24905.7
A Briefe Treatise of the Plague VVherein is shewed,
1603_(James I)_STC (2nd ed.) 9209
ORDERS, thought meete by his Maiestie, and his Priuie Counsell, to be executed
throughout the Counties of this Realme, in such Townes, Villages, and other
places, as are, or may be hereafter infected with the Plague, for the stay of further
increase of the same.

43
WEAR, Knowledge and Practice, pp. 117, 119.
84 MAURA RATIA and CARLA SUHR

1603_Lodge, Thomas_STC (2nd ed.) 16676

A TREATISE of the Plague:
1603_Thayre, Thomas_STC (2nd ed.) 23930
A TREATISE OF THE PESTILENCE:
1604_Herring, Francis_STC (2nd ed.) 13248
MODEST DEFENCE OF THE CAVEAT GIVEN TO THE WEARERS OF impoisoned Amulets,
as Preseruatiues from the Plague:
____________________________________________________________________

1625_Bradwell, Stephen_STC (2nd ed.) 3537

A VVATCHMAN FOR THE PEST.
1630_Boraston, William_STC (2nd ed.) 3372
A NECESSARIE AND BRIEFE TREATISE OF THE Contagious disease of the Pestilence,
1630_Anon._STC (2nd ed.) 19192
A TREATISE OF THE PLAGUE,
1630_Royal College of Physicians_STC (2nd ed.) 16770
THE KINGS Medicines for the Plague.
1636_Bradwell, Stephen_STC (2nd ed.) 3536
PHYSICK FOR THE SICKNESSE, Commonly called the PLAGVE.
1640_Woodall, John_STC (2nd. ed.) / 25961
THE CVRE OF THE PLAGUE BY AN ANTIDOTE CALLES AURUM VITÆ.
1641_Sherwood, Thomas_Thomason, E.176[10]
THE Charitable Pestmaster, OR, The cure of the PLAGUE,
____________________________________________________________________

1665[?]_Anon._WING (2nd ed.) P2337

THE PLAGVES Approved PHYSITIAN.
1665_Anon._WING (2nd ed.) P3674
THE PROPHECIES, AND Predictions, FOR London's Deliverance:
1665_Anon._WING (2nd ed.) S3717
THE SHUTTING UP Infected Houses As it is practised in ENGLAND Soberly Debated.
1665_Barker, Richard, Sir_WING (CD-ROM, 1996) B778
Consilium Anti-Pestilentiale: OR, Seasonable Advice, CONCERNING Sure, Safe,
Specifick, and Experimented MEDICINES, both for the Preservation from, and Cure
of this Present PLAGUE.
1665_Bèze, Théodore de_WING (CD-ROM, 1996) B2196
A LEARNED TREATISE OF THE PLAGUE:
1665_Gadbury, John_WING (CD-ROM, 1996) G86
London's Deliverance predicted: IN A Short Discourse Shewing the Causes of
PLAGUES IN GENERAL;
1665_Kemp, W._WING (CD-ROM, 1996) K260
Verbal and Visual Communication in Title-Pages 85

A BRIEF TREATISE Of the NATURE, CAUSES, SIGNES, PRESERVATION FROM, AND CURE
OF THE Pestilence.
1665_(Kephale, Richard)_WING (2nd ed.) K330
MEDELA PESTILENTIÆ: Wherein is contained several Theological Queries
CONCERNING THE PLAGUE,
1665_M. R._WING (2nd ed.) R45
THE MEANES OF Preventing, and Preserving from, and CURING of that most Conta-
gious Disease, called the PLAGUE:
1665_(W. W.)_Not in WING.
THE Christians Refuge: OR HEAVENLY ANTIDOTES against the PLAGUE.
1665_T. D._WING (CD-ROM, 1996) D88
Food and Physick, FOR EVERY Housholder, & his Family, During the Time of the
PLAGUE.
1665_Thomson, George_WING (2nd ed.) T1026
LOIMOLOGIA. A Consolatory Advice, And some brief OBSERVATIONS Concerning
the Present Pest.
1665_W. J._WING (CD-ROM, 1996) J47
A COLLECTION OF Seven and Fifty approved RECEIPTS Good against the PLAGUE.
1665_Wharton, Thomas_WING (2nd ed., 1994) W1577
DIRECTIONS FOR THE PREVENTION and CURE of the PLAGUE.
1666_Thomson, George_WING (CD-ROM, 1996) T1027
ËOIMOTOMIA: OR THE PEST Anatomized

In the first group of texts from 1603-1604, emphasis on visual ordering of

elements on the title page instead of textual labelling is evident. For example,
in “A SHORT DIA|LOGUE” and “MODEST DE|FENCE”, words are broken over two
lines to create the inverted triangle shape. Visually the most prominent textual
labels ‘dialogue’ and ‘defence’ are divided between two lines consisting of
different-sized typefaces. What catches the reader’s attention is the first line of
the title printed with the largest type instead of the textual labels. All title
pages, except one (A briefe treatise), use the inverted triangle, or a similar type
of shape with decreasing line width, for the highlighted elements. Some texts
use the same design for the other elements of the title as well. A characteristic
of this style is that the type size decreases by each line. Three or sometimes
four different sizes are used to form the triangle shape. Sometimes the triangle
is formed only after the indefinite article ‘A’, which stands on its own in the
first line (in three texts). The article is highlighted with the biggest typeface in
“A | NEVV TREATISE | OF THE PESTILENCE”, but in two other texts the second line
is printed with a bigger type which gives emphasis to the row containing the
genre label ‘treatise’.
86 MAURA RATIA and CARLA SUHR

The most highlighted labels are mainly genre labels: ‘treatise’ occurs in
three texts as the most highlighted item; only in A new treatise of the pestilence
is the label printed with the second-largest typeface. The label ‘treatise’ ap-
pears to have been considered important by contemporary printers in plague
writing; it is also a term used by present-day historians with reference to medi-
cal plague writing of the time.44 However, as the word ‘treatise’ is commonly
the second or third element in the title, it entails that it is printed in the first or
second line on the title page which is always highlighted according to the prin-
ciples of early seventeenth-century design. Other highlighted textual labels are
found in texts A short dialogve, Modest defence and Orders. Looking at the
titles in Table 1, Orders attributed to James I stands out from the rest as the
highlighted section is three times as long as in the other texts. The text repre-
sents a different genre, but it is still part of the medical domain. Plague orders
had been put together by physicians throughout the latter part of the sixteenth
century under the instructions of the Privy Council.45 A fleuron precedes the
word ‘orders’ to stretch the first line to achieve a perfect form of the triangle.
Other highlighted textual items, but to a lesser degree, can be found in, for
instance, “CERTAINE | RVLES, DIRECTIONS, OR ADVERTISMENTS”, which point to
the text type of instruction. The premodifier ‘certaine’ gets most emphasis as
it starts the title. Later in the title Certaine Rvles (1603) also features the tex-
tual label ‘caveat’ “to those that weare about their neckes impoisoned Amulets
as a Preseruatiue from the Plague”, but it is not highlighted. The word is pre-
ceded by a text-organising device or a tag “WITH”. A reply published the next
year, “MODEST DE|FENCE OF THE | CAVEAT GIVEN TO | THE WEARERS OF |
impoisoned Amulets, as | Preseruatiues from | the Plague”, focuses on this
topic, and the label ‘caveat’ is highlighted, although not as much as elements
in the first and second rows. In the latter part of the title the tags “Wherein”
and “As likewise” are followed by textual labels ‘point’ and ‘opinion’, refer-
ring to the text type of argumentation. In Orders by James I, the long title is
arranged as two equally sized inverted triangles. Labels ‘(also / an) aduise’ and
‘(containing) rules’, preceded by tags, can be found in the first part of the sec-
ond inverted triangle drawing attention to the word ‘aduise’. The rest of the

44
TOTARO, Suffering in Paradise, p. 15; WEAR, Knowledge and Practice, p. 278.
45
M. HEALY, “Discourses of the plague in early modern London”, in: Epidemic Disease in
London, ed. J.A.I. CHAMPION (London, 1993: Centre for Metropolitan History Working Papers
Series 1), pp. 19-34, <http://www.history.ac.uk/cmh/epiheal.html>.
Verbal and Visual Communication in Title-Pages 87

triangle formation is in black letter which cannot be found elsewhere in the

material.
Tags are used before topical labels as well. However, topical labels in title
pages are not visually as prominent as textual labels, which suggests that they
were, indeed, understood differently by contemporaries. The following types
of occurrences are common: “containing the Causes, Signes, Preseruatiues and
Cure” and “Containing the nature, signes, and accidents”. These topical labels
are embedded in the text and not visually highlighted. The italics in the previ-
ous example, however, seem to be more than only a stylistic device. Italics are
also used elsewhere to draw attention to the same topical label, for instance in
Modest defence only the last two lines are italicised: “Preseruatiues from the
Plague”. The topical labels ‘preservatiues’ as well as ‘cure(s)’ carry in them-
selves specific information about the contents. They indicate the presence of
recipes or recipe paraphrases and, for this reason, it was practical to highlight
them.
The layout of title pages changes considerably when analysing the second
group of texts from 1625 to 1641. Of the seven texts, only one uses the in-
verted triangle for highlighted elements of the title. Also, words are not broken
over two lines anymore, at least not in the highlighted part of the title. The
most prominent items in the titles are not textual labels but topical labels, espe-
cially the word ‘plague’ (4 out of 7 texts). As the topical label ‘plague’, ‘pesti-
lence’, or ‘pest’ does not occur at the beginning of the title in this group of
texts, the biggest emphasis in some texts (3 out of 7) is still reserved for topical
labels or lines of text appearing earlier, which reflects the principles of design
of the earlier period. For example, “THE | KINGS | Medicines for | the Plague”
emphasises the word ‘kings’ over ‘plague’ which is written with such small
type that it is hard to spot in comparison to the topical label ‘kings’ as well as
‘medicines for’. In one of these texts the label ‘pestilence’ appearing only on
sixth row is printed with a very small type, but it is the only word in italics. In
contrast, three other texts feature the topical label ‘plague’ with much larger
type, even on fifth row. “PHYSICK FOR THE SICKNESSE, Commonly called the
PLAGUE” features the topical label ‘physic’ referring to medicines as promi-
nently as the word ‘plague’. It seems that the focus on the visual ordering of
the title had shifted towards highlighting topical labels.
During this period we also begin to see headline titles. A necessary require-
ment for these titles are postmodifying labels that describe the contents more
accurately: “A VVATCH-MAN FOR THE PEST. TEACHING The true Rules of Preser-
88 MAURA RATIA and CARLA SUHR

vation from the Pestilent Contagion” and “THE Charitable Pestmaster, OR, The
cure of the PLAGUE”. Either a part of the headline title is highlighted the most,
for instance “VVATCH-MAN”, or, as in The charitable pestmaster, the topical
label ‘plague’ is the most noticeable item, whereas the headline title receives
less emphasis.
Medicines are advertised in “THE KINGS Medicines”, “THE CVRE OF THE
PLAGUE BY AN ANTIDOTE CALLED AURUM VITÆ” (“an antidote” in decorative
swash capitals) and, indirectly, in “THE Charitable Pestmaster, OR, The cure of
the PLAGUE”, where the word ‘cure’ acts as an indicator of recipes (one of the
three chapters is dedicated to making remedies). Textual labels that have not
been emphasised with a bigger type include “(TEACHING) Rules”, “(Together
with) passages”, “(Conteining) instructions”, “(Together with a little) treatise”.
The textual label ‘description’ in The cvre of the plague is intriguing as it oc-
curs without any tag words (the only instance in the whole material) towards
the end of the long title: “The description, order, and use whereof, together
with the said Antidote [...]” referring to the author’s “Aurum vitæ”.
In general, textual labels were conveyed differently from topical labels, for
instance, sometimes tags leading to textual labels were highlighted: “TEACHING
Rules” (1625). A similar case can also be found in the previous dataset: “WITH
A caueat” (1603). In comparison, tag words were not highlighted when they
introduced topical labels. Topical labels introduced by tag words are similar to
the ones in the first group of texts: “(with the) Causes, Signes, and Cures”,
“(CONTAYNING) The Causes, Signes, Symptomes, Prognosticks, and Cure”,
“(With) Signes and Symptoms”, “(Being) cure”, “(together) Antidote”.
The last and the biggest group consists of 15 texts from 1665-1666. Head-
line titles in this subgroup are common, elaborated on for instance with tag
words “IN” and “OR” acting as visual cues: “London’s Deliverance predicted:
IN A Short Discourse Shewing the Causes of PLAGUES IN GENERAL” and “THE
Christians Refuge: OR HEAVENLY ANTIDOTES against the PLAGUE”, where the
headline part of the title has been underlined. These texts highlight the topical
label “PLAGUES” or “PLAGUE” with a larger type at the expense of the headline.
In The Christians Refuge the topical label ‘(heavenly) antidotes’ seems vague,
but it places the text firmly in the religious register.
This subgroup of texts also contains foreign titles, which are a type of
headline titles, followed by tag words and textual and topical labels: “Consili-
um Anti-Pestilentiale: OR, Seasonable Advice”, “MEDELA PESTILENTÆ: Wherein
is contained several Theological Queries CONCERNING THE PLAGUE”, “LOIMO-
Verbal and Visual Communication in Title-Pages 89

LOGIA. A Consolatory Advice” and (by the same author) “ËOIMOTOMIA: OR THE
PEST Anatomized” (the letter ‘l’ has been transcribed with the equivalent Greek
letter). Visually the most prominent labels in these titles are textual labels ‘(sea-
sonable / consolatory) advice’ and topical labels ‘plague’ and ‘pest’ due to
their larger type size. In Medela Pestilentiæ, the textual label ‘(theological)
queries’ is also highlighted, but less than the topical label ‘plague’.
The prominence of the topical labels ‘plague / plague’s’, ‘pestilence’, or
‘pest’ continues from the earlier period – these words are among the most
highlighted items in 9 texts; in 5 texts only these labels, not others, are empha-
sised as much. This indicates that these topical labels were used as organisa-
tional devices to cater to the needs of the audience.
A novel characteristic in the visual layout of the title page is that more
labels and longer stretches of the title are highlighted than in earlier periods.
For instance, in “Consilium Anti-Pestilentiale: OR, Seasonable Advice, CON-
CERNING Sure, Safe, Specifick, and Experimented MEDICINES, both for the
Preservation from, and Cure of this Present PLAGUE”, the most highlighted
items are both ‘seasonable advice’ and the topical label ‘plague’, which are of
equal type size, but the topical label “PLAGUE” printed with all capitals is what
catches the reader’s eye. The tag word ‘concerning’ is also highlighted, as well
as ‘medicines’. ‘Preservation’ and ‘cure’ are not emphasised by a bigger type
face but by their italics. Similarly, in “LOIMOLOGIA. A Consolatory Advice,
And some brief OBSERVATIONS Concerning the Present Pest” the foreign title
is highlighted less than ‘consolatory advice’ and main emphasis falls on “Pres-
ent Pest”, which is printed with the largest type and underlined, and then on
OBSERVATIONS. The foreign title “LOIMOLOGIA” is underlined as well, but the
typeface is much smaller.
New textual labels emerge in this period among the most emphasised
items: “PROPHECIES, AND Predictions” referring to the medieval tradition of
providing prognosis for illness based on planetary movements.46 Another title
advertises “A COLLECTION OF Seven and Fifty approved RECEIPTS Good against
the PLAGUE” with equal emphasis on labels ‘receipts’ and ‘plague’. “Food and
Physick” are highlighted in one title, pointing towards the regimen genre.
Plague treatises often dealt with preventative means because of the severity of
the disease and the texts contained practical, regimen-style instructions.47 In

46
I. TAAVITSAINEN, Middle English Lunaries: A Study of the Genre (Helsinki, 1998: Mé-
moires de la Société Néophilologique 47), p. 100.
47
WEAR, Knowledge and Practice, p. 301; PAHTA and RATIA, “Category 2”, p. 98.
90 MAURA RATIA and CARLA SUHR

“THE MEANES OF Preventing, and Preserving from, and CURING of that most
Contagious Disease, called the PLAGUE”, the labels ‘mean(e)s’, ‘curing’ and
‘plague’ are highlighted with decreasing emphasis. As the volume of published
treatises on plague increases, the topics also become more varied, for instance
the practice of quarantine is criticised in “THE SHUTTING up Infected Houses”
with the word ‘shutting’ as the most highlighted element. Interestingly, the
topical label ‘plague’ is not in the title, but only a reference to “the present
Infection” is made. Other highlighted textual labels in this subgroup of texts
not mentioned yet are the genre label ‘treatise’ (two occurrences) and ‘direc-
tions’ in “DIRECTIONS (FOR THE) PREVENTION and CURE of the PLAGUE”, where
the topical label ‘plague’ receives as much emphasis as ‘directions’, and the
topical label ‘prevention’ is printed with a bigger type than ‘cure’.
Religious textual or topical labels begin to emerge in this period. In Medela
pestilentiæ, the textual label ‘theological queries’ refers to the discourse form
of the text, whereas the title “THE Christians Refuge: OR HEAVENLY ANTIDOTES
against the PLAGUE” retains the headline type of quality as the topical label
‘(heavenly) antidotes’ is rather vague.48 In three other texts religious genre
labels appear closer to the bottom of the page, for instance, above imprint and
separated by horizontal lines, or right above a biblical quotation: “Chiefely, a
Godly and Penitent PRAYER unto Almighty God”, “Also some Prayers, and
Meditations upon Death”, “Together with several Prayers and Meditations”.
The genre labels ‘prayer(s)’ and ‘meditations’ are highlighted in the first two
titles. In “A LEARNED TREATISE OF THE PLAGUE: WHEREIN Questions” the label
‘questions’ refers to both medical and religious matters: “Whether the PLAGUE
be Infectious, or no: And Whether, and how farr it may be shunned of Chris-
tians, by going aside? are resolved”.
Overall, the material becomes increasingly varied in the group of texts
from 1665-1666: some title pages continue to present a detailed description of
the contents on their title pages whereas others are very concise. Ëoimotomia,
for example, boasts a long list of textual labels on the title page: ‘historical
account’, ‘reflections’, ‘observations’, ‘directions’, ‘apology’, and ‘word’. All
the shorter titles in the whole dataset – A briefe treatise of the nature and
Loimologia included in toto in Table 1 – are from this period. This complexity
can be seen in title page design as well. For example, Medela pestilentiæ fea-
tures at least nine different typefaces as well as italics for highlighting through-

48
A plague sermon by John Featly was published during the same year titled A Divine
Antidote against the Plague (1665).
Verbal and Visual Communication in Title-Pages 91

out the title, which is in stark contrast to title pages from the early seventeenth
century making use of a much smaller variety of types.
The use of italics in the data set in general is rather consistent. Author’s
credentials and place names are frequently in italics or, even more often, the
author’s name is in small capital letters which stand out in comparison to the
surrounding italic typeface. Foreign titles are commonly in italics and also in
upper case; headline titles can also contain italicised elements. In plague trea-
tises of later periods epigraph or motto quotation (when present) is always in
italics. Later texts also feature types of larger italic swash capitals and more
variety in the use of typeface. Direct references to audience also emerge, which
can also be written in italics, for instance “Fitted for the Poorer sort” (in Di-
rections for the prevention and cure of the plague). A specific usage for italics
can be found in a number of texts: for tag words preceding textual or genre
labels “WITH a caveat” (1603), “TEACHING Rules” (1625), “Chiefely, a Godly
and Penitent PRAYER” (1665) and “Learned treatise: WHEREIN, The two Ques-
tions” (1665). Highlighting these tag words suggests that textual labels were
differentiated from topical labels, since it was only done for textual labels. This
practice of highlighting tag words was, in fact, common in early modern news
pamphlets.49

Discussion

Verbal and visual communication in title pages of medical texts had a

unique development: in our analysis of medical texts individual textual labels
were highlighted early on rather systemically. This verbal turn, based on our
findings, happened earlier in medical texts than in popular pamphlets, that is,
not in the 1640s but already at the turn of the seventeenth century. Medical
texts addressing literate and also professional audiences were perhaps among
the first ones to highlight textual labels and in this way to rely on their audi-
ence’s ability to recognise these labels. This development, however, appears to
have taken place a little later in plague treatises, i.e. in the group of texts from
1625-1641. In comparison to other medical texts, plague treatises seem to rep-
resent a more conservative design and, in our data, their title pages often made
use of the inverted triangle to organise the main part of the title (in texts from
1603-1604), which practice was already becoming less frequent in other medi-
49
SUHR, “News and relations”.
92 MAURA RATIA and CARLA SUHR

cal texts. Across the data, established genre labels were used surprisingly little,
instead ambiguous labels were abundant such as ‘method’, ‘directions’, and
‘observations’.
In the course of the seventeenth century, topical labels informing the read-
ers about the contents became the most highlighted elements at the expense of
textual labels. The emphasis on textual instead of topical labels in early seven-
teenth-century plague treatises could also, however, stem from the fact that
textual elements occur in the title before topical labels: ‘(new / briefe / -) trea-
tise’, ‘(short) dialogue’, ‘(certaine) rules’, ‘orders’, ‘(modest) defence’. The
layout of the page with the inverted triangle and decreasing type size entailed
that topical labels would not be highlighted, at least not by using a bigger type
size. Italics could be used for that function, but in our material it happened
seldom. When italics were used, the entire line was highlighted, even if it
broke up words, conforming to contemporary aesthetic principles of design.
This practice of highlighting the entire line and, at times, breaking up words
can still be seen in some later texts (from 1665) alongside new practices of
highlighting specific labels instead of whole lines. In later texts, however,
individual topical labels were highlighted in italics to a greater extent, for in-
stance “Preservation and Cure” (Consilium Anti-Pestilentiale, 1665) and
“Antidotes, Signes and Symptoms” (Medela Pestilentiæ, 1665). Especially the
labels ‘cure’ and ‘antidotes’ must have appealed to the readers as they were
used to advertise medicines and recipes. Actual advertisements on proprietary
medicines started to appear only in the latter part of the seventeenth century.50
One distinctive feature emerging in both data sets is the ‘stacking’ of high-
lighted elements, resembling the practice found in other genres such as popular
pamphlets.51 Overall, more gets highlighted than before: in addition to textual
labels, topical labels are highlighted. Various types of visual emphasis are used
in comparison to earlier texts. Underlining elements, for instance foreign titles
and parts of a headline title, only appears in the material from 1665 onwards.
In earlier texts, horizontal lines were sometimes used to distinguish the imprint
from the title. Later the same practice was adopted to highlight a biblical cita-
tion right before the imprint or sometimes the author’s credentials or the in-
tended audience: “Published for the Benefit of all FAMILIES”. In the early part
of the century highlighted elements are often in upper case, at least for the first

50
See M. RATIA and C. SUHR, “Medical pamphlets: Controversy and advertising”, in: Early
Modern English Medical Texts, pp. 180-201, at pp. 182-183.
51
See SUHR, Publishing for the Masses; SUHR, “News and relations”.
Verbal and Visual Communication in Title-Pages 93

two or three lines of the title, but this practice changes and becomes more
varied in later texts with sometimes only topical labels, such as ‘plague’, in
upper case and also printed with the biggest type. So in the latter part of the
seventeenth century, different types of highlighting co-occurred and, at the
same time, highlighting intensified as a greater variety of types became avail-
able for printers.

Conclusion

Our study on the title pages of early modern specialised medical texts
verified our initial hypothesis: the shift from visual to verbal communication in
these title pages of more literate texts took place earlier than it did in popular
news pamphlets. The macro-analysis places the shift at the turn of the sixteenth
and seventeenth centuries, but the micro-analysis of plague treatises gives a
slightly later date. These results indicate that this shift was gradual and adopted
at different paces in different kinds of texts; further analyses of other specific
types of medical texts are clearly needed for a more detailed view of the devel-
opments in this register of writing. Both analyses show that medical texts em-
ployed a wide range of textual labels, which reflects the variety of genres cov-
ered by medical writing in this period, but likely also signals the level of learn-
ing and sophistication of the majority of the expected consumers of the books.
Diachronic trends could also be identified for the seventeenth century, when
the turn to verbal communication had taken place. Textual labels were high-
lighted more in the early part of the century, but this practice gave way to high-
lighting headlines and topical labels, both of which were characteristically
stacked together with other highlighted elements.
Our study shows that medical texts addressing literate and professional
audiences started to highlight textual labels earlier than popular texts aimed at
less educated readers, but research should be done on other literate genres as
well. For example, title pages of literary books have a similar timeline when it
comes to changes in layout features, but there is as yet no study of highlighting
textual elements that could verify whether the changes in visual features are
connected to a shift to more verbal communication in these title pages.
 Quantifying Contrasts: A Method of
Computational Analysis of Visual Features
on the Early Printed Page*

JUKKA TYRKKÖ

Introduction

ccording to G. Thomas Tanselle, the “visual display of language” has

A been the object of increasing interdisciplinary attention over the last

few years.1 In historical linguistics, too, the pragmaphilological study
of paratextual features in manuscripts and early printed books has enjoyed
something of a renaissance in recent years,2 with various corpus linguistic and

*
This study was conducted while the author was funded by the Institute for Advanced
Social Research, University of Tampere.
1
G.T. TANSELLE, Bibliographical Analysis: A Historical Introduction (Cambridge, 2009),
p. 61. See, e.g., T. HIIPPALA, Modelling the Structure of a Multimodal Artefact (doctoral thesis,
University of Helsinki, 2013) <http://urn.fi/URN:ISBN:978-952-10-9427-9>
2
See, e.g., C. SUHR, Publishing for the Masses: Early Modern English Witchcraft Pamph-
lets (Helsinki, 2011: Mémoires de la Société Néophilologique de Helsinki 83); R. CARROLL, M.
PEIKOLA, H. SALMI, M. VARILA, J. SKAFFARI, and R. HILTUNEN, “Pragmatics on the page: Visual
text in Late Medieval English books”, European Journal of English Studies 17 (2013), pp. 54-71;

......................................................................................................................................
Verbal and Visual Communication in Early English Texts, ed. Matti PEIKOLA, Aleksi
MÄKILÄHDE, Hanna SALMI, Mari-Liisa VARILA, and Janne SKAFFARI, Utrecht Studies in Medieval
Literacy, 37 (Turnhout: Brepols, 2017), pp. 95-122.

DOI 10.1484/M.USML-EB.5.114132
96 JUKKA TYRKKÖ

digital editing projects developing increasingly sophisticated models for anno-

tating the non-textual features of the page along with the textual ones.3 To
incorporate the visual and material aspects of the original documents into the
corpus edition, some recent corpora have included facsimile images of the
original artefacts, while others have introduced layered annotations that record
visual and material features such as layout, hand and script switches, changes
of typeface in printed documents, illustrations, decorations and other graphic
elements, and so on. However, while detailed manual annotations are superbly
useful in opening up entirely new windows onto the material dimensions of the
original documents, there is a substantial drawback when it comes to the time
and effort that the painstaking annotation of minute features requires.4 Another
perhaps less frequently cited disadvantage concerns the types of information
that manual annotators can comfortably record. While it is reasonably easy,
though not effortless, to assign categorical group membership to a given item
– for example, to record that it is positioned in the margin, or whether or not it
is a decorative element of the page – it is much more labour intensive to count
all elements – not merely words – on the page or to record their sizes in a pre-

J. TYRKKÖ, “Printing houses as communities of practice: Orthography in early modern medical

books”, in: Communities of Practice in the History of English, ed. J. KOPAZCYK and A. JUCKER
(Amsterdam, 2013), pp. 151-176; A. MEURMAN-SOLIN, “Taxonomisation of features of visual
prosody”, in: Principles and Practices for the Digital Editing and Annotation of Diachronic
Data, ed. A. MEURMAN-SOLIN and J. TYRKKÖ (Helsinki, 2013: Studies in Variation, Contacts
and Change in English 12) <http://www.helsinki.fi/varieng/series/volumes/14/meurman-
solin_c/>; J. TYRKKÖ, V. MARTTILA and C. SUHR, “The Culpeper Project: Digital editing of title-
pages”, in: Principles and Practices for the Digital Editing and Annotation of Diachronic Data,
ed. A. MEURMAN-SOLIN and J. TYRKKÖ (Helsinki, 2013: Studies in Variation, Contacts and
Change in English 12) <http://www.helsinki.fi/varieng/series/volumes/14/tyrkko_marttila_suhr/>;
T. WALKER and M. KYTÖ, “Features of layout and other visual effects in the source manuscripts
of An Electronic Text Edition of Depositions 1560-1760 (ETED)”, in: Principles and Practices for
the Digital Editing and Annotation of Diachronic Data, ed. A. MEURMAN-SOLIN and J. TYRKKÖ
(Helsinki, 2013: Studies in Variation, Contacts and Change in English 12) <http://www.helsinki.
fi/varieng/series/volumes/14/walker_kyto/>; C. CLARIDGE, “From page to screen: The relevance
of encoded visual features in the Lampeter Corpus”, in: Principles and Practices for the Digital
Editing and Annotation of Diachronic Data, <http://www.helsinki.fi/varieng/series/volumes/14/
claridge/>. See also the Pragmatics on the Page website at <https://www.utu.fi/en/units/hum/
units/English/research/projects/Pages/Pragmatics-on-the-Page.aspx>.
3
See V. MARTTILA, Creating Digital Editions for Corpus Linguistics: The Case of Potage
Dyvers, a Family of Six Middle English Recipe Collections (doctoral thesis, University of Hel-
sinki, 2014) <http://urn.fi/URN:ISBN:978-951-51-0060-3>, pp. 145-146, 177.
4
See S.R. REIMER, “Unbinding Lydgate’s Lives of Ss. Edmund and Fremund”, in: The
Book Unbound: Editing and Reading Medieval Manuscripts and Texts, ed. S. ECHARD and S.
PARTRIDGE (Toronto, 2004), pp. 169-189, at p. 176.
Quantifying Contrasts 97

cise manner.5 Luckily, these are precisely the types of features that modern
computers are very capable of analysing quickly and efficiently.
In recent years, new computational methods of image analysis have been
developed in rapid response to the needs of diverse fields such as natural sci-
ences, medicine and artificial intelligence, among others. In the humanities,
too, computational image analysis has become a useful tool in archaeology, the
forensic analysis of documents, art and cultural history, and media studies, to
name a few fields. So far their use has been limited in historical linguistics, but
it is to be expected that, as scholarly interest in visual and paratextual features
of historical texts gains more ground, and as the data sets grow larger and lar-
ger, making manual analysis increasingly difficult, these methodologies will
become commonplace in our field as well. Accordingly, in this chapter I will
concentrate on the role and practical use of computational analysis of visual
features, introducing a simple but effective method for quantifying a variety of
visual features using ImageJ, an image processing tool originally developed for
medical imaging, such as radiological image analysis, at the National Institute
of Health, and Imageplot 1.0, an ImageJ plugin developed by Software Studies
Initiative for the analysis of cultural image data.6 To illustrate the methods in
practice, I will use a small collection of early and late modern surgical books,
as well as one specific book in particular, to give two examples of how the
analytical procedure can be used to discover visual trends in the body text of
the page over three hundred years.7 Applications of the same basic methodol-
ogy to other research questions will also be discussed, along with some illustra-
tive examples thereof.

5
For discussion of the manual measuring of visual elements from early printed books, see
TYRKKÖ et al., “The Culpeper Project”.
6
For ImageJ, see C.A. SCHNEIDER, W.S. RASBAND and K.W. ELICEIRI, “NIH Image to
ImageJ: 25 years of image analysis”, Nature Methods 9 (2012), pp. 671-675. For Imageplot 1.0,
see L. MANOVICH, “How to compare one million images?”, in: Understanding Digital Human-
ities, ed. D. BERRY (New York, 2012), pp. 249-278.
7
The data, drawn from an ongoing project exploring the correlation between linguistic and
visual complexity, should be considered a sample. The first public presentation of the findings
took at the IAUPE conference in London, June 2016.
98 JUKKA TYRKKÖ

Text and Paratext8

Although the recent rise of paratextual interests in historical linguistics can

be attributed in part to new digital archives and repositories that have afforded
ever-growing numbers of scholars easy access to digital facsimiles, as well as
advances in text annotation and digital humanities methodologies, it is impor-
tant to emphasise the theoretical and ideational significance of the paratextual
dimension when it comes to the study of both language and books. Numerous
scholars have pointed out that focusing on the text alone means leaving out a
considerable part of the communicative event that a full physical document is.
According to Bonnie Mak, the physical page “transmits ideas ... but more sig-
nificantly influences meaning by its distinctive embodiment of those ideas”,
while Ruth Carroll et al. go even further, arguing that “the appearance of the
page is integral to the reader’s construal of meaning”.9 Indeed, while most
studies of historical paratext take the text as the primary layer of information
and may then elaborate on it with additional quantitative features,10 it is also
possible to re-imagine this fundamental and seemingly natural order of things
by beginning the analysis with an unsupervised computational examination of
layout features, the relative prominence of printed characters and other visual
features on the page, and treating textual information as a second, albeit equal-
ly important, layer.11 Mark Bland illustrates this with the example of an
eighteenth-century poem, “printed in folio and set in large type with obvious
spaces between the lines”.12 Although, Bland argues, the literary person might
identify the text as a poem if asked to name the most obvious thing about the
document in front of them, such an answer already presumes reading the words
as words. For the bibliographer, Bland continues, the most obvious things are
in fact the size of the type and the spaces between the lines, and the things that

8
For the seminal work on paratext, see G. GENETTE, Paratexts: Thresholds of Inter-
pretation, trans. J.E. LEWIN (Cambridge, 1997), and for a valuable discussion of the term’s ap-
plicability to early artefacts, see W. SHERMAN, “On the threshold: Architecture, paratext and early
print culture”, in: Agent of Change: Print Culture Studies after Elizabeth L. Eisenstein, ed. S.
ALCORN, E.N. LINDQUIST, and E.F. SHELVING (Amherst, MA, 2007), pp. 67-81.
9
B. MAK, How the Page Matters (Toronto, 2011), p. 5; CARROLL et al., “Pragmatics on
the page”, p. 55.
10
See, e.g., TYRKKÖ et al., “The Culpeper Project”.
11
See, e.g., SUHR, Publishing for the Masses; MEURMAN-SOLIN, “Taxonomisation”.
12
M. BLAND, A Guide to Early Printed Books and Manuscripts (Hong Kong, 2010), p. 3.
Quantifying Contrasts 99

they imply: here, the use of more paper, which made the book more expen-
sive.13 According to Anneli Meurman-Solin,

it is possible to claim that, in regarding texts as specimens of language use, we tend

to marginalise features which we label as ‘non-linguistic’ and therefore consider
them less significant in the process of ensuring the authenticity of data. However,
I propose that we refer to these features by using the concept of ‘indirectly linguis-
tic’ in order to stress that many of them provide information which is not only
useful and important but often also indispensable for producing a correct linguistic
analysis.14

Examples of such indirectly linguistic features include changes in typeface or

the colour of ink which can, among other things, signal the contemporary un-
derstanding of a word’s status as a foreign item or a technical term, or the use
of tables and diagrams, in which relations between words and phrases are indi-
cated by means of spatial positioning and connecting lines and brackets.
Going beyond the individual applications of specific indirectly linguistic
features, one can argue that there is a broader cultural relevance to the paratext-
ual and visual aspects of the page. Features such as layout, the variance of
typographic and visual elements used, and the amount of content packed onto
the page anchor a text to specific print cultural tradition and text type. As will
be discussed in the section ‘Examples: Quantifying change on the early medi-
cal page’, these details are of considerable importance to historical linguists
and philologists who wish to contextualise the books from which they draw
linguistic evidence, particularly as they relate to the sociolinguistic characteris-
tics of the books’ intended readerships.

Computational and Quantitative Analysis of Visual Data

Concurrently with the previously discussed renaissance of paratextual

studies, recent years have also witnessed considerable leaps and bounds in the
use of quantitative data in the humanities. Statistical methods, visualisation
techniques and corpus-driven approaches have enriched the scope of linguistics
by allowing us to query ever larger data sets, up to and including ‘big data’,
and to observe data using techniques such as ‘distant reading’ and ‘culturom-
13
BLAND, A Guide, p. 3.
14
MEURMAN-SOLIN, “Taxonomisation”.
100 JUKKA TYRKKÖ

ics’,15 which sidestep traditional approaches to humanities data completely and

instead endeavour to identify previously undetected patterns. What these meth-
ods, and others like them, have in common is that they encourage a theory-
neutral or data-driven perspective to the primary data and emphasise the need
to look for tendencies, patterns and trends, which may (or may not) reveal
something new, or confirm something already known, about a complex data
set. While it is arguable whether such methods alone can make much of a con-
tribution to our understanding of human languages, texts, or cultures, they
undoubtedly offer intriguing and not infrequently surprising insights into for-
ests that we humans cannot see for all the trees.16
By contrast, the study of paratextual and visual features of texts largely
remains a field where the analysis is typically carried out manually and only
rarely from a quantitative perspective at all.17 Indeed, one of the challenges
associated with traditional philological scholarship is that it has been difficult
to integrate with modern data-driven paradigms, which makes cross-disciplin-
ary approaches difficult to operationalise. As an example, in the larger project
of which the present chapter is a pilot study, the present author is interested in
exploring links between linguistic and visual complexity, that is, in seeing
whether the gradual simplification of the printed pages over the early modern
and late modern periods correlates in a significant way with the standardisation
of English spelling and simplification of sentence structures. The increasing
availability of substantial collections of digitised, computer-readable text col-
lection such as EEBO-TCP and Project Gutenberg have made large-scale dia-
chronic analysis of linguistic changes a reality, but until now the lack of easy-
to-use computational methods of visual analysis has meant that similar analysis
of the visual features has been difficult, if not impossible.
In the emerging field of cultural analytics, associated in particular with the
work of Manovich, the automatic analysis of large cultural data sets is treated

15
For distant reading, see F. MORETTI, Graphs, Maps, Trees: Abstract Models for a Literary
History (London, 2005); for ‘culturomics’, see J.-B. MICHEL et al., “Quantitative analysis of
culture using millions of digitized books”, Science 331 (2011), pp. 176-182. See also the website
of the Visualizing English Print project, which is developing tools and techniques for large-scale
topic modelling of Early Modern English (<http://vep.cs.wisc.edu/projects.html>).
16
It should be emphasised that none of the proponents of data-driven methods actually
claim that such methods alone would suffice.
17
See <http://www.sharpweb.org/main/research/> for a reasonably up-to-date list of on-
going digital research into book history.
Quantifying Contrasts 101

much in the same way as textual data is in corpus linguistics. As Lev Manovich
notes,

... having at our disposal very large cultural data sets which can be analysed auto-
matically with computers and explored using interactive interfaces and visualiza-
tion opens up exciting new possibilities. These possibilities can potentially trans-
form our understanding of cultures in many important ways. Instead of being fuzzy
and blurred, our horizon (knowledge of a cultural field as whole) can become razor
sharp and at the same time acquire a new depth (being able to sort and cluster
millions of artefacts along multiple dimensions). This would enrich our understand-
ing of any single artefact because we would see it in relation to precisely delineated
larger patterns.18

It is worth emphasising that the computational methods discussed in this

chapter are presented primarily as a useful tool which will afford the historical
linguist or book historian one more angle from which to approach paratext-
uality. Although automatically harvested quantitative data can reveal previ-
ously unknown patterns and trends, especially when applied to large data sets,
they cannot, and are not meant to, replace an experienced scholar’s understand-
ing of the artefacts. Furthermore, the successful application of these methods
requires not only an understanding of what the documents are like, but also of
the principles and practices that governed their use in the original context. For
example, we must be cognisant of the fact that during the early modern period
and well into the late modern, authors exerted remarkably little influence over
the paratextual arrangement of their books.19
The methods discussed in this chapter differ in one fundamental way from
the more conventional approach to text-based corpora and digital editions.
Rather than annotating the descriptive metadata into a text corpus in XML or
some other annotation model, both the analytical framework and the resulting
output are essentially independent of the text. While this is less than ideal if the
research question is primarily concerned with examining type size or other
directly text-related factors of paratext, the benefit is that the data is consider-
ably less complicated to handle and query than a typical multi-layered XML

18
MANOVICH, “How to compare one million images?”, p. 252. See also L. MANOVICH,
“Media visualization: Visual techniques for exploring large media collections”, in: The Inter-
national Encyclopaedia of Media Studies, 6, Media Studies Futures, ed. K. GATES (London,
2013), DOI: 10.1002/9781444361506.wbiems 144.
19
SUHR, Publishing for the Masses, pp. 64-66.
102 JUKKA TYRKKÖ

document. For purposes of cross-disciplinary analysis, the quantified visual

data can be merged with other types of continuous or categorical variables with
ease.
Finally, the somewhat contentious question of using facsimile images as
objects of research should be addressed briefly, because all scholars who have
experience of working with facsimile photographs know their limitations and
pitfalls.20 The potential problems are numerous and occasionally serious: mis-
takes in photography, parallax effects, issues with contrast thresholds and ren-
dering which can make some text invisible or emphasise features that are virtu-
ally invisible to the naked eye, and so on.21 However, for all these shortcom-
ings, facsimile images have become an invaluable tool for scholars worldwide,
and some have proposed that the use of digital images and the methods devel-
oped for that purpose constitute a new field of study in itself. As Sarah Werner
notes,

if we could use digital tools to estrange ourselves from our books, to defamiliarise
what we think we know, we might learn something new about how they were made
and how they are used.22

As I see it, this is not to suggest that studying original artefacts should be dis-
couraged, but rather that there are benefits to be reaped from approaching
familiar objects of inquiry from new perspectives as well. While it goes with-
out saying that digital images ought to be derived from reliable sources to
ensure their authenticity and the accuracy of their metadata, and that they must
be of a reasonably high quality to be useful – the utility of visual analysis tools
in assessing this will be discussed later – a much more important point is that
they enable new types of analyses which are practically, if not theoretically,
impossible to carry out using analogue data.

20
I discuss all facsimile images under the same rubric for brevity, though naturally there is
a wide range when it comes to the quality of photography, from high-definition full-colour images
taken at scholarly libraries to the grainy black-and-white images produced by mass digitisation
projects.
21
In the present study, the quality of images was deemed sufficient for the worked example.
Some techniques for improving image quality in the visual analysis software will be discussed in
the section ‘Examples: Quantifying change on the early medical page’.
22
S. WERNER, “Where material book culture meets digital humanities”, blogpost in Wynken
De Worde, posted 29 April 2012 at < http://sarahwerner.net/blog/2012/04/where-material-book-
culture-meets-digital-humanities/>
Quantifying Contrasts 103

ImageJ and Imageplot

The methods of visual analysis discussed in this chapter make use of the
freeware image analysis software ImageJ.23 Described by the developer as “an
open source image-processing program for multidimensional image data with
a focus on scientific imaging”, ImageJ was originally developed for the Na-
tional Institute of Health for medical imaging, but it has subsequently been
adopted by a wide and diverse community of scientists and researchers. As a
Java-based application ImageJ does not require installation as such, and it runs
on MS Windows, Mac OS and Linux. The source code is freely available, and
there are hundreds of open source plugins and macros that expand the basic
capabilities of ImageJ. One of these is ImagePlot, a macro script developed for
ImageJ by Software Studies Initiative for the analysis of cultural image data
such as paintings and posters.24
The basic procedure of using ImageJ and ImagePlot is relatively simple.
First, a set of images is collected and, depending on the quality and nature of
the images, as well as on the research questions, the images are prepared for
analysis. This can mean, among other things, the extraction of specific content,
such as title pages or pages containing engraved images, from a larger data set,
cropping the images to include only relevant parts, the slicing of the image into
tiles, the thresholding of the page, and so on.25 Once all the images are ready
for analysis, they are placed in a folder. ImageJ can then be used to analyse
specific features in each file, either one by one, if image specific adjustments
are required, or in batch mode, in which case the same analysis is carried out
on each file. ImageJ produces an output file in tabular format with each row
representing a file, each column representing a unique feature, and the cells
containing the quantitative values. ImageJ includes numerous different analyti-
cal algorithms, and several can be run on the data set to obtain a wide range of
descriptive values. Finally, the data set can be exported to an external spread-
sheet or statistical package, or analysed by running the ImagePlot macro, which
produces scatter plot visualisations of the selected quantitative values with

23
See W.S. RASBAND “ImageJ”, U.S. National Institutes of Health, Bethesda, Maryland,
USA (1997-2014) <http://imagej.nih.gov/ij/> and SCHNEIDER et al., “NIH Image to ImageJ”.
24
See <http://lab.softwarestudies.com/p/imageplot.html>; MANOVICH, “How to compare
one million images?”.
25
An image may be split into tiles to allow the analysis of its structural components.
Thresholding refers to the setting of cut-off points based on the size, shape and darkness of
elements to be included in the shape count.
104 JUKKA TYRKKÖ

thumbnails of the original images as markers. Both ImageJ and ImagePlot

come with extensive and well-written manuals, but the basic steps of the analy-
sis will be discussed in the following.
What types of analyses relevant to book history and the history of paratext
could be accomplished using ImageJ? The basic functionalities of the software
focus on measurements and various types of particle analyses, which can be
used to count items and analyse their shapes according to various size and
shape parameters. Depending on the thresholds set, a scholar could find all
pages featuring decorative initials or small woodcuts used as text-organising
devices, identify pages containing marginalia, or extract large images such as
frontispieces, to name just a few possibilities. ImageJ can produce detailed
analysis of the brightness, saturation and colours of items on the page, which
would make it possible to identify different inks, distinguish between original
text and later annotations, and so on. A wide variety of image enhancement
tools are also available, which would allow the user to improve the quality of
less-than-ideal photography improving the visibility of red or faded ink. Impor-
tantly, because these processes, and many others, can be carried out on a large
number of images and the results can be saved as quantitative data, these meth-
ods make it possible to accurately analyse trends and baseline tendencies in
large diachronic and synchronic data sets.

Examples: Quantifying Change on the Early Medical Page

In this section, I provide three introductory examples: the first featuring 42

files from the surgical texts categories of the Early Modern English Medical
Texts (2010) and Late Modern English Texts (forthcoming) corpora,26 the sec-
ond illustrating the analysis of a single surgical book, and the third demonstrat-
ing diachronic trends using 115 medical title pages from the early modern
period. The data sets are not large by any means, but this is purposeful to place
emphasis on the basic procedures of the method.

26
Early Modern English Medical Texts, ed. I. TAAVITSAINEN, P. PAHTA, T. HILTUNEN, M.
MÄKINEN, V. MARTTILA, M. RATIA, C. SUHR, and J. TYRKKÖ. Corpus, CD-ROM (Amsterdam,
2010); Late Modern English Medical Texts, ed. I. TAAVITSAINEN, P. PAHTA, T. HILTUNEN, M.
MÄKINEN, V. MARTTILA, M. RATIA, C. SUHR, and J. TYRKKÖ. Corpus (forthcoming). The choice
of corpora is essentially trivial when it comes to this introductory discussion. Suffice it to say that
the two medical corpora used are particularly familiar to the present author as one of their
compilers.
Quantifying Contrasts 105

Fig. 1 Section of the image cropped for analysis.

106 JUKKA TYRKKÖ

Changes on the Surgical Page

The question at hand in all three examples is a seemingly simple one. How
many shapes – that is, characters, images or other continuous forms – appear
on the page, and what percentage of the area of the main text block is covered
in ink? As will be shown, with these two simple metrics alone, we can begin to
examine the stylistic changes that took place on the printed page over three
centuries, and by so doing provide contextual evidence for linguistic and book
historical analysis.
Although the full page, that is, the entire page including the margins, could
be used in the analysis, the decision was made to focus exclusively on the main
block of text, to highlight the fact that computational image analysis technique
can extract useful descriptive information out of even seemingly very mundane
parts of the text. The analysis was carried out using facsimile copies from
Early English Books Online (EEBO) and Eighteenth Century Collections Online
(ECCO).27 All images were converted into black and white, setting the contrast
threshold high enough to render the background, or negative space, white. The
first full page of text from the main body of the book was selected for analy-
sis.28 The page was cropped into the form of a rectangle which includes the
entire main block of body text, exclusive of side and bottom marginalia and
running heads, and catchwords (see Fig. 1). Some of the images were not per-
fectly horizontal in the original photography due to tight binding or camera
position, and in these cases, when the image was cropped to include the main
text block, a small amount of extra white space was unavoidably included in
one or more corners of the image. The effect was considered negligible.
Some of the sample pages included indented items, decorative initials, or
section headers, which introduced some additional negative space. In every
case, this was a common paratextual characteristic of the book in question and
not an anomaly specific to the selected page. Consequently pages with trivial
amounts of paratextual content within the text block were not replaced.

27
The images used for figures 1 and 2 were photographed by the author. Miniature images
are used under the principle of fair academic use.
28
The sampling could be improved by selecting a number of similar pages from each book
and calculating central tendencies for each descriptive feature. In the present case study, only
single pages were selected, but the pages were subjectively evaluated as typical full pages of text.
Quantifying Contrasts 107

Fig. 2 A visualisation of what ImageJ sees on the page.

108 JUKKA TYRKKÖ

The format of the images was converted from tiff to jpg using ImageJ’s
internal batch conversion macro (Process6Batch6Convert).29 The files were
collected in a single folder and a batch image analysis was carried out using the
ImageShapes macro provided with ImagePlot 1.1.30 The ImageShapes macro
uses ImageJ’s internal command “Analyze Particles...”, only running it in batch
mode on default or modified settings.31 All images within the target folder will
be analysed using the same settings, which may or may not be ideal; if neces-
sary, the same analysis could be carried out by analysing each image separately
with customised settings. ImageShapes analyses five features in each image:
the number of shapes,32 the area of the image in pixels, the mean size of the
shapes, the percentage of area covered by the shapes, and the median colour on
the page (which was almost invariably white). Only the count of shapes and the
per cent of area covered will be examined here.
The default settings of the “Analyze Particles...” tool in ImageJ can be
altered by editing the macro script. For example, if the original image has small
dark particles that should be excluded from the analysis, the minimum size of
a shape can be increased. Likewise, a maximum size can be set to exclude large
shapes, such as ornamental decorations and illustrations. Because the minimum
and maximum sizes can be set freely, it would be possible to set the minimum
size above the size of a single character, which would allow the counting of
decorations and images only. Other important settings to consider include
“Exclude on Edges” and “Include holes”. The former, if selected, excludes any
items that touch the edge of the image; while useful at times, if the selected
area is cropped very tightly, numerous letter shapes may be inadvertently ex-
cluded from the analysis. The latter setting controls whether areas of negative
space within shapes, such as counters and eyes in typefaces, are to be counted
as part of the shape or as separate shapes. In the present study, “Exclude on
Edges” was deselected and “Include Holes” was selected. Fig. 2 gives the
“Analyze Particles...” output of the selection shown in Fig. 1.
29
A macro is a pre-prepared sequence of instructions that can be installed and executed
instead of the user having to manually repeat the same set of actions for each file. Macros are used
because they make the performance of routine tasks more efficient and less prone to human error.
30
The amount of memory allocated to ImageJ can be changed from “Edit -> Options ->
Memory and threads”. On reasonably powerful modern computers, set the memory allocation to
at least 2000Mb to prevent crashes.
31
External macros like ImageShapes need to be installed (Plugins -> Macros -> Install)
before they can be executed (Plugins -> Macros -> name of installed macro).
32
A shape is a continuous form. Because the analysis is entirely unsupervised, letter forms
that break into more than one separate shape, such as <i>, are counted as multiple shapes.
Quantifying Contrasts 109

Fig. 3 Scatter plot of Year and Count of shapes

Once the image analysis was done, the quantitative data was exported from
ImageJ to the statistics tool JMP. Fig. 3 gives the count of shapes in the block,
plotted against the timeline. The scatterplot shows a slight but significant de-
cline, which means that there is a declining trend in the number of characters
printed over the two centuries (r=-0.46, R2=0.21, p=**).33 This is a reflection
of the smaller format of the later surgical books: while most of the sixteenth-
century books are folios or quartos, most books in the corpora from the later
seventeenth century and the eighteenth century are octavos.
A strong negative correlation was discovered when we look at the percent-
age of the page covered by the shapes or, to put it another way, the amount of
ink on the page (r=-0.85, R2=0.74, p=***). As Fig. 4 illustrates, the trend is
very strong with no obvious outliers (z>2.5). The format of the book is of no
consequence here, because the coverage is calculated as a proportion of the
selected area. This can be verified from the labels in Fig. 4, which show that
the earliest quartos and octavos show similar levels of coverage to the folios.
This suggests that the extent to which the central block of the page (that is, area

33
In the notation, r indicates Pearson correlation, R2 is coefficient of determination and p
= statistical significance. In short, the linear correlation coefficient indicated the strength and the
direction of the relationship between two variables, and the coefficient of determination shows
how much of the variance of one variable can be predicted from the other.
110 JUKKA TYRKKÖ

Fig. 4 Scatter plot of Year and Coverage (%).

excluding the margins) was covered with ink was a stylistic choice. The practi-
cal means of achieving this effect included the use of black letter types, thicker
roman types, and tighter line spacing, as well as the use of decorated initials,
other ornamental decorations and text dividers, and so on.
Although the discovery of a trend toward a less crowded page is nothing
new to most students of early printing, the example shows how computational
analysis allows us to quantify this trend in precise terms and to visualise the
findings in various ways. As Bland notes, “over time ... slight adjustments are
made to the formal aspects of presentation that cumulatively affect the appear-
ance of the page in quite radical ways”.34 Just like today, printers and booksell-
ers were fully aware of the contemporary conventions, and any deviation from
the normal genre-specific presentation was immediately recognisable and,
more importantly, meaningful to both the craftsmen and the customers. The
printers and booksellers were naturally keen to follow each other’s work to see
whether a new stylistic innovation appeared to go over well with customers. In
the age of the printing press, any new innovation was likely to require invest-
ment in the form of new sets of type and commissioning of image plates, and
few printers had the resources to try new things very often.

34
BLAND, A Guide, p. 4.
Quantifying Contrasts 111

Fig. 5 Cluster dendrogram of sample pages.

The scatter plot not only shows a declining trend, but also a tighter cluster-
ing toward the right-hand side of the plot, which suggests – though, given the
small sample size, certainly does not prove – that late modern printers followed
the contemporary trend more closely than their fellow printers had in the first
two centuries of print. Again quoting Bland, we are reminded that “many early
modern books were printed within what, on reflection, are well-established
112 JUKKA TYRKKÖ

conventions, and these are formats and layouts that must have seemed instinc-
tive to members of the trade”.35
For a different view of the same data, Fig. 5 gives a simple cluster dendro-
gram (Ward’s method, simple linkage) using the two variables of shape count
and coverage to organise the pages into clusters based on similarity. With pub-
lication years as labels, it is easy to see that the dataset divides into three main
clusters (from top to bottom): one comprising primarily of sixteenth-century
books, the second comprising some sixteenth-century books and primarily
seventeenth-century books, and the third consisting entirely of eighteenth-cen-
tury books. As a rough generalisation, we might interpret the dendrogram as
illustrating the aesthetic transition from the heavy, ‘dark’ page of the sixteenth
century to the light and airy page of the eighteenth, with a transitional period
in the middle.
Finally, the same data also allows us to see that there is no significant
correlation between the number of shapes in the text block and the percentage
of paper covered (Fig. 6; r=0.28, R2=0.08, p=na). Given the earlier observation
that both the number of shapes and the amount of ink decline over the timeline,
the only explanation is that there was an aesthetic trend toward lighter, less
blocky typefaces, as well as toward allowing more space between lines. As
Martyn Lyons notes,

by the late eighteenth century ... publishers began to favour streamlined designs
rather than ostentatious decorations of baroque art. New and clearer typefaces were
designed and lettering became more rounded and vertical; notes fell into disfavour,
leaving more space for wide margins.36

As discussed earlier, ImagePlot makes it possible to visualise data as a

scatter plot using thumbnails of the analysed images as markers. A standard
tab-separated data file is prepared with the names of the files in one column
and any desired variables in subsequent columns.37 These need not be restricted
to features measured using ImageJ (or other visualisation software), or indeed
to visual features at all. ImagePlot can take any kind of quantitative data, such

35
BLAND, A Guide, p. 119. For a related study supporting this observation, see TYRKKÖ,
“Printing houses as communities of practice”, where the present author examined the gradual
disappearance of non-standard spelling in early modern medical writing.
36
M. LYONS, Books: A Living History (London, 2013), p. 110.
37
The file references can also be given as full paths, in which case ImagePlot can retrieve
them from multiple folders.
Quantifying Contrasts 113

Fig. 6 Scatter plot of Count and Coverage.

as frequencies of linguistic features, and turn it into a scatter plot with the
corresponding pages (or other images) as markers. In addition to the data file,
the images need to be placed in a folder with each file name corresponding
with an item in the data file. Various features of the scatter plot can be adjusted
as desired. Fig. 7 gives the timeline on the horizontal axis and the coverage on
the vertical axis similarly to Fig. 4. Note how the thumbnails appear to get
lighter as one moves from left to right; this is not a reflection of the ink being
darker in the images to the left, nor of the photographs having higher contrast.
On the other hand, the thicker typefaces used in the earlier books naturally lead
to a lower percentage of the page being left for negative space.

Variation within a Book

The same method can naturally be applied to pages within individual

books, or to large collections of full-length books. This can be particularly
useful when the research question concerns the arrangement of images and
other non-textual paratexts, such as diagrams, within a book, and especially so
when large collections of books can be queried at once. For example,
computationally generated information about the presence of large images
114 JUKKA TYRKKÖ

Fig. 7 ImagePlot visualisation of sample pages by Year and Coverage.

within a book may be one useful indicator of its price, allowing the researcher
to link linguistic practices with audience design.
As an example, we take the full 594 pages of Joannes De Vigo’s Most
Excellent Workes Of Chirurgerye (1543) and again carry out the same analysis
of shape counts and coverage.38 The procedure is exactly the same as before,
with the exception that each page – or, in fact due to the original photography,
a page spread consisting of a verso and recto – was first extracted from the pdf
file. The marginalia were included in the present study.
Fig. 8 gives the ImagePlot visualisation, which suggests that there is a
slight decline in the number of elements on the page as the book progresses,
but it also alerts us to an obvious issue with the primary data; some of the
pages low on the vertical axis appear to be incomplete, as a result of either
mistakes in photography or at some point thereafter in the digital editing pro-
cess.

38
J. DE VIGO, The Most Excellent Workes of Chirurgerye (London, 1543: STC 24720).
Quantifying Contrasts 115

Fig. 8 ImageJ visualisation of Pages in De Vigo’s Most Excellent Workes Of Chirur-

gerye by Year and Count of shapes

The initial run of the analysis revealed a number of pages with compara-
tively low shape counts. Upon closer inspection it became evident that a num-
ber of pages in the middle of the book had been incorrectly photographed with
only the lower half of the page showing. This is a good example of how a po-
tential problem in corpus data can be discovered by including visual analysis
in the workflow. When large collections of texts are digitised automatically
using optical character recognition, incorrectly photographed pages are often
processed along with the rest of the pages without anyone noticing the mistake.
Here, the troublesome images showed shape counts below 3,000, and the deci-
sion was made to use that shape count as a threshold for discarding pages from
further analysis. As Fig. 9 shows, the decline observed earlier appears to be a
real phenomenon (r=-0.31, R2=0.17, p=**): the number of shapes on the page
declines over the course of the book. Note that this is not only the result of a
few image plates toward the end of the book, but that instead the decline ap-
pears to be quite stable.
The ImageShapes macro also allows us to count the mean size of the
shapes in each image. As Fig. 10 shows, the mean size grows gradually
116 JUKKA TYRKKÖ

Fig. 9 Scatter plot of Page and Count of shapes.

(r=0.38, R2=0.15, p=*), suggesting that the back pages of the book may contain
large images or illustrations which register as large shapes.
The relationship between the count of items and the mean size of shapes is
nicely illustrated in Fig. 11; the fewer items there are on a page, the larger they
tend to be. Naturally, the tabular data used for the visualisation can be queried
outside ImagePlot for more data. In Fig. 11, the page numbers have been added
to illustrate how, with only a few exceptions, low shape count and larger mean
size tend to correlate with a higher page number, and vice versa.
There are several points to make about this second example. First, the
initial visualisation (Fig. 7) demonstrated how computational image analysis
can help us identify problems in large data sets by drawing our attention to
outliers and unlikely artefacts, which can then be examined in more detail and
dealt with as appropriate. Secondly, it should be emphasised that the analysis
was carried out using only the visual images. To accomplish the same using a
linguistic corpus, one would first need to key-in or, hopefully, scan and OCR an
early sixteenth-century book of nearly 600 pages, an undertaking which would
take a considerable time under the most favourable conditions, then annotate
Quantifying Contrasts 117

Fig. 10 Scatter plot of Page and Mean size of shapes.

all the non-textual elements into the corpus, being careful to count all the items
and measure their sizes. To put this task into perspective, the combined shape
count in De Vigo’s Most Excellent Workes Of Chirurgerye is roughly 1.23
million shapes. And thirdly, although the placement of images can naturally be
easily verified in individual books, doing so with data sets sufficiently large to
allow generalisations about diachronic trends in print culture would be
extremely time consuming and virtually impossible to turn into quantifiable
format.
The methods outlined here are naturally only a starting point. For example,
it is relatively simple to analyse the structure of a page by automatically slicing
the image into a predetermined number of tiles and then performing the previ-
ously described procedure on each tile, giving each section of the page a shape
count and coverage figure. The quantitative data can then be tabulated both
within the book and across a longer timeline to determine whether the overall
balance of the page layout has changed over time. Similarly, if the number of
items within the body text is known, an analysis of the full page can be carried
out and the two sets of figures can be compared, which gives the number and
118 JUKKA TYRKKÖ

Fig. 11 Scatter plot of Count of shapes and Mean size of shapes (page numbers as
labels).

size of items in the margins. If the two aforementioned methods are combined,
the contents of each of the four margins can be studied separately.

Early Modern Title pages

In the final example, 115 title pages of books included in the Early Modern
English Medical Texts corpus are used to illustrate the trend away from illus-
trated title pages over the 200-year timeline. Books without title pages were
naturally left out of the analysis, as were books where the facsimile image of
the title page was incomplete or corrupted. The title pages were used in their
entirety, including any marginal space. The images were processed following
the procedure discussed in the subsection ‘Changes on the surgical page’.
The diachronic trend is shown in Figs. 12 and 13 (r=-0.41, R2=0.169, p=*).
It is immediately apparent that title pages with large illustrations, registering as
coverage of over 30 percent of the page, appeared predominantly in the first
Quantifying Contrasts 119

Fig. 12 Scatter plot of Coverage by Year.

half of the sixteenth century and only very rarely thereafter.39 The most con-
spicuous outlier in the seventeenth century is John Gerarde’s The Herball, or
Generall historie of plantes (1633).40 First published in 1597, the book is noted
for its hundreds of illustrations, which originally appeared in a number of con-
tinental botanical books. The 1633 edition was heavily edited by the apothe-
cary Thomas Johnson.
One of the strengths of visualisations such as Fig. 13 is that much informa-
tion is made available at a glance. Although the thumbnail images could be
thought of merely as illustrations, they also provide us with additional evi-
dence. For example, it is immediately apparent that many of the more textual
title pages in the early sixteenth century would appear to show the inverted
pyramid arrangement typical of early hand-press books.41 By contrast, none of
the seventeenth century books continue this trend, both because it had become

39
C. SUHR and J. TYRKKÖ, “Investigating the cultures and practices of print: Corpus-based
taxonomies of medical title-pages 1500-1800”, a conference presentation at ICAME37, Hong
Kong, 26 May 2016, report that throughout the sixteenth century, nine out of ten medical title
pages included a decorative image of some kind. That number fell to 55% in the first half of the
seventeenth century and to less than 20% by the latter half of the century.
40
J. GERARDE, The Herball, or Generall historie of plantes (London, 1633: STC 11751).
41
See, e.g., G. PROOT, “Converging design paradigms: Long-term evolutions in the layout
of title pages of Latin and vernacular editions published in the Southern Netherlands, 1541-
1660”, The Papers of the Bibliographical Society of America 108 (2014), 269-305.
120 JUKKA TYRKKÖ

Fig. 13 ImagePlot visualisation of Coverage by Year.

old-fashioned and because it was considered a laborious and unnecessary exer-

cise for the typesetter.
For the historical linguist, these images can provide invaluable – and read-
ily available – evidence of the respective books’ intended audiences. In the
beginning of the sixteenth century, nearly all printed medical books were pro-
duced in small numbers and at considerable expense for a wealthy readership
of professional practitioners. Then, as we move along the timeline, medical
books are printed in greater volume for a widening audience of literate medi-
cos, such as surgeons, apothecaries and midwives, and this is reflected in the
paratextual and visual aspects of the books. There is a gradual transition to-
ward an increasingly sparse style of presentation, which comes to be associated
with use-literature and the emerging sense of academic publishing, spear-
headed by the Philosophical Transactions of the Royal Society.42 Indeed, the

42
The Royal Society of London was founded in 1660, and the Philosophical Transactions,
often considered the first academic journal in the modern sense, was first published in 1665.
Quantifying Contrasts 121

only books from this latter end of the timeline with long, wordy title pages are
books specifically targeted for less affluent readers, such as Aristoteles Master-
piece, or the Secrets of Generation displayed in all the parts thereof (1684), a
midwifery manual by an anonymous author, and John Durant’s Art and Nature
Joyn Hand in Hand, or the Poor Mans Daily Companion (1697), a medical
textbook for non-professionals.43

Conclusions and a Look to the Future

The foregoing discussion has hopefully demonstrated some of the potential

benefits of computational image analysis. As noted in the beginning, rather
than presenting a detailed analysis of a specific research question, the main
objective of this paper was to introduce these methods and to offer some ideas
about their possible uses in linguistic, philological and book historical re-
search. The methods discussed in this chapter do not require much computa-
tional know-how, but the tools introduced, as well as others like them, will also
accommodate the needs of a more experienced scholar who may be interested
in developing complex macro scripts to suit the needs of a particular research
project. Also, because the visual data can be exported into other applications,
it can be linked or merged with other types of data as needed. While relatively
small data sets were used in the present study in an effort to focus attention on
the methods, the same methodology can be applied to very large data sets and
combined with linguistic evidence, where available.
As regards the present author’s own research interests, the main question
relates to the challenge of quantifying the ideational or cultural connection
between linguistic, paratextual, and visual features over the early modern and
late modern periods. In order to analyse those connections in a data-driven,
statistically well-argued manner, it will be necessary to analyse very large
collections of visual data from the same sources used as linguistic primary
data, and this can only be managed using computational methods. To conclude,
the importance of the visual and paratextual dimension to linguistic analysis is
only beginning to be recognised, and with the rapid development of digital

43
Anon., Aristoteles Master-piece, or the Secrets of Generation displayed in all the parts
thereof (London, 1684), WING A3689; J. DURANT, Art and Nature Joyn Hand in Hand, or the
Poor Mans Daily Companion (London, 1697), WING D2681A.
122 JUKKA TYRKKÖ

humanities methodologies and the growing archives of computer-readable

primary data, the opportunities for new discoveries are truly exciting.
 Part III

Communicating through Script and Typography

 Stating the Obvious in Runes

YIN LIU

he first obvious statement I will make is that this chapter starts from a

T very different place than the others in this volume. The texts I will dis-
cuss do not appear on the parchment or paper page, but have instead
been carved onto bones, boxes, rings, and monuments. They are a handful of
English runic inscriptions from the fourth to ninth centuries, and what they
have in common is that they are strikingly self-referential. Not only do they
defy modern expectations that texts normally appear on ‘pages’, these inscrip-
tions also challenge our assumption that writing is intended to convey verbal
information from writer to reader, and raise the question of what it means to
read a written text.

Anglo-Saxon Runic Literacy: Context and Questions

The runic alphabet – the futhorc in Old English – was originally designed
for carving and was used by the early Germanic peoples throughout Europe,
primarily in the northwest. I will confine my discussion to a very small subset
of the English runic epigraphic corpus, which is itself quite small. The practice

......................................................................................................................................
Verbal and Visual Communication in Early English Texts, ed. Matti PEIKOLA, Aleksi
MÄKILÄHDE, Hanna SALMI, Mari-Liisa VARILA, and Janne SKAFFARI, Utrecht Studies in Medieval
Literacy, 37 (Turnhout: Brepols, 2017), pp. 125-139.

DOI 10.1484/M.USML-EB.5.114133
126 YIN LIU

of runic writing in early medieval England followed a trajectory distinct from

the development of runes in Scandinavia or continental Europe, and it must be
carefully noted that what is true of English runes may not be true of runes
elsewhere, and vice versa. What we do know of medieval English runes is that
they occurred in both epigraphic and manuscript contexts; they were used to
write both English and Latin; they appeared in explicitly Christian contexts
from at least the seventh century onward; and they occurred sometimes in
mixed-script inscriptions, that is, objects carved with both runic and roman
letters. The very limited nature of the extant English corpus – fewer than a
hundred inscriptions recognisably runic in whole or in part – means that there
is a great deal that we cannot know with certainty about the use of runes in
early medieval England. René Derolez has suggested that the texts that survive
amount to “less than one percent” of a hypothetical total of all runic texts in the
five centuries after Anglo-Saxon settlement in England, and consequently a
single newly discovered inscription would have the potential to overturn any
number of our current theories.1
Thus many questions about English runic literacy persist. How widespread
was runic literacy in early medieval England? Was it a popular or an elite
script? What was its relationship with the roman alphabet and with Latin liter-
acy? For example, does the existence of an inscribed object such as the Fal-
stone memorial, in which the same English text is written in roman letters on
the left side and runes on the right, suggest that some readers could read roman
but not runic, others runic but not roman – or did such ‘multiliteralism’, to use
Thomas Bredehoft’s term, have other functions that had little to do with reach-
ing two types of readers?2 Above all, why would someone in Anglo-Saxon
England choose to write in runes rather than in roman script?

1
R. DEROLEZ, “Runic literacy among the Anglo-Saxons”, in: Britain 400-600: Language
and History, ed. A. BAMMESBERGER and A. WOLLMANN (Heidelberg, 1990), pp. 397-436, at pp.
400-401; R.I. PAGE, “Quondam et futurus”, in: ID., Runes and Runic Inscriptions: Collected
Essays on Anglo-Saxon and Viking Runes, ed. D. PARSONS (Woodbridge, 1995), pp. 1-16, at pp.
13-16.
2
C.E. FELL, “Anglo-Saxon England: A three-script community?”, in: Proceedings of the
Third International Symposium on Runes and Runic Inscriptions: Grindaheim, Norway, 8-12
August 1990, ed. J.E. KNIRK (Uppsala, 1994), pp.119-137; R.I. PAGE, An Introduction to English
Runes, 2nd edn. (Woodbridge, 1999); T.A. BREDEHOFT, “Multiliteralism in Anglo-Saxon verse
inscriptions”, in: Conceptualizing Multilingualism in England, c. 800 - c. 1250, ed. E.M. TYLER
(Turnhout, 2011: Studies in the Early Middle Ages 27), pp. 15-32.
Stating the Obvious in Runes 127

Stating the Obvious

A somewhat unlikely group of texts suggests particular answers to these

questions: English runic inscriptions whose messages are so obvious, it seems,
as to be mind-numbingly banal. A typical text of this sort is found on an eighth-
century silver-gilt finger-ring found at Wheatley Hill, County Durham, in
1993.3 The runic inscription on the hoop, although partly obscured by the set-
tings for the stones that were attached later, is still readable: “hring ic hattæ”
(“I am called a ring”).4
This is rather like opening a book to read the sentence “You are reading a
book” or digging up at an archaeological site a stone painstakingly engraved
with the statement “This is a stone”. It may seem at first to be an irritatingly
juvenile Anglo-Saxon practical joke, except that the Wheatley Hill inscription
fits a pattern in a handful of other runic and non-runic early English texts in
which dead or inanimate objects identify themselves.5 Indeed, the very earliest
English (or proto-English) text extant is a runic inscription on an astragalus, a
mammal’s ankle-bone, discovered in an early fifth-century cremation urn at
Caistor-by-Norwich. Also in the urn were some personal grooming tools, black
and white gaming pieces, and many other astragali, from sheep, that may have
been used as gaming-pieces also. The inscribed astragalus was the largest of
the set and the runes on it were the subject of some debate for a time, but it is
now generally agreed that they say “raihan”, that is, “from a roe deer”. And
indeed this is the only astragalus in the set that comes from a deer rather than
a sheep.6 Two later Anglo-Saxon runic objects similarly self-identify as animal
bones. The Brandon antler is a deer antler refashioned as a tool handle, and is
inscribed with the words “wohs wildum deoræ an” (“grew on a wild animal”).7

3
British Museum number 1995,0902.1; PAGE, Introduction, p. 169; E. OKASHA, “Anglo-
Saxon inscribed rings”, Leeds Studies in English 34 (2003), pp. 29-45, at p. 31; T. LOOIJENGA,
Texts and Contexts of the Oldest Runic Inscriptions (Leiden, 2003), p. 292.
4
Since this volume is not directed primarily at runologists, for ease of reading I do not
follow the usual conventions for runic transcription here but present the texts in edited form, with
word division and slightly normalised spelling. All translations are my own.
5
LOOIJENGA, Texts and Contexts, p. 109, with other examples from the European corpus.
6
J. HINES, “The runic inscriptions of Early Anglo-Saxon England”, in: Britain 400-600:
Language and History, pp. 437-455, at pp. 441-442; C. HILLS, “The archaeological context of
runic finds”, in: Old English Runes and Their Continental Background, ed. A. BAMMESBERGER
(Heidelberg, 1991), pp. 41-59, at pp. 53-54; PAGE, Introduction, pp. 18-19, 179-180; LOOIJENGA,
Texts and Contexts, pp. 284-285.
7
PAGE, Introduction, pp. 169-170; A. BAMMESBERGER, “The Brandon Antler runic in-
128 YIN LIU

The Southampton bone is inscribed “catæ”, the meaning of which is somewhat

doubtful but which is most often considered to be a Frisian word meaning
“knuckle-bone”.8 It was discovered in a former rubbish pit associated with the
early medieval coastal settlement of Hamwic, and the function of the inscrip-
tion is obscure. However, its existence raises the question of how many other
apparently casual inscriptions of this sort, on ephemeral objects such as bits of
bone or wood, may be now lost.
These very short texts, which have been described neutrally as ‘auto-
referential’ and rather pejoratively as ‘simplistic’ and ‘self-evident’, have ana-
logues elsewhere in northern Europe: a stringed instrument labelled as a lyre,
a small box labelled as a jewel-case, combs labelled as combs, and the insis-
tently redundant inscription on a bone fragment from Lund, Sweden – “bin is
þita bin is þit[a]” (“this is bone, this is bone”).9 These self-referential objects
have elicited bafflement from modern scholars; of inscriptions such as that on
the Caistor-by-Norwich astragalus, John Hines has remarked, “one cannot
imagine that they had much utilitarian value in interpersonal transactions”.10 Of
Anglo-Saxon inscribed rings, Elisabeth Okasha notes, “in a society with a
limited level of literacy, such an undertaking does not appear to have enormous
practical advantage”.11 We might be tempted to dismiss these texts as casual
graffiti, scratched by idlers with knives onto whatever objects happened to be
at hand, except that even the very early Caistor-by-Norwich astragalus was
deliberately placed in a cremation urn, with at least two different sets of gam-
ing pieces and other luxury items, and the Wheatley Hill inscription seems to
have been incorporated into the original design of the ring – although later
gemstone settings partly obscure it, suggesting that its legibility was no longer
important.

scription”, Neophilologus 86 (2002), pp. 129-131. The transliteration here is Bammesberger’s,

with the concession that the last word could be on; Page suggests wohs wildum deoran. Either
way the meaning is the same.
8
PAGE, Introduction, pp. 168-169; LOOIJENGA, Texts and Contexts, p. 324. The alterna-
tive, Old English cat or catte, meaning ‘cat’, makes much less sense.
9
The term ‘autoreferential’ is from J. HINES, “Some observations on the runic inscriptions
of Early Anglo-Saxon England”, in: Old English Runes and Their Continental Background, pp.
61-83, at p. 74; ‘simplistic’ and ‘self-evident’ are from PAGE, Introduction, p. 169, where these
examples are mentioned. For other examples see also LOOIJENGA, Texts and Contexts, p. 109.
When in Turku, Finland, at the conference at which this paper was presented, I was delighted to
see a fragment of a fourteenth- or fifteenth-century wooden vessel unearthed at the Aboa Vetus
site and inscribed, in runes, bua el, probably intending bulle, ‘bowl’.
10
HINES, “Some observations”, pp. 74-75.
11
OKASHA, “Anglo-Saxon inscribed rings”, p. 40.
Stating the Obvious in Runes 129

There is certainly nothing casual, either, about the well-known inscription

on the front panel of the Franks (Auzon) Casket, a carved whalebone box from
eighth-century Northumbria. Around images alluding to the Germanic legend
of Weland and the visit of the Magi to the Christ child is carved, in runes, not
a description of the pictures but a two-line poem about a monstrous and terrify-
ing fish experiencing sorrow when lifted by waves onto a rocky slope. The
poem is, in fact, a riddle, and the word on the left-hand side of the front panel
gives the answer: “hronæsban” (“whalebone”): the box was made from the
bone of a beached whale.12 Unlike the other texts on the box, the front panel
inscription does not identify the subject of the images but the material of which
the box itself was made. It is as if a picture were provided with a caption that
said: “This picture was printed on acid-free paper”. Like objects inscribed with
declarations such as “from a roe deer” or “I grew on a wild animal”, the Franks
Casket confounds our expectations by stating the obvious: it is bone declaring
itself to be bone. Whether or not an Anglo-Saxon handling the box would have
known from its look and feel that the substance was whalebone (in fact, proba-
bly sperm whale mandible, but even modern archaeologists have difficulty
determining exactly what sort of whalebone a small object was made from), the
inscribed riddle, unusual among Anglo-Saxon riddle texts in overtly providing
the solution, would surely have produced a pleasant jolt of recognition as the
casket revealed its history.13
At the other end of the scale from single words scratched onto bits of bone
are inscriptions on stone monuments such as crosses and grave markers. For
example, the seventh- or eighth-century Crowle stone (Lincolnshire) displays
the word licbæcun, otherwise unattested but most likely a compound meaning,
literally, “corpse-monument”, that is, “memorial stone”. This is not so glaring
an example of a self-evident statement when we realise that the stone is a frag-
ment of what was probably a cross, and that it probably featured a longer in-
scription declaring that it was a monument erected by someone or in memory
of someone. Thus, for example, the ninth-century Great Urswick stone (Lanca-
shire) reads “Tunwini setæ æfter Torohtredæ bekun æfter his bæurnæ; gebidæs
þer saulæ” (“Tunwini set up for Torhtredæ a memorial for his son; pray for his
soul”), and then “Lyl þis w ...” before the inscription breaks off, but presum-
12
British Museum number 1867,0120.1; PAGE, Introduction, pp. 172-179; C.E. KARKOV,
The Art of Anglo-Saxon England (Woodbridge, 2011: Boydell Studies in Medieval Art and
Architecture), pp. 145-153.
13
V.E. SZABO, Monstrous Fishes and the Mead-Dark Sea: Whaling in the Medieval North
Atlantic (Leiden, 2008), pp. 55-56.
130 YIN LIU

ably it was “Lyl þis wrohte” (“Lyl made this”). The fragmentary Falstone in-
scription (Lancashire), c. 800, repeats the same text twice, once in runic and
once in roman: “... æfter Hroethberhtæ becun æfter eomæ; gebiddæd þer sau-
le” (“[someone set up] for Hrothberht a memorial for his uncle; pray for his
soul”). The inscription on the ninth-century Overchurch stone (Cheshire) can
be reconstructed to read “folc arærdon becun; gebiddaþ fore Æþelmunde”
(“the people raised up a memorial; pray for Æthelmund”). Also from the ninth
century, the text of Thornhill III (West Yorkshire), carved on a stone slab,
features two female names: “Gilsuiþ arærde æfte[r] Berhtsuiþe bekun on ber-
gi; gebiddaþ þær saule” (“Gilswith raised up for Berhtswith a memorial on a
hill; pray for her soul”). It is reasonable to suppose that the word becun in all
these inscriptions refers to the stone memorial itself.14
These very conventional formulas on monuments provide a context for one
of the most studied of runic texts, the inscription on the Ruthwell Cross. What
has drawn the most attention on this early eighth-century stone cross are the
runes around the borders of the lower east and west sides, which represent
quotations from the Old English poem now known as The Dream of the Rood,
known also from a full but later version in the Vercelli Book (MS Vercelli,
Biblioteca Capitolare CXVII) of the late tenth century. The cross itself speaks:

+ ondgeredæ hinæ God alme ttig þa he walde on galgu gistiga,

modig fore allæ men; buga ic ni dorstæ ...

ahof ic riicnæ kyninc, heafunæs hlafard; hælda ic ni dorstæ;

bismæradu unket men ba ætgadre; ic wæs miþ blodi bistemid ...

+ Krist wæs on rodi; hweþræ þer fusæ fearran kwomu

æþþilæ til anum; ic þæt al biheald.

saræ ic wæs miþ sorgum gidroefid; hnag ...

miþ strelum giwundad; alegdun hiæ hinæ limwoerignæ;

gistoddun him æt his licæs heafdum; bihealdun hiæ þer,

heafunæs dryctin ...

14
A. BAMMESBERGER, “Three Old English runic inscriptions”, in: Old English Runes and
Their Continental Background, pp. 125-136, at pp. 128-131 (Overchurch); FELL, “Anglo-Saxon
England”, pp. 128-129 (Falstone, Crowle); for all, see PAGE, Introduction, pp. 55-56, 59, 141-
142, 149-151.
Stating the Obvious in Runes 131

Almighty God stripped himself when he wished to ascend the gallows,

courageous before men; I dared not bow down ...

I raised the mighty king, the lord of heaven; I dared not bend down;
we two were degraded together; I was soaked with blood;

Christ was on the cross; yet there came from afar

nobles to that one man; I beheld it all.

I was hard beset with sorrows, bent down ...

wounded with arrows; they laid him down weary of body,

stood at the head of his corpse, looked on him there,

lord of heaven ...15

The Ruthwell Cross is atypical of such inscribed stone monuments in that

it does not seem to be a memorial stone for a particular individual from a local
community, but instead is a surface for displaying images and texts alluding to
figures and scenes from the life of Christ. It features a number of inscriptions
in both roman and runic, and with one notable exception they all serve to iden-
tify the images in the carved panels; for example, around a panel showing
Christ triumphant, with two animals submissive under his feet, is the Latin
inscription in roman capitals: “IHS XPS IVDEX AEQUITATIS BESTIAE ET DRACO-
NES COGNOVERVNT IN DESERTO SALVATOREM MVNDI”, that is, “Jesus Christ,
judge of equity; beasts and dragons recognised in the desert the saviour of the
world”. The exception is, of course, the English text, which is inscribed in
runes on the borders of generic vine-scroll carvings. This inscription interprets
not the significance of the images on the cross, but the significance of the cross
itself. Thus the runes declare the cross to be a cross, a representation of the
wooden cross on which Christ died, but they do so in an indirect way: the cross
itself speaks and tells its story in such a way that a Christian audience would
recognise its identity. The Vercelli Dream of the Rood, albeit in passages that
do not correspond to the text on the Ruthwell Cross, repeatedly refers to the
cross as a beacen, a word whose primary meaning in Old English was ‘sign’,
and which had the specific meaning ‘monument’, as we have seen.

15
Partially reconstructed text adapted from É. Ó CARRAGÁIN, Ritual and the Rood:
Liturgical Images and the Old English Poems of the Dream of the Rood Tradition (London,
2005), pp. XXII-XXIII, XXVI-XXVII; compare D.R. HOWLETT, “A corrected form of the
reconstructed Ruthwell Crucifixion Poem”, Studia Neophilologica 80 (2008), pp. 255-257.
132 YIN LIU

The runic inscriptions on both the Franks Casket and the Ruthwell Cross
are riddling texts: that is, they tell a story in oblique ways so that the reader is
led to a moment of recognition when the answer to the riddle is revealed as the
very object on which the riddle is written. They have, therefore, a certain affin-
ity with another set of well-known Anglo-Saxon texts featuring speaking inani-
mate objects, the Exeter Book riddles (in MS Exeter, Cathedral, Dean and
Chapter Library 3501). Catherine Karkov has commented on the close relation-
ship between speech and writing in Anglo-Saxon England, a society in which
communicative acts were overwhelmingly more likely to be oral than written;
in her words, “writing is a representation of voice”.16 Thus, in the Exeter Book
riddles, birds and weapons and household implements and books speak out,
challenging the reader or listener to identify them: “saga hwæt ic hatte” (“say
what I am called”). The dreamer in The Dream of the Rood begins by describ-
ing what he sees – “ic gesawe syllicre treow” (“I saw a most wondrous tree”)
– just as Exeter Book Riddle 31 declares “ic seah sellic þing” (“I saw a won-
drous thing”).17 It is worth noting that runes appear sporadically in the Exeter
Book riddles, as hidden clues; for example, the solution to Riddle 24 is pro-
vided in a series of runes that can be unscrambled to provide the solution:
“higoræ” (“magpie”). The runic inscription on the Chessell Down pail (Isle of
Wight) may be similarly decoded as an anagram: “bws.ecccæææ” could be
unscrambled to read “Becca, Wecca, Secca”, a series of masculine personal
names of which Becca and Secca are attested from Widsith line 115.18 Not all
Anglo-Saxon riddling inscriptions are runic; in roman (with a few mysterious
intrusive runes) is the inscription on the Orpington sundial: “... ðan ðe secan
can hu + / ... tellan and healdan +” (“for the one who can seek out how / to
count and to keep”), and the solution in Latin (now mostly effaced): “(h)orolo-
gium” (“timepiece”).19

16
C.E. KARKOV, “Art and writing: Voice, image, object”, in: The Cambridge History of
Early Medieval English Literature, ed. C.A. LEES (Cambridge, 2012: New Cambridge History
of English Literature), pp. 73-98, at p. 82.
17
The Vercelli Book, ed. G.P. KRAPP (New York, 1932: Anglo-Saxon Poetic Records 2),
p. 61; The Exeter Book, ed. G.P. KRAPP and E. VAN KIRK DOBBIE, (New York, 1936: Anglo-
Saxon Poetic Records 3), p. 196. More recent editions are The Dream of the Rood, ed. M.
SWANTON, 2nd edn. (Exeter, 1987); The Old English Riddles of the Exeter Book, ed. C. WILLIAM-
SON (Chapel Hill, 1977).
18
LOOIJENGA, Texts and Contexts, pp. 280-281.
19
PAGE, Introduction, pp. 223-224; E. OKASHA, “Script-mixing in Anglo-Saxon inscrip-
tions”, in: Writing and Texts in Anglo-Saxon England, ed. A.R. RUMBLE (Cambridge, 2006:
Publications of the Manchester Centre for Anglo-Saxon Studies 5), pp. 62-70, at p. 64.
Stating the Obvious in Runes 133

A riddle is a form of semantic cryptography; it must be decoded to be un-

derstood. Rather than communicating meaning in the clearest way possible, it
deliberately makes meaning difficult to discover.20 There are other indications
that the Franks Casket is a cryptographic object; its right panel features a runic
inscription in which the usual vowel runes are replaced by other symbols in a
basic substitution code.21 It can also be argued that the runic inscription on the
Ruthwell Cross is deliberately cryptic. Page complains that the runic text
“looks absurd and is maddeningly hard to read”, and Calvin Kendall suggests
that the use of runes and their counterintuitive layout “seem designed to inhibit
easy understanding”.22 Unlike the Latin inscriptions, which follow the orienta-
tion of the panel borders and thus present a more or less continuous linear text,
the runic Dream of the Rood inscriptions on the side faces of the cross shaft are
read horizontally down the vertical borders, breaking up the text into groups of
two or three characters, so that, for example, “hælda ic ni dorstæ” (“I dared not
bend down”) appears like this:

Fig. 1 Rune-groups on the Ruthwell Cross.

The effect is, if not cryptographic, at least obfuscatory. I suspect that medieval
Anglo-Saxons would not have found these inscriptions particularly easy to read
either. Indeed, it is highly likely that a literate person in early medieval Eng-
land, where Latin literacy was normally a precondition for vernacular literacy,
would have found it much easier to read the Latin inscriptions in roman script
on the Ruthwell Cross than to read the English inscriptions in runes.

20
The scholarship on riddles is vast, but for the Anglo-Saxon context see especially N.
PORTER STORK, Through a Gloss Darkly: Aldhelm’s Riddles in the British Library MS Royal
12.C.XXIII (Toronto, 1990); J.D. NILES, Old English Enigmatic Poems and the Play of the Texts
(Turnhout, 2006); D. BITTERLI, Say What I Am Called: The Old English Riddles of the Exeter
Book and the Anglo-Saxon Riddle Tradition (Toronto, 2009).
21
PAGE, Introduction, pp. 86-88, 178-179.
22
PAGE, Introduction, p. 147; C.B. KENDALL, “From sign to vision: The Ruthwell Cross and
‘The Dream of the Rood’”, in: The Place of the Cross in Anglo-Saxon England, ed. C.E. KAR-
KOV, S. LARRATT KEEFER, and K.L. JOLLY (Woodbridge, 2006), pp. 129-144, at p.140, note 54.
134 YIN LIU

A debate continues over whether the runic inscription on the Ruthwell

Cross was part of the original design of the cross or added later, and to this
debate our observations about the obscurity of the runic text are relevant. One
argument for the view that the runes were added later is that everything else on
the cross shows a high degree of craftsmanship, so the peculiar layout of the
runic text can be ascribed to a less skilled artisan, who was trying to squeeze
the poem into the borders of the east and west panels and sacrificed readability
to do so. Page, advancing this theory, adds, apologetically, “this is a heretical
view and not shared by art historians”; however, Okasha calls it conclusive,
and Bredehoft assumes it. On the other hand, if readability was not a concern
of the rune-carver, indeed if the runic text was intended to be difficult to read,
then there is every reason to believe that it was an integral part of the monu-
ment’s original design, as Éamonn Ó Carragáin has argued.23 It is perhaps
significant that runologists, who try to read the semantic content of a runic text,
tend to think that the Ruthwell Cross inscription was a later addition, whereas
art historians, who consider an inscription in the context of an artefact’s deco-
rative design, tend to think that it was original.

Implications

A strong body of evidence suggests that, in early Anglo-Saxon England

(before the Alfredian reforms of the late ninth century), knowledge of runic
writing was not the most basic form of literacy but rather a more specialised
skill. Contrary, perhaps, to popular modern assumptions, it was Latin literacy
and knowledge of the roman script that was more basic. Some mixed-script
inscriptions suggest that their carvers were more familiar with roman than with
runic. On the oak coffin made c. 698 at Lindisfarne to house the remains of St.
Cuthbert, the runic inscription ihs xps for the name of Christ is clearly a trans-
literation of the nomina sacra in a roman script.24 The strangely mixed charac-
ter of the bilingual inscription on the back panel of the Franks Casket, which
begins as Old English in runic script, then switches to Latin in roman script,

23
PAGE, Introduction, p. 147; OKASHA, “Script-mixing”, p. 66; BREDEHOFT, “Multiliteral-
ism”, p. 24; for the opposing argument see Ó CARRAGÁIN, Ritual, pp. 47-53.
24
R.I. PAGE, “Roman and Runic on St Cuthbert’s coffin”, in: St Cuthbert, His Cult and His
Community to AD 1200, ed. G. BONNER, D. ROLLASON, and C. STANCLIFFE (Woodbridge, 1989),
pp. 257-265, at p. 264; LOOIJENGA, Texts and Contexts, p. 286; OKASHA, “Script-mixing”, p. 66;
M.P. BARNES, Runes: A Handbook (Woodbridge, 2012), p. 46.
Stating the Obvious in Runes 135

and then ends with one Latin word in runes, has been convincingly explained
by Page: the carver was following a Latin text in roman script, both translating
into Old English and transliterating into runes as he went; halfway through he
slipped into copying the exemplar directly, and when he recovered in the last
word he did so only partially, transliterating into runes but remaining in
Latin.25 Anomalies in the orthography of the Falstone inscription suggest that
the carver, while skilled in both scripts, “was more at home in roman than ru-
nic”; for example, the last letter of gebiddaed (gebiddað) is d in both roman
and runic, even though runic thorn (þ) would have been more appropriate.26 In
the small set of mixed-script inscriptions that survive, it is much more common
for a predominantly roman text to include some runes than for a predominantly
runic text to use a few Roman letters. In a number of runic inscriptions on
memorial stones, rings, and fittings, serifs on the runes show that the artisans
were influenced by the (roman) insular bookhands used in manuscripts.27 As
Okasha notes, “it seems inconceivable that an Anglo-Saxon composer of a text
should be basically literate in runes and know how to write [R]oman script
only as a secondary skill. This goes against everything we know of literacy in
Anglo-Saxon England”.28 David Parsons has argued that runic literacy in Eng-
land was explicitly facilitated by Christian monastic interests, that is, by pre-
cisely those people in Anglo-Saxon society who would have had the highest
level of skill in Latin literacy and the use of roman script.29
It has been assumed that “runic began as a practical script, invented for
day-to-day purposes” and that “there were numbers of Anglo-Saxons who were
runically literate”, implying that the choice to write in runes would be moti-
vated by the desire to reach more people than could read roman script, and
perhaps people who were less well educated.30 Likewise, mixed-script inscrip-
tions seem to suggest two literacies in Anglo-Saxon England – one runic in
English, the other roman in Latin – that were distinct but overlapping, so that,

25
PAGE, Introduction, p. 176; see also OKASHA, “Script-mixing”, p. 65; BREDEHOFT, “Mul-
tiliteralism”, pp. 20-21.
26
FELL, “Anglo-Saxon England”, p. 128, citing PAGE, Introduction, p. 142, and A. CAMP-
BELL, Old English Grammar (Oxford, 1959), p. 29.
27
OKASHA, “Script-mixing”, pp. 68-70; PAGE, Introduction, pp. 103-104.
28
OKASHA, “Script-mixing”, p. 66. See also J. HINES, “New light on literacy in eighth-
century East Anglia: A runic inscription from Baconsthorpe, Norfolk”, Anglia 129 (2011), pp.
281-296, at p. 295.
29
D.N. PARSONS, Recasting the Runes: The Reform of the Anglo-Saxon “Futhorc” (Upp-
sala, 1999).
30
PAGE, Introduction, pp. 103, 102.
136 YIN LIU

again, the choice to write in runes would be an extension to a wider reader-

ship.31 But the evidence points in the other direction. In a society where literacy
was already a highly specialised technical skill, runic literacy would have been,
at least in Christianised Anglo-Saxon England, even more specialised. Further-
more, as Hines has remarked, runic writing in Anglo-Saxon England, espe-
cially earlier in the period, was strongly associated not with mundane uses but
with prestige objects, such as high-status grave goods, jewellery, or monu-
ments; this accords with Tineke Looijenga’s argument that, in northern Europe
generally, runes began as an elite script and demonstrated the skills of an arti-
san class.32
One possible answer, then, to the question of why an Anglo-Saxon artisan
would have chosen runic writing, is that runes were commonly used as a dis-
play script for social elites – that is, a script used primarily as decoration and
to raise the status of an object – and that the value of the object may have been
enhanced by the addition of a runic inscription, whether or not most people
could read it.33 Further evidence for this function of runes is the existence of
many undecipherable and baffling objects that appear to be carved or scratched
with what might be runes, and the frequent occurrence of errors or mistakes in
runic inscriptions that have been apparently carved by someone who did not
really understand the text. If these sorts of inscriptions are taken as evidence of
Anglo-Saxon runic literacy, they would seem to indicate a very low level of
literacy, not to mention shoddy craftsmanship; but, on the other hand, they are
just what we would expect if some less skilled artisans were attempting to add
prestige to their work by decorating it with a script in which they were not fully
competent. One did not have to be able to read the runes for them to fulfil this
decorative function.

31
BREDEHOFT, “Multiliteralism”.
32
HINES, “Some observations”, pp. 71-73; LOOIJENGA, Texts and Contexts, p. 107. N.
HOLDER, “Inscriptions, writing and literacy in Saxon London”, Transactions of the London and
Middlesex Archaeological Society 49 (1998), pp. 81-97, at p. 94, notes that London is an
exception, but the more mundane runic objects found there point at least to artisans who aspired
to the creation of higher-status objects. By contrast, mundane uses of Scandinavian runes are
exemplified by the many informal messages scratched onto pieces of wood excavated in and
around Bergen, Norway; these come not only from a different runic tradition but also a later one
than the Anglo-Saxon inscriptions.
33
HINES, “Some observations”, p. 73; HOLDER, “Inscriptions”, p. 93; OKASHA, “Anglo-
Saxon inscribed rings”, pp. 40-41. For English writing in Roman letters as a display script in
twentieth-century contexts, see T. MCARTHUR, The English Languages (Cambridge, 1998), pp.
14-16, 27.
Stating the Obvious in Runes 137

Decoration also explains the existence of apparently pointless alphabetical

inscriptions such as “fuþorcglaæe” on the eighth-century Malton pin (North
Yorkshire) or the futhorc sequence on the Thames scramasax (seax of Beag-
noth, London).34 There is no practical purpose for inscribing the beginning of
an alphabet on the head of a pin, or a full alphabet on the blade of a cutting tool
or weapon. The inutility of the inscriptions is highlighted by their ineptness.
On the Malton pin, the futhorc sequence switches to an inexplicable vowel
series; the futhorc sequence on the Thames scramasax contains numerous er-
rors. When modern scholars are confronted with a painstakingly crafted runic
inscription that makes no sense, they are tempted to suggest that the runes were
being employed because they were believed to have magical properties. But,
given the overwhelming lack of evidence for magical uses of runes in Anglo-
Saxon England, and their appearance so often in overtly Christian settings, the
‘rune-magic’ hypothesis is the least likely; and some have remarked that if
runes had magical properties, one would expect the practitioners of such magic
not to make so many mistakes with their runes.35 It is much more probable that
the futhorc inscriptions on the Malton pin and Thames scramasax are decora-
tive, that in both cases the artisan was better at using runes for ornament than
for communicating semantic information.
Elites, like other social groups, are formed by including certain people and
excluding others. Choosing a specialised script that only a tiny minority can
read creates that exclusiveness; the cryptic nature of these texts separates the
adept from the rest of the population. In making this argument I agree with
Okasha, who, in discussing Anglo-Saxon inscriptions that use both roman and
runic letters, suggests that runes were chosen because they were regarded as
decorative, cryptic, archaic, or esoteric,36 and Seth Lerer, who argues that, for
the Anglo-Saxons, runes were a link to a mythology of writing, representing
“an alien or ancient form of communication”.37 We must remember that early

34
Both objects are currently held by the British Museum: the Malton pin, number
2000,0508.1; the Thames scramasax, number 1857,0623.1. On the Malton pin, see LOOIJENGA,
Texts and Contexts, p. 294, and compare the futhorc inscription on the Brandon pin, PAGE,
Introduction, p. 81. On the Thames scramasax, see PAGE, Introduction, p. 80.
35
BARNES, Runes, p. 49.
36
OKASHA, “Script-mixing”. This also accords with the use of runes in manuscripts: see R.
DEROLEZ, Runica Manuscripta: The English Tradition (Brugge, 1954), pp. XXII-XXV, and L.
BRAGG, “Runes and readers: In and around ‘The Husband’s Message’”, Studia Neophilologica
71 (1999), pp. 34-50.
37
S. LERER, Literacy and Power in Anglo-Saxon Literature (Lincoln, NE, 1991: Regents
Studies in Medieval Culture), p. 17. See also B. TILGHMAN, “Writing in tongues: Mixed scripts
138 YIN LIU

medieval English runes were created in a society where literacy of any sort was
a specialised skill, and writing itself a form of esoteric knowledge.
In conclusion, let us consider one last Anglo-Saxon runic inscription, the
text on the Baconsthorpe object (Norfolk). We are unsure what, exactly, this
small copper-alloy object is; it has been called a set of tweezers, but Hines
suggests that it is a kind of book clip or page-turner.38 The inscription on the
flattened head reads (edited) “rede se þe cuinne, bæu þas rune awrat” (“read
whosoever can, Beaw wrote these runes”). This can be compared with the
cryptographic message (in roman script) from the scribe on f. 16v of MS Lon-
don, British Library Cotton Vitellius E XVIII, which, when deciphered, reads:
“Ælfwine me wrat . ræde ðu ðe cenne” (“Ælfwine wrote me; read me, whoever
can”).39 I am reminded of the coded Scandinavian runic inscriptions on bone
and wood, recently deciphered by Nordby, which, when decoded, say “ráð þat”
(“read this”).40 In Anglo-Saxon England, likewise, any runic inscription drew
attention to the skill of the writer and, if deciphered, to the skill of the reader.
In some cases, the message of the text was quite incidental.
When we ask questions about runic literacy in early medieval England, we
always risk making assumptions about medieval literacy that reflect modern
ideas of literacy. We find it easy to assume, for example, that a person writes
something intending it to be read (even if the only reader is the writer, as in the
case of a private journal); that writing involves depositing some sort of mean-
ingful information in the text; that reading involves extracting that information
with minimal confusion or loss of meaning; and that the primary function of
writing is to enable this kind of information transfer. We also often expect the
information in question to be conveyed by the written text at its most abstract
level (the ‘code’), not by the material form of the writing or indeed by the act
of writing itself.41 To modern scholars from societies where functional literacy

and style in Insular art”, in: Insular & Anglo-Saxon Art and Thought in the Early Medieval
Period, ed. C. HOURIHANE (University Park, PA, 2011: Index of Christian Art Occasional Papers
13), pp. 92-108, at pp. 93-94.
38
HINES, “New light”, p. 282.
39
P. PULSIANO, “The prefatory matter of London, British Library, Cotton Vitellius E.XVIII”,
in: Anglo-Saxon Manuscripts and Their Heritage, ed. E.M. TREHARNE and P. PULSIANO (Brook-
field, VT, 1998), pp. 85-116, at p. 99.
40
K.J. NORDBY, “Ráð þat, if you can!”, Futhark 3 (2012), pp. 81-88.
41
This is, of course, the familiar ‘conduit metaphor’ of communication described by M.J.
REDDY, “The conduit metaphor: A case of frame conflict in our language about language,” in:
Metaphor and Thought, ed. A. ORTONY (Cambridge, 1979), pp. 284-310, and discussed by G.
LAKOFF and M. JOHNSON, Metaphors We Live By (Chicago, 1980), pp. 10-12. As Lakoff and
Stating the Obvious in Runes 139

is the norm rather than the exception, the act of reading becomes deceptively
transparent: we try to look through the code in order to see the message. Deco-
rative writing, indecipherable writing, or writing that conveys an utterly banal
message will seem, if these are our prejudices, to be practically meaningless –
at least, nonfunctional or pointless.
Why, then, would a medieval artisan state the obvious in runes? The an-
swer is that anything written in runes was not so obvious. In other words, there
is a strong likelihood that when an Anglo-Saxon artisan chose to inscribe a text
in runes rather than in roman letters, the choice was motivated in at least some
cases not by a desire to maximise readability but by a desire to obstruct read-
ability. The point of runic inscriptions that state the obvious was not to convey
information but, first, to demonstrate the carver’s cleverness in mastering an
esoteric script – allowing both the carver and the reader to participate in elite
culture – and, secondly, to give the reader the sense of smug satisfaction that
comes from solving a puzzle and belonging to the exclusive minority who can.
The precondition for such strategies – a society where literacy was a special-
ised skill, but where literacy in one of two available scripts was even more
specialised – was present in Anglo-Saxon England.42 These runic texts present
reading as a challenge, drawing attention not to the meaning of the message but
to the mysterious process by which these scratches and signs can create mean-
ing in the first place, and by which bone and stone and metal can be made to
speak. In doing so, they recover even for us, twenty-first-century literates, that
moment of wondrous recognition that we experience every time we read, when
the marks on the page become words.

Johnson observe, “The CONDUIT metaphor does not fit cases where context is required to
determine whether the sentence has any meaning at all and, if so, what meaning it has” (p. 12).
For the persistence of the metaphor in linguistic models of communication, see P.L. BLACKBURN,
The Code Model of Communication: A Powerful Metaphor in Linguistic Metatheory (Dallas,
2007). Related to these issues in many ways is the distinction between semantics and pragmatics;
see the Introduction to Semantics versus Pragmatics, ed. Z. GENDLER SZABÓ (Oxford, 2005), pp.
1-14.
42
It may have been present in other times and places as well, but it is striking that this sort
of autoreferentiality appears in the runic corpus not only in England but also elsewhere in
northern Europe. Whether it is exclusively associated with runic writing is a wider question that
deserves further study.
 Labours Lost:
William Caxton’s “Otiose” Sorts, c. 1472-1482

ANYA ADAIR

Introduction

t is often observed that incunabula stand as a bridge between late medieval

I manuscripts and the printed books of the sixteenth century. The bibliogra-
pher Curt F. Bühler, writing in 1949, reminds his readers of what follows
from the position of early printed books thus caught between two technologies:
“they partake of the problems of both”.1 A similar truth has been brought home
in our own age by the advent of digital publication: this modern transposition
of the means of production involves dealing with both the values and assump-
tions of the book-as-artefact, and the capacities and constraints of the powerful
new technology through which it is to be represented. These values can them-
selves become clearer, their terms and significance more explicit, in the act of
recreating a text through a new process. “Before a text can be encoded”, Lou
Burnard writes, “it must first be decoded”.2 That is to say, the technological

1
C.F. BÜHLER, Standards of Bibliographical Description (Philadelphia, 1949), p. 4.
2
L. BURNARD, “On the hermeneutic implications of text encoding”, in: New Media and the

......................................................................................................................................
Verbal and Visual Communication in Early English Texts, ed. Matti PEIKOLA, Aleksi
MÄKILÄHDE, Hanna SALMI, Mari-Liisa VARILA, and Janne SKAFFARI, Utrecht Studies in Medieval
Literacy, 37 (Turnhout: Brepols, 2017), pp. 141-164.

DOI 10.1484/M.USML-EB.5.114134
142 ANYA ADAIR

shift forces those responsible for the new representation to make explicit and
assign a value to each of its visual elements. The move from manuscript to
early print involves a similar interpretive exposure: printers working to encode
mechanically the scribal forms of manuscript had first to decode the meanings
of these forms. Examining the choices they made within (and sometimes de-
spite) the constraints of their new technology helps to make explicit fifteenth-
century understandings of some obscure features of manuscript and print pro-
duction.
This chapter is concerned with phenomena found in both English manu-
script and English print during the first decade of William Caxton’s book pro-
duction: the representation of the same letter in various forms, and the more
perplexing appearance of ornamentation on certain letters. In fifteenth-century
English manuscripts, the letter d (for example) might appear in both looped and
unlooped forms, and in forms for use in ligatures; or it might be adorned by a
final flourish when it appears at the end of a word. Individual scribes might
further modify the shape of the letter according to its interaction with surround-
ing letters, obeying a complex of influences in which the efficiency of their
ductus (the direction and sequence of the strokes that form each letter) and the
aesthetic of the finished word are contributing factors. In early English print,
the attempt to reproduce this scribal versatility by mechanical means resulted
in a profligate production of pieces of type (known as sorts) and a consequent
bewilderment of choice for the typesetter: each letter was cut in several forms;
the resulting sorts then had to be separately stored in the printer’s case, and
carefully selected in the setting process.
Especially interesting is the transfer of apparently decorative or otiose
marks from fifteenth-century manuscript (where they are problematic in them-
selves) to the rigidity of pre-cast print. In manuscript study, otiose strokes are
marks added to some letters (in the form of macrons, hairline crosses, or loops
at the end of words) which do not seem to act as marks of abbreviation, but
which are nevertheless regularly used by many English scribes.3 Their purpose
remains uncertain: explanations offered by palaeographers and editors have
ranged from the grammatical and orthographic to the merely ornamental.
Whether they should be taken to affect the spelling of the words on which they
appear remains an unsettled issue. Individual scribes may have deployed the

Humanities: Research and Applications, ed. D. FIORMONTE and J. USHER (Oxford, 2001), p. 43.
3
On otiose strokes in manuscript, see, e.g., M.B. P ARKES, English Cursive Book Hands,
1250-1500 (Berkeley, 1980), p. XXVI.
Labours Lost 143

strokes according to an internally consistent system, but no single schema is

discoverable from the totality of the fifteenth-century manuscript evidence.4
Given the technical challenges and associated expense that such strokes present
to the process of printing in metal type, it is intriguing that they were still con-
sidered sufficiently important to the English book for many of them to be
painstakingly reproduced in Caxton’s early works – and this at some cost.
David R. Carlson writes drily that the

grotesquely oversized bills of fount met with in fifteenth-century printing ... some-
times seem designed as if meant to undo exactly what Gutenberg’s technical dis-
covery had made it possible to do in the first place.5

Lotte Hellinga, similarly concerned with the manageability of the enterprise,

complains that though the type used in setting the Gutenberg Bibles required
some 280 individual letter designs “logically a type-case should contain hardly
more than 70”.6 Hellinga’s economy would leave little room for multiple ver-
sions of individual letterforms; in the early decades of print, however, printers
moved enthusiastically in the opposite direction. Caxton commissioned and
used several different types over the course of his career as a printer. His Type
1, used for the earliest of his English printing, required at least 162 different

4
For an example of a consistent system used by one scribe in this period, see R. DAHOOD,
“Abbreviations, otiose strokes and editorial practice: The case of Southwell Minster MS 7”, in:
New Perspectives on Middle English Texts: A Festschrift for R.A. Waldron, ed. S. POWELL and
J.J. SMITH (Cambridge, 2000). My own forthcoming study of the manuscript production of the
Carthusian scribe William Darker (fl. 1481-1512/1513) supports the claim that a scribe’s
deployment of such marks could be complex, consistent and linguistically motivated. On the dif-
ficulty of discerning any over-arching system across the work of multiple unconnected scribes,
see V. NELSON, “Problems of transcription in the ‘Speculum Vitae’ MSS”, Scriptorium 31 (1977),
pp. 254-259.
5
D.R. CARLSON, “A theory of the early English printing firm: Jobbing, book publishing,
and the problem of productive capacity in Caxton’s work”, in: Caxton’s Trace: Studies in the
History of English Printing, ed. W. KUSKIN (Notre Dame, IN, 2006), pp. 35-68, at p. 49.
6
L. HELLINGA, “Printing”, in: The Cambridge History of the Book in Britain, 3, 1400-
1557, ed. L. HELLINGA and J.B. TRAPP (Cambridge, 1999), pp. 65-108, at p. 70.
144 ANYA ADAIR

letterforms; Type 2 nearly doubled this amount, demanding over 250.7 Why
this expensive enthusiasm?
It resulted in part from the decision to reproduce certain letters both singly
and as a series of pairs: both Types 1 and 2 contained the combinations la, le,
li, lo, lu in addition to single l. This generosity of sort allowed the printed
works more accurately to capture the visual effect of manuscripts in which
certain letters overlap or nestle attractively together. More sorts result from the
cutting of some letters both with and without apparently decorative strokes: in
Caxton’s Type 1, the sorts for l include the letter with and without a curved
head stroke, and in crossed and uncrossed varieties. The same options have
been cast for the doubled ll (see Fig. 1).

Fig. 1 Caxton’s “Type 1: 120” letter l, as recorded by William Blades.

In the 1480s, Caxton abandoned these expansive founts in favour of less varied
type; but for at least a decade, what ruled in English print was a wide variety
of differentiated letterforms along with an array of apparently decorative op-
tions.
In the brief tenure of these visual elements – expensively created, painstak-
ingly deployed and quickly discarded – can be traced something of the govern-
ing principles of early English print development. The mechanical reproduction
of the forms fashioned by a scribal hand implies more than a turn from the
scribe to the metalworker, from the language of ductus, stroke, and the flow of
ink from moving nib to the logic of metal matrix, sort and type. It more essen-
tially involves a re-evaluation by printers of established visual features of or-
thography and letterform. How is the visual result of a fluid hand to be cap-
tured by a mechanic? What aspects of this fluidity were valued and so retained
by the printer? And can those decisions help us to recreate something of the
meaning they held for the original scribe and his readers?

7
The assessment here, and the reproductions of Caxton’s Type 1 and Type 2 that follow,
make use of the classifications and systematic reproductions of fount in W. BLADES, The Life and
Typography of William Caxton, England’s First Printer, 2 vols. (Cambridge, 1861-1863, repr.
2014), esp. plates XI, XIII, XVI, XVIII, XXI, and XXIII. Where my own reckoning of Caxton’s sorts
differs from that of Blades, note is made; for a more comprehensive critique of the work of Blades
and his pen facsimilist, G.I.F. Tupper, see J.A. DANE, The Myth of Print Culture: Essays on
Evidence, Textuality and Bibliographical Method (Toronto, 2003), pp. 75-87.
Labours Lost 145

Making and Using Types 1 and 2

Johannes Veldener, a punch-cutter and printer in Flanders who worked

closely with Caxton, is believed to have been responsible for cutting Caxton’s
Types 1 and 2.8 The precise process of making type in this era, however, re-
mains obscure. Joseph Dane reminds us that “there is very little evidence re-
garding the nature of fifteenth-century printing beyond the products them-
selves”.9 Despite this uncertainty, we may be reasonably sure of the basics, as
Pamela Robinson summarises them:

A letter was first cut in relief on one end of a long rectangular piece of steel known
as a punch. The punch was then struck into copper to form a matrix or mould in
which to cast type. Each resulting piece of type formed a small piece of metal ...
with a letter on one end.10

Deciding which letters to cut, strike and cast was a matter of no small moment:
the process was a complex and costly one. Including fine detail on multiple
designs presumably added significantly to its expense and difficulty.
Caxton’s Type 1: 120 and Type 2: 13511 are both similar to a script broadly
known as lettre bâtarde, which was used in de luxe manuscripts in France and
Burgundy in the fifteenth century. Characterised by features such as its single
compartment a and short r, the script had a “prickly appearance” created by the
tapering descenders and its preference for broken, angular strokes rather than
curved ones.12 In the hands of a master scribe it could be stately and even, and
was associated with manuscript productions of the highest quality.13 The exact
provenance of Caxton’s early designs, however, remains uncertain. Type 1 was
once thought to have been modelled on the hand of Colard Mansion; David
8
L. HELLINGA, Caxton in Focus: The Beginning of Printing in England (London, 1982),
p. 51.
9
J. DANE, Out of Sorts: On Typography and Print Culture (Philadelphia, 2011), p. 19.
10
P. ROBINSON, “Materials: Paper and type”, in: A Companion to the Early Printed Book
in Britain 1476-1558, ed. V. GILLESPIE and S. POWELL (Cambridge, 2014), pp. 61-74, at p. 64.
11
The body-size measurements following the colons here help to distinguish the types by
representing, in millimetres, the average measurement of twenty lines of type (from the foot of
m on the first line to the foot of m on the twenty-first line below).
12
ROBINSON, “Materials”, p. 65.
13
Not everyone has thought so well of the bâtarde form: Updike condemns Type 1 for its
“rough, angular, awkward design, which shows clearly its relation to current Flemish handwriting,
which was rough, angular, and awkward too” (D. UPDIKE, Printing Types, Their History, Forms,
and Use: A Study in Survivals, 1 (Cambridge, 1922), p. 115).
146 ANYA ADAIR

Aubert, another producer of de luxe manuscripts, is now considered a more

likely candidate. Caxton certainly worked with Mansion in Bruges; he is sup-
posed to have formed a relationship with Aubert in Ghent.14 If Type 1 is based
on the hand of Aubert, Type 2, a slightly larger and statelier variety, appears to
belong to a more general Flemish style.15
In order that it suit printing in English, any Continental model of script had
to be augmented by the English letters w, yogh, and the y-form that functioned
to represent the English thorn. Caxton could not simply reuse founts fully pro-
duced but intended for Latin. Nor is his a Continental fount minimally modi-
fied by the addition of just the three missing English letters. The founts he
commissioned are replete with specifically English features, though these
mostly take the form of details whose significance might escape the modern
eye. In Type 1, for example, a tailed form of d is found in addition to simple d,
and the ascenders of h and l appear in both crossed and uncrossed varieties.
Such features do not appear in the scripts of either Mansion or Aubert; they are
not seen in Type 1 settings of French-language texts.16 These are uniquely
English choices, and seem to have been created for English-language printing.
None, it could be argued, affects in any meaningful way the spellings encoded
by the type. The marks – and thus the sorts on which they appear – would seem
to be otiose. But they were perceived as being so needful to both Type 1 and
Type 2 that they were cut and cast, and deployed with great care in the setting
of Caxton’s early English works.
Once the type is cast and sorted for use in the printing workshop, the type-
setter is required to choose the correct sort not merely in absolute orthographic
terms, but according to its interaction with surrounding letters: certain combi-
nations will appear in ligature; others will take a decorated form only in certain
14
For the argument that Caxton worked with Aubert in Ghent to produce works printed with
Type 1, see L. HELLINGA, “William Caxton, Colard Mansion, and the printer in Type 1”, Bulletin
du bibliophile (2011), pp. 86-114.
15
On the relationship of Caxton’s early sorts to Burgundian and Flemish scripts, see the
account in ROBINSON, “Materials”, p. 65. On the founts available in the Low Countries, and the
general appearance of type in this period, see W. and L. HELLINGA, The Fifteenth-Century
Printing Types of the Low Countries, trans. D.A.S. REID (Amsterdam, 1966). For a representative
account of Caxton’s borrowing of a general manuscript aspect into his early typography, see N.F.
BLAKE, “Manuscript to print”, in: Book Production and Publishing in Britiain 1375-1475, ed.
J. GRIFFITHS and D. PEARSALL (Cambridge, 1989), pp. 403-432, at p. 404.
16
The absence from the French of decorated sorts in crossed ll and h and tailed d is not
merely a matter of spelling: though the letter d, for example, is less common at word-end (where
the tailed form would be used in English), when the letter does appear in this position, it carries
no tail.
Labours Lost 147

positions or beside certain other letters. For a scribe, this sort of contextual and
aesthetic integrity is a matter of practised movement, whose economy is a
physical one, constrained by the muscles of the hand, the quality of the nib, and
the flow of ink onto the support. Moreover, a scribe can decide, once a word or
line is complete, that its letters require some adornment. The scribe’s organic
process, responsive to the emerging appearance of the written line and folio,
becomes a mechanical one when transferred to the printing workshop, depend-
ent upon choices limited by the sorts in the case. The setter’s imagined antici-
pation of the printed appearance of a word suggests the most appropriate com-
bination of sorts; constraints in selection depend upon the space available in
the line and the forme (the block of type composited to make a page), and how
many of each sort remain in the type case. This is a very different economy.
The result of all these processes in the production of works in Types 1 and
2 – from the commissioning of the fount, its cutting, striking and casting,
through to the product of the setter’s labour – was a printed page of hybrid and
unique appearance. The aspect of the Continental scripts, with their broken
strokes and sharply-pointed minims, was softened and made familiar to an
English audience by the horizontals and curves of English features: the hairline
crosses and trailing tails common in Anglicana and the English interpretations
of Textura and Secretary.
But given their Burgundian models, Robinson argues, these types “must at
first have appeared foreign-looking to English readers”. Indeed, she suggests,
“unlike printing in the vernacular elsewhere, it took a long time before type-
faces were created here that gave printing in England ‘a style of its own’”.17
This assessment, however, may undervalue the effect of the apparently decora-
tive features so painstakingly added to the early founts. The adorning strokes
and English letters gave Caxton’s first types an English aspect, even if their
underlying letter shape did not. This Burgundian wall with English stucco,
established by Type 1, was further domesticated in Type 2 with the addition of
a two-compartment Anglicana form of a, a looped d, a far greater number of
nasal macrons (thick lines above the letters m and n), and a tail for word-final
g to match that already available for word-final d. Yet neither Type 1 nor Type

17
ROBINSON, “Materials”, pp. 65-66. Robinson here refers to the claim of L. HELLINGA,
“Printing types and other typographical materials”, in: Catalogue of Books Printed in the XVth
Century Now in the British Library, Part XI: England, ed. L. Hellinga (’t Goy-Houten,
Netherlands, 2007), p. 335. For a discussion of Caxton’s introduction of Continental type to
England, see N. BARKER, “Caxton’s Typography”, Journal of the Printing Historical Society 11
(1976-1977), pp. 114-143.
148 ANYA ADAIR

2 was especially practical for print: both had a great many ligatures, the angle
of certain tall letters interacted awkwardly with more upright varieties, and
subtly (for the typesetter, probably irritatingly) different forms of certain letters
were required for use at word-end. Why did Caxton bother in the first case;
why did he so surprisingly (from the privileged perspective of hindsight) per-
sist in the second? Perhaps his Continental experience and perspective suggest
an answer: he was well positioned as an outsider, as it were, to see what was
English about his native scripts. As a businessman, he was equally well placed
to foresee the need to tailor his product to his intended market. The result of
his assimilation of Burgundian script with English interpolations and overlay
created a kind of trans-Channel visual accent that characterises his early types.
In the tiny ‘otiose’ marks of Types 1 and 2 there is evidence of a matter of
great contemporary importance: the Englishness of the whole text.

The Setting of Type 1 and the Deployment of ‘Otiose’ Sorts

Assessments of Caxton’s technical prowess as a printer have not been

glowing.18 A lack of technical proficiency certainly appears in the rather erratic
treatment of letters in the fount of Type 1, which must have been commis-
sioned and produced in 1472, and which was used in the printing of the Re-
cuyell of the Historyes of Troye and The Game and Playe of Chesse in 1473-
1474, in Bruges or Ghent.19 In the Type 1 fount, some letters are commissioned
in a range of forms and combinations, while others appear in relatively few
varieties.20 As a result of this uneven distribution of sorts, setting occasionally

18
See, e.g., UPDIKE, Printing Types, Their History, Forms, and Use, pp. 1, 113; T. ATKIN
and A.S.G. EDWARDS, “Printers, publishers and promoters to 1558”, in: A Companion to the
Early Printed Book in Britain, 1476-1558, pp. 27-28.
19
R. LEFÈVRE, [here begynneth the volume intituled and named] the recuyell of the
historyes of Troye, trans. W. CAXTON ([Bruges, 1473-4]: STC 15375); J. DE CESSOLIS, The Game
and Playe of Chesse, ([Bruges, 1473-4?]: STC 4920). Three French texts were also produced with
this type. For a detailed discussion of the chronology and geography of Caxton’s Continental
Type 1 printing, see L. HELLINGA, William Caxton and Early Printing in England (London,
2010), pp. 26-51.
20
A further variability was the existence of more than one matrix for a single letter-form:
slight variation in appearance caused by this production contributes to what William Blades calls
the “freedom” in the appearance of the type. (BLADES, The Life and Typography of William
Caxton, pp. 1, 111.) Such variations are not of concern here: my interest is in the deployment of
forms that would have been differentiated in the sort-case.
Labours Lost 149

produces inconsistent spacing within and between words.21 Where tall-s ap-
pears, for example, it is often followed by an apparent gap as the sort-slug (the
moveable piece of metal holding the letter) accommodates the right-facing
shoulder at the top of the letter. Unexpected mid-word gaps such as this are
accentuated by the use elsewhere of sorts containing a closely placed pair of
letters. In Fig. 2, compare especially the spacing of the letters within the word
shall on line one (created with three sorts sh-a-ll) and schall on line two (cre-
ated with four sorts s-ch-a-ll). Combined sch, though available, seems to have
been cast in numbers insufficient to meet the setter’s needs.22

Fig. 2 Recuyell of the Historyes of Troye, Huntington 62222, The Huntington Li-
brary, San Marino, California, p. 5, detail (EEBO image 3).23

For present purposes, this clumsiness is useful: it allows for a clearer as-
sessment of the ways in which typesetters responded to the constraints of sort
availability. The case of shall is revealing: all seven instances of the <schall>
spelling (which tend to produce that unexpected gap following the tall-s sort)
appear on the page upon which the word is first used. The appearance of this
word was apparently unsatisfactory; the solution was to reject the orthography
entirely, and settle on sh spellings: the remaining 821 uses of shall in the Re-

21
The vertical alignment of the letters in relation to one another, however, thanks to the skill
of Veldener’s uniform casting, is remarkably even. See the account in W.J. PARTRIDGE, “The use
of William Caxton’s Type 3 by John Lettou and William De Machlinia in the printing of their
‘Yearbook 35 Henry VII, c.1481-1482’”, The British Library Journal (1983), pp. 56-65, at p. 56.
What Partridge calls “Type 3” is labelled Caxton’s Type 2 by William Blades and others, a more
standard categorisation followed in this study.
22
I cannot find more than two uses of this sort on a page, even where multiple sch spellings
appear.
23
This image (and all others from the Recuyell of the Historyes of Troye) is reproduced from
Huntington Library 62222 with the permission of the Huntington Library. For ease of reference,
images are also labelled as they appear in the Early English Books Online database, which
reproduced the Huntington text. See EEBO Raoul Lefèvre, hEre begynneth the volume intituled
and named the recyell of the historyes of Troye (http://eebo.chadwyck.com).
150 ANYA ADAIR

cuyell are spelled <shall> or <shal>. Constraint of sorts appears in this case to
have determined and fixed the spelling.
The subject of the following examination of sort varieties and their deploy-
ment is Caxton’s Type 1 as it appears in the Recuyell of the Hystories of Troye.
The deployment of certain sort varieties that appear to be otiose is of interest;
such sorts are found among the options for the lower-case letters d, h, l, and n.
For the letter d, a tailed and a tailless sort are available; for h and l there are
crossed and uncrossed sorts, as well as versions of l with and without a curved
top, and for n, there is a sort for the plain letter and one for n surmounted by a
macron. To analyse the deployment of these marks, pages were selected from
the Recuyell in groups of four, spaced at approximately twenty-page intervals;
one-fifth of the total text was in this way examined.24 Using the images and
transcription available in the Early English Books Online database, all in-
stances of relevant letters were recorded, with note made of the sort deployed,
the orthography and setting of the word, and its position within a line. The
resulting database of over 5,000 individual sort deployments suggested broad
patterns of use that were investigated via an analysis of the minutiae of the
setting context on individual pages.

Fig. 3 Recuyell of the Historyes of Troye, Huntington 62222, The Huntington Li-
brary, San Marino, California, p. 3, detail (EEBO image 2).

24
The interest of the study is in broad deployment patterns: attention was not paid to quire
division. Hellinga’s chronology of the spread of the two-pull press and her assessment of evidence
for one-pull press use in Caxton’s early works confirms that one-page printing and setting was
used for the Recuyell (L. HELLINGA, Texts in Transit: Manuscript to Proof and Print in the
Fifteenth Century (Leiden, 2014), pp. 11-13 and esp. p. 14). No copy-text survives for any
English-language Type 1 settings. The text is taken to have been (most probably) set seriatim, in
the order in which pages appear in the finished text. Two presses may have been in simultaneous
use during production of this work and the other Type 1 works, with both presses drawing on the
Type 1 sort-case; this, if true, would have affected the availability of sorts. For an account of the
setting order for incunabula, see HELLINGA, “Notes on the order of setting a fifteenth-century
book”, Quaerendo 4 (1974), pp. 64-69.
Labours Lost 151

Two Sorts of d

The tailed-d sort is reserved for use at the very end of a word. Almost 90%
of word-final d appearances are set with the tailed sort. The preference for this
mark is clearly strong, yet an apparent omission rate of over 10% is surpris-
ingly high. Two mechanical considerations present possible explanations. First,
an occasional need to squeeze more letters into a single line or forme (in which
case the tailed-d sort, with its greater width, might be sacrificed to save space);
and secondly, the possibility that tailed-d sorts were in limited supply, and
occasionally ran out. One of these scenarios is certainly implicated by the fact
that on at least two pages the tailless-d sort appears with increasing frequency
towards the end of the forme.25 In a few heavily abbreviated lines, spatial con-
siderations also seem to be urgent: note the tailless-d sorts in Fig. 3, line 2
(“englissh whiche sche comanded me to amende and mo-”). These lines, ap-
pearing toward the end of the page, may omit tailed-d because of a shortage of
the sorts; equally, the absence may control the length of a line, part of a reper-
toire of printers’ strategies that includes modifications to spacing, abbreviation
use, spelling and phrasing.26
But though the average number of words ending in d remains relatively
stable across the book (around 35 per page) the average number of tailless-d
sorts deployed in word-final position decreases considerably towards its end.
On the first 200 pages, an average of 5 word-final ds per page lack a tail (these
are found predominantly in clumps towards the end of certain pages); in the
second 200 pages, this decreases to about 3.5 per page; by the final 300 pages,
the average rate of omission is barely over one per page. An omission rate of
almost 15% in the first portion of the book has fallen to under 2.5% by its end.
This final rate would be the expected one for ordinary setting error – a kind of
compositor-typo – and need imply nothing more than momentary inattention.
Setters can perhaps be seen here to become more adept at selecting sorts and
spacing their lines over the course of the printing; given the sharp increase in
the presence of word-final tailed-d sorts towards the end of the book, it is also
possible that more of these sorts were cast during the printing period.27 Cer-

25
EEBO images 2[b] and 61[a]. On the greater inaccuracy of text towards the end of printed
pages (argued to result from mechanical constraints) see L. HELLINGA, “Manuscripts in the hands
of printers”, in: Manuscripts in the Fifty Years after the Invention of Printing, ed. J.B. TRAPP
(London, 1983), pp. 3-11.
26
BLAKE, “Manuscript to print”, p. 409.
27
The possibility that tailed-d sorts occasionally ran out towards the end of a forme might
152 ANYA ADAIR

tainly, consistency in the application of this letterform increases over time, a

fact which might argue its relative importance to the typesetter.
Within the orthographic norms of the period, words ending in a final d may
also often be spelled with a final de: the pairs <world> / <worlde> and <yeld>
/ <yelde> are attested in the Recuyell, for example. Throughout the Recuyell,
spellings with final d are generally preferred to those with de. In some uses, a
final e relates to the length of the preceding vowel (as in the alternative spell-
ings <maad> and <made>); or may help to clarify meaning (<dede> is the only
recorded spelling of ModE ‘deed’, while both <ded> and <dede> can represent
ModE ‘dead’, for example). In some fifteenth-century scribal practice, a tailed
word-final d can plausibly be interpreted as a mark of abbreviation for a final
e,28 and thus might function orthographically to mark just such phonological
differences as these. But the setters of Caxton’s workshop were either unaware
of this possibility, or unwilling to make use of its potential. No pattern exists
to suggest that for Caxton or his setters, the tailed d is an abbreviation for a
spelling in de. This would, in fact, make little sense within the print-room
economy, as a biting-de sort is available, and is approximately the same width
as the tailed d. The only apparent reason for using a tailed-d sort was to give
words an English finish.

Two Sorts of h

In English vernacular manuscripts of the fifteenth century, a hairline often

adorns the ascender of h, especially where the letter is placed towards the end

Fig. 4 Caxton’s Type 1: 120 letter h, as recorded by William Blades.

of a word. The deployment of this stroke is a matter of scribal preference: some

scribes mark every h with a hairline, others mark only the ascenders appearing
towards the end of a word, and still others use no hairlines at all. In type, the

also support the hypothesis that two concurrent presses were used in the setting of the two English
books; more type sitting in formes would increase the demand on the sort-case.
28
In the editing of medieval texts, decision to expand such strokes to word-final e remains
within the judgment of individual editors and commentators; little consensus exists.
Labours Lost 153

choices available to all future setters are fixed at the time of casting. Caxton’s
Type 1 options for h can be seen in Fig. 4. In addition to these combinations,
there are two sch sorts, one with and one without a cross; crossed and un-
crossed varieties of a th sort; an uncrossed sh sort; and an uncrossed ch sort.29
The constraints of this array are immediately obvious. If the spelling ch
should appear, for example, the ascender of h cannot bear a cross-stroke unless
two sorts are used in place of one. The combination ht, conversely, is only
available in a crossed variety unless set as two sorts. These casting decisions
appear to have been the result of a clear sense of where the crossed form of h
should be placed: at word end (where ht and th are common letter combina-
tions) and not in the middle of a word (where all ch combinations appear).
Caxton, commissioning the fount, seems to have been sure that all words with
ch in the last syllable would take a final e: a crossed variety of the sort was
therefore not cast. This same casting decision and spelling preference is appar-
ent in Type 2: no crossed ch is available in this fount either.30 In the intention
of the commissioner at least, the use of the crossed h varieties should be where
the words end in the cluster ht, or the letter h (except in the combination sh).31
The deployment of crossed and uncrossed th is revealing of setters’ atti-
tudes. Though the sort almost never appears at the beginning of a word (the,
that, this etc. do not receive crosses), in word-final position the exercise of
some discretion is evident. Of the 92 recorded instances of <with> / <wyth> in
the Recuyell sample, only half contain crossed h (most of these are produced
with three sorts, w-i-th or w-y-th, but some appear to use four, w-i-t-h). Usage,
as a general rule, clumps together: a page will generally display either consis-
tently crossed or consistently uncrossed final th, not a mixture of the two. The
consistency in the patterns suggests that there may have been two setters work-
ing on the Recuyell. A possible reason for the unwillingness of at least one
setter to use the crossed th may be found in the execution of the sort itself: the
punch has been cut with a very fine hairline stroke, placed somewhat high and
29
W. Blades appears to have been unaware of the crossed th sort.
30
In a selection of 52 pages taken from the Type 1 setting of the Canterbury Tales, there
are only three words in which ch appears at word end – in each case, the uncrossed sort is used.
31
This widespread preference can be seen in, for example, the English-language works
copied by Carthusian scribe William Darker, working c. 1481-1512 (throughout MS London,
British Library, Add. 22121; MS Cambridge, University Library, Ff.vi.33; MS Oxford, Bodleian
Library, Laud Misc. 517; MS Oxford, Bodleian Library, Laud Misc. 38; MS Glasgow, University
Library, Hunterian T.6.18). The aesthetic created by the two ascenders of h and the preceding tall-
s seems to have determined the consistent omission of the hairline cross in manuscript production;
neither t nor c in manuscript is tall enough to interfere with the hairline.
154 ANYA ADAIR

crookedly. Avoiding the imperfect sort while keeping the word-final crossed h
involved the greater labour of setting the word with four sorts.
The clear intention to mark with a cross only ht combinations that are
absolutely word-final is compromised when final e or es appears: the only
option for combined ht is a crossed sort. The word <knyghtes> causes particu-
lar problems. Perhaps to avoid the need for eight separate sorts, it is often set
with the crossed-ht sort (k-n-y-g-ht-e-s); at other times, the te sort is used,
preceded by a separate uncrossed-h sort (k-n-y-g-h-te-s). This latter setting has
two consequences: uneven letter spacing (a gap creeps in between the h and t),
and a mismatch between the morphological boundaries and the visual effect of
the typesetting.32 Neither effect seems to have been pleasing, for setters tend to
avoid such choices where they can. The instability of setting suggests that the
want of uncrossed ht was felt. A similar instability affects the setting of
<myghte> and <mighty>: setters are often unwilling to forego the more eco-
nomical and visually appealing single sort, even when it means breaking the
convention that would allow crossed ht only in absolute word-final position.
Selection of the most appropriate sort, then, can be a difficult decision, and
appears to depend upon personal preference as well as an emerging house
style. The mechanical constraints identified here appear to have had an effect
upon the spelling itself. The difficulty created by the absence of uncrossed ht
may have influenced the fact that an e rarely follows ht in the final syllable
(<myght>, for example, far outnumbers <myghte>; similar patterns appear in
<ryght> / <ryghte> and <fought> / <foughte>). In Type 2, fewer consonant
clusters with h are cut; most such sequences are set with closely fitting individ-
ual sorts.33 This move may have relieved the spelling pressure: the strong Type
1 preference for word-final th above word-final the weakens considerably in
Type 2.34

32
Morphologically and syllabically, the word is knyght-es; the visual effect of the close-set
te sort faintly obscures this expected division.
33
William Blades’ survey of Type 2 sorts provides the initial basis for comparison of
consonant clusters (W. BLADES, The Life and Typography of William Caxton, 2, plates XI and
XIII); the broader survey of Type 2 deployments, which followed the method used for Type 1,
focused on the Westminster printing of the Canterbury Tales (G. CHAUCER , The Canterbury
Tales ([Westminster, 1477]: STC 5080), from which 62 pages were selected for examination, taken
in four-page blocks at 30-page intervals throughout the text.
34
The 6:1 ratio of th to the spellings in texts set with Type 1 shifts to approximately 3:1 in
texts set with Type 2.
Labours Lost 155

Many Sorts of l

A wide range of sorts is available to represent the letter l in Type 1: with

three different varieties of single l (straight; with looped head stroke; crossed),
this is among the most varied of the letters in the fount. Setters seem to have
found some of the distinctions rather too fine. There is a preference to use l
with looped head-stroke (the third glyph in Fig. 1) at the beginning of a word;
straight l (the second glyph in Fig. 1) tends to appear mid-word. But straight l
is used from time to time to begin words, and l with looped head-stroke appears
frequently mid-word. If these two sorts were separately stored at the beginning
of the print run, it is tempting to suppose that they were thoroughly mixed by
its end.
As was seen in Fig. 1, a crossed and an uncrossed sort are available for
both single and double l, giving setters four “spelling” choices (in addition to
a range of sorts containing l in combination with other letters). At the end of
the fifteenth century, double versus single l remains a highly variable ortho-
graphic feature. Interaction of the letter with word-final e is a further complica-
tion: <batayll>, <batayle> and <bataylle> are all attested spellings in the Re-
cuyell. The word <all> / <alle> confirms this variability: inclusion or omission
of final e follows no consistent pattern. Occasionally, a spelling in <alle> will
become dominant for a page or two; at other times, <all> will be ascendant.
More common than a preference for either, however, is a more or less random
mix of the two, and spelling appears to be constrained by space more than any
other consideration.35 One point of near-absolute consistency emerges, how-
ever: a spelling in <all> will be set with a crossed-ll sort, and a spelling in
<alle> will be set with an uncrossed-ll sort. This remains consistent across the
whole book. Surprisingly, though <all> requires only two sorts (a-ll), and is

Fig. 5 Caxton’s Type 1: 120 letter n, as recorded by William Blades.

35
In the freer space of the verse-line setting of the Canterbury Tales in Type 2, the <alle>
spelling is almost universal, and <all> is only found in prose.
156 ANYA ADAIR

thus the most efficient setting, it appears in only slightly above half of the
instances of the word.
In the case of <all> / <alle>, the consistent appearance of a final e where
the ll is not crossed suggests that the cross might function as an abbreviation
for a final e. Likewise, in the words batayll(e), fyll(e), hell(e), and wyll(e),
crossed ll is only and always used where no final e appears. But in other ortho-
graphic pairs the abbreviation hypothesis is less convincing: <counceyll> with
a crossed ll is at times spelled <counceyl> with no stroke at all, and <foull>
appears, crossed, on the same opening as uncrossed <foul>. Taken as a whole,
the weight of setting evidence suggests that crossed-ll may well have been
taken by the majority of setters to be a mark of abbreviation, but also that it
was tending towards a more decorative function. Since spelling with final e is
in no case absolutely required, readers need not understand crossed-ll to be an
abbreviation in order that the text should be comprehensible.
Whatever their function, crossed ascenders imply a different series of
choices in the printer’s workshop from those of the scriptorium. A scribe may
be presumed to loop the nib back through the ascenders of final letters or add
a quick hairline in a final flourish both fluid and reflexive. The setter of type,
however, must choose letterforms before setting them: in reaching for a
crossed-ll sort, therefore, he has already decided not to include the following
letter. Removing an e that made a line too long, the setter must exchange the
preceding sort also. The constraints of line-length determine not just the letter
to be omitted, but the sort selected for the preceding letters.

The Problem of Nasal Macrons

In manuscripts, macron over a word-final nasal holds a particularly ambivalent

position. Scribes writing in English often form the stroke as a heavy, separate
mark that is most suggestive of abbreviation (a similar macron placed above a
vowel always abbreviates a following m or n), but modern editors have been
wary of expansion: in many cases, the addition of this stroke to a word-final
nasal appears less orthographic than decorative. In the Type 1 fount, only n is
cast with a macron option (see Fig. 5); the sort was perhaps intended for use
Labours Lost 157

Fig. 6 Recuyell of the Historyes of Troye, Huntington 62222, The Huntington Li-
brary, San Marino, California, p. 678, detail (EEBO image 340).

in Latin passages, where it might more usefully function as an abbreviation.

But for setters wishing to mimic the English deployments of the stroke, a
problem results from the fact that (from the English perspective) the wrong
version of n has been cut with a macron. The Continental scripts used as a
basis for Caxton’s Types 1 and 2 affect a pointed trailing minim on nasals at
the ends of words. In Fig. 6, this can be seen in line 1 <vpon> and <hym>, and
in line two <whom>. Like the tailed d and crossed h and ll of English, this
trailing form is associated with a word-end aesthetic – so closely, in fact, that
the setter can omit from the tight second line the space that would usually
divide the elements of the compound hym self: the pointed final minim signals
to the reader that the first word is complete, even in the absence of a space.
In fifteenth-century English-language manuscripts, a nasal macron (when
used) is likewise consistently associated with word-final position. But in Type
1, the only variety of n cast with a macron lacks the trailing second minim of
a word-final form. To be useful for an English aesthetic, the n-with-macron
sort must appear at word-end; to fit the bâtarde style in this type, it cannot be
used in this position.
The setter of the Recuyell, setting the word th-a-n at the end of a line early
in the text, and wishing either to abbreviate thanne or (more probably) to deco-
rate this word- and line-final nasal in the English style, faced a dilemma: he
could have either a trailing minim or a macron, but not both. His solution, seen
in Fig. 7, is more creative than visually pleasing, but certainly demonstrates a
felt need for a sort not yet available. Here, the trailing minim sort for n is used,
and is followed by a macron placed over empty space. A similar impetus may
be behind the use of the a-macron sort preceding the final n of <mân> in line
2 of this figure: mann is an unlikely spelling, and in fifteenth-century manu-
script, the word man is one of the few frequently associated with a nasal ma-
cron.36

36
Speculating on the script of the copy-text in the absence of any survival is risky, but
whatever the script, these heroic early efforts of the setter to place a macron with word-final
158 ANYA ADAIR

Fig. 7 Recuyell of the Historyes of Troye, Huntington 62222, The Huntington Li-
brary, San Marino, California, p. 19, detail (EEBO image 20).

Despite the best efforts of the typesetter on this page, the n-macron is virtu-
ally unused in the Recuyell. Precisely what this macron might have meant to
Caxton remains unclear; but the importance of its association with final nasals
is implied by the cutting, in Type 2, of forms of word-final n and m with nasal
macrons. The inadequacy of the original Type 1 fount, and the attempts of
setters to cope with this inadequacy, is perhaps the clearest evidence of a con-
flict between the requirements of an English style and those of the type’s Con-
tinental models.

Type 2 Developments

Type 2, cut sometime before 1476 and brought by Caxton to his Westmin-
ster press, was used for all but liturgical printing until about 1484; in 1479, the
type was recast with additional ligature sorts (including wa, we and wo): this
version (Type 2*) corrected many of the earlier letter-spacing issues.37 In Type
2, there is also a wider array of decorative possibilities. A comparison of the
sorts available for Type 2 letter n (Fig. 8) with the Type 1 options for this letter
(Fig. 5) demonstrates the direction of the change: more choices, not fewer.
There are more options for nasal macrons (both for n and for m); the tailed-d
sort is matched by a tailed-g sort of similar aspect; and a crossed-ll option
appears for straight, as well as looped, varieties of the letter.
These further options in decoration appear to have had an effect upon
spelling. In Type 1 works, there is already a preference to spell present partici-
ples without a final e, but the preference is slight: ing is only about 1.5 times

nasals suggests that it must have borne these English idiosyncrasies – and may further imply that
the setter himself was not familiar their use.
37
For a discussion of the chronology of Caxton’s types and publications, see P. NEEDHAM,
The Printer & the Pardoner: An Unrecorded Indulgence Printed by William Caxton for the
Hospital of St. Mary Rounceval, Charing Cross (Washington, DC, 1986), pp. 83-91.
Labours Lost 159

Fig. 8 Caxton’s Type 2: 135 letter n, as recorded by William Blades.

more common than inge. With Type 2, the availability of a tailed-g sort (con-
sistently and only used in word-final position) may have encouraged this trend
away from final e: in early English Type 2 works,38 ing appears almost three
times as often as inge. The tailed-g form cannot plausibly be accounted an
abbreviation; setters using Type 1 felt no compulsion to provide a final e where
they could place no tail. But the look of the decorated letter seems to have been
an attraction in itself, and may have been enough to influence the spelling of
the setters.
Among the more striking of the additions in Type 2 is a two-compartment
Anglicana a. This addition, and a greater flamboyance in the appearance of w,
contributed to the more English appearance of the type. As with the variously
decorated forms discussed in Type 1, the deployment of these alternatives was
not random. Hellinga in 1982 distinguished a hierarchy in the deployment of
sorts for a: “When Type 2 was used in the Cordiale in Bruges, the double a
was consistently used in every word beginning with a, and the round a in all
other positions, in the middle and at the end of words”.39 This system is analo-
gous to the Type 1 preference set out above for the wider l with looped head
stroke at the beginnings of words; the Type 2 letter a provides is a more visu-
ally striking instance of the same impulse. This system for the a of Type 2,
Hellinga notes, is not found in any

of the twenty books that form Caxton’s output in the first years in Westminster.
When Type 2 was taken to Westminster, Caxton’s new compositors ... were appar-
ently unfamiliar with the function of the two a’s protocol.40

In Westminster, Hellinga is able to distinguish two compositors, one who be-

gins only unusual or interesting words with double-compartment a, and another
who avoids the sort altogether:

38
These early Type 2 works include R. LEFÈVRE, The History of Jason, trans. W. CAXTON
([Westminster, 1477]: STC 15383); The Dictes or Sayengis of the Philosophhres ([Westminster,
1477]: STC 6826); G. CHAUCER, The Canterbury Tales ([Westminster, 1477]: STC 5080).
39
HELLINGA, Caxton in Focus: The Beginning of Printing in England, p. 59.
40
HELLINGA, Caxton in Focus: The Beginning of Printing in England, pp. 59-61.
160 ANYA ADAIR

He might use a few double a’s very rarely and quite indifferently, probably because
they had become mixed with the round a’s in the type-case ... Also, shortage of
type could force him to use double a’s.41

Satoko Tokunaga, examining Caxton’s Westminster productions in more detail

using type-recognition software, suggests a more complex interaction of the
two setters over the course of the printing, and confirms that deployments of
double-compartment a suggest the likelihood of at least two (and possibly
three) workmen being involved in printing with Type 2.42 Compositors, it ap-
pears, could be as individual as scribes in their interpretation and application
of the variety available to them. The inconsistencies in deployment caused by
this variable are increased by the use of Type 2, and competing interpretations
of the same letterforms have arisen: the tasks of commissioning and casting
these subtly different pieces of type, sorting and housing them, and recasting
them as they wore out, begin to feel like labours lost.

What’s a Flourish to a Printer?

In order to create multiple versions of the same letter, printers do both

more and less than scribes. On the one hand, the work of designing, cutting,
striking and casting sorts is multiplied by the requirement of decorative ver-
sions. Then there is the issue of keeping a greater number of sorts organised
and available. On the other hand, since the movement of selecting one letter
above another is about the same, and since a single forme, set once, allows
multiple copies, the labour involved in the selection of decorated letterforms at
the point of setting, even taking into account the large number of compartments
required per case, is relatively light.
But the inconsistencies seen in the deployment of the various decorative
and stylistic options demonstrate a major complication: decisions made at the
point of fount commissioning rely (for their final execution) on the consistency
of setters whose individual interpretations and evaluations of the text’s visual

41
HELLINGA, Caxton in Focus: The Beginning of Printing in England, pp. 61-62.
42
S. TOKUNAGA, “Early English printing and the hands of compositors”, International
Journal of English Studies 5 (2005), pp. 65-76. On this digital method more generally, and for
Tokunaga’s suggested revisions to the classification of the Type 2, see S. TOKUNAGA, “A digital
approach to the history of the book: The case of Caxton”, Poetica (Tokyo) 60 (2003), pp. 149-
160.
Labours Lost 161

features may differ substantially. The inconsistent deployment of crossed th in

Type 1 and the sudden drop in the use of two-compartment a after the arrival
of Type 2 in Westminster demonstrate the ease with which labour undertaken
early in the process could be wasted. Little wonder, then, that Caxton soon
allowed his otiose sorts to expire, and that Wynkyn de Worde was never tempt-
ed to resuscitate them after Caxton’s death, despite the continued use of the
strokes in manuscript well into the sixteenth century. The effect of a scribe’s
distinctive twist of nib becomes in the printer’s workshop a major effort of
forethought and co-operation, in which any deviation from the subtleties of the
shared system threatens the economy of the whole.
Despite their drawbacks, however, the apparently otiose sorts of Types 1
and 2 have been seen to function within a visual system that suggests their
value to the wider dimensions of meaning in print and manuscript culture. In
her examination of capital letters in Type 1, Hellinga found a hierarchy in the
deployment of thick and thin capital letter varieties of R and N in Caxton’s
early verse and prose printing. “Protocol”, she writes, “gave precedence to the
wide and haughty ones, while the spindlier forms played a humbler part in the
elaborate system”.43 This decorum is more subtly reproduced in the lower-case
letters. From the preference for curved-l at the beginnings of words to the in-
creasingly consistent placement of tailed-d at their ends, a careful ordering of
letter varieties is clear in Type 1. Minor differences between the designs for
each letter, closely examined, resolve into a detailed system of signals, through
which the expected shape of words and the divisions between them is indicated
or reinforced. Hellinga’s observation of a tendency to place the spindly form in
the middle of a passage and the more exuberant wide letter at the beginning of
a verse line participates in a pattern replicated at the level of the word. Wider
or more ornate versions of certain letters are associated with word beginnings,
while flourished, pointed or tailed forms come to mark their ends. Mid-word,
decorative features tend to be absent. The mere spacing of the letters (either
between words or within them) becomes only one among many signals of word
division. In this way, the details of letterform and decoration, and their place-
ment within a word, work to guide the reader’s eye through the text.44

43
HELLINGA, Caxton in Focus: The Beginning of Printing in England, pp. 55-57.
44
A hint towards the system argued for here is also provided by Tokunaga: “It is noticeable
that the double compartmented <a> is used [on the word and] when the conjunction is preceded
by a slash (/) or punctuation mark (+). Such graphical apparatus might have possibly been of great
help to readers, allowing them to pause and take breath where there is a short interruption in the
flow of a sentence. The typesetter who made use of the double <a> for the conjunction and,
162 ANYA ADAIR

A still more subtle deployment of word-end signalling arises in compound

words. We have already seen how the spacing that generally divides elements
can be dispensed with in extremis, provided there is some other signal marking
the end of the first word (for instance <hymself> and <woldproue>, Fig. 6, line
2). Here, a tapering final minim on m and a tail on d signal that <hym> and
<wold> are separate words, even where scant following space is provided. (An
analogous space-saving device is the use of punctuation in place of a space
between words, rather than in addition to it; see, for example, the punctuation
of Fig. 2 line 2.) Other word-end forms can separate the elements of a com-
pound from one another: the setting of <notwithstanding>, for example, often
contains a crossed-th sort, a form overwhelmingly associated with the end of
a word. The combination of this mark and the absence of a space hints at some
ambivalence in the word’s form (is it one lexeme, or two?) and helps the eye to
navigate a word that by itself takes up a third of a line. A similar tendency is
seen in the setting of <allway>, in which the crossed-ll is common, though no
space ever separates the first element from the second. In both cases, the visual
signal associated with word-end functions in concert with the absence of a
space to mark a single word compounded of two.
But for all the usefulness of the features at a local level, it is their function
in creating the overall aspect of the printed page that most strongly justifies
their inclusion in the founts of type produced throughout the first decade of
English print. Their appearance suggested an aesthetic not only scribal, but of
a specifically English scribal kind. As guides to smooth reading, they were
surely successful precisely because the words they helped to shape had the
comforting familiarity of English form. But this labour of love (or at least, this
labour to win England’s love) was soon to be lost: simplicity trumped subtlety,
and the brief life of English otiose sorts came to a close.
In 1480, a smaller version of Type 2, Type 4: 95, was adopted. This more
economical type dispensed with many letterform variants (two-compartment a
and looped-d, for example) and greatly reduced the number of crossed and
decorated forms.45 A limited survey of pages from the Cronicles of Englond

combined with a slash or punctuation, was presumably intent on creating pages that might
facilitate the reader’s appreciation of the prose text, by making distinctive use of the double
compartmented <a>” (TOKUNAGA, “Early English printing and the hands of compositors”, p.
156).
45
David R. Carlson has suggested that the relative value of manuscript was increased after
the advent of print precisely because of manuscript’s ability to produce the kinds of ligatured
letterforms that are here disappearing. D.R. CARLSON, “Manuscripts after printing: Affinity,
Labours Lost 163

(1480) suggests that the few decorative features still available in this fount
(tailed-g and tailed-d) were being less and less consistently used: the even
spacing of a line has entirely obliterated the subtle hierarchies in the visual
aesthetic of the word. Tailed-g is rare; tailed-d appears at random. Type 4 was
the last of the founts to preserve the sorts in significant numbers; it was also
the fount in which they became truly otiose.

Conclusion

The cradle of print, suspended between two technologies, was swayed by

the hand of the scribe as well as by that of the printer. “The absence of any
apparent change in product was combined with a complete change in methods
of production, giving rise to the paradoxical combination ... of seeming conti-
nuity with radical change”.46 This paradox resulted, in part, from the efforts of
printers and typesetters to encode old features by new methods. Their troubles,
temporary successes, and final decisions in this endeavour reveal their under-
standing of English manuscript: one that is sensitive to even the smallest of its
features.
A standard explanation of Caxton’s orthographic choices is given by Si-
mon Horobin: “he needed to print books that could be read by the widest possi-
ble audience and so was concerned not to use spellings that would cause diffi-
culties for his readers”.47 This study has demonstrated that mechanical idiosyn-
crasies appear also to have played a significant role in spelling choices and
their standardisation; it also suggests that helping readers to avoid difficulties
in understanding was as much a matter for the careful typesetter as it was for
the writer of copy-text or selector of dialect. An appreciation of the function of
apparently otiose marks to shape words and contribute to an English aesthetic
has been revealed, and something too of the printer’s understanding of what

dissent and display in the texts of Wyatt’s Psalms,” in: Prestige, Authority, and Power, ed.
Felicity Riddy (Woodbridge, 2000), pp. 171-188, at pp. 185-186. The prestige that he identifies
in these forms surely attached also to the otiose marks that were abandoned with them.
46
E.L. EISENSTEIN, The Printing Press as an Agent of Change: Communications and
Cultural Transformations in Early Modern Europe, 2 vols. (Cambridge, 1979), 1, p. 51. On the
subject of Caxton’s print as reproductive of and responsive to prior traditions, see W. KUSKIN,
Symbolic Caxton: Literary Culture and Print Capitalism (Notre Dame, IN, 2008), esp. pp. 3-5.
47
S. HOROBIN, “Mapping the words”, in: The Production of Books in England, 1350-1500,
ed. A. GILLESPIE and D. WAKELIN (Cambridge, 2011), pp. 59-78, at p. 75.
164 ANYA ADAIR

Englishness might mean in terms of page presentation. The marks traced in this
study are deployed meaningfully, and with unexpected semantic as well as
aesthetic weight.
Caxton’s cosmopolitanism, and his position relative to England as a kind
of outsider, produced a type combining new features with old, the foreign with
the familiar. To the technologically hybrid nature of the early printed book,
Caxton adds another level of hybridity: one that combines the high-status aes-
thetic manuscript culture from the Burgundian court and Flemish scriptorium
with the deliberate and pointed application of English-style writing features.
Otiose strokes, as they appeared in the eyes of a few early-adopting printers,
have been decoded for us precisely as they were being encoded into another
form. The complexity of their value and meaning, and their very instability in
manuscript culture, is written by the sorts that encoded them. As the new tech-
nology grew to privilege the simple, the efficient, and the standardised in its
new aesthetic, the schema implied by these sorts became finally illegible, their
meaning and values overrun by a very different technical idiom. For the brief
tenure of Caxton’s Type 1 and Type 2, however, the otiose sort was not otiose
at all.
 Code-Switching, Script-Switching, and
Typeface-Switching in Early Modern English
Manuscript Letters and Printed Tracts

SAMULI KAISLANIEMI

Introduction

lthough we still commonly use italics to mark foreign words or phrases,

A it is not generally known that this convention derives from historical

typographical practices for textual highlighting. These practices in
printed texts in turn partly stem from medieval scribal traditions of associating
scripts with languages. Just as Tudor books printed in black letter use roman
typeface to flag words and phrases in other languages, Early Modern English
manuscripts written in secretary script use italic script for the same purpose. In
analogy of the term code-switching, I call this practice ‘script-switching’ (in
print, ‘typeface-switching’; these terms are defined at more length below).
This chapter is a pilot study of early modern English practices of script-
and typeface-switching, connected to contemporary practices of code-switch-
ing. In the first section, I define my terms and give an overview of script- and
typeface-switching, explaining their role in the early modern European world

......................................................................................................................................
Verbal and Visual Communication in Early English Texts, ed. Matti PEIKOLA, Aleksi
MÄKILÄHDE, Hanna SALMI, Mari-Liisa VARILA, and Janne SKAFFARI, Utrecht Studies in Medieval
Literacy, 37 (Turnhout: Brepols, 2017), pp. 165-200.

DOI 10.1484/M.USML-EB.5.114135
166 SAMULI KAISLANIEMI

of ‘national’ scripts and typefaces. In the next main section, I present two small
case studies of the practice in early modern English manuscripts and printed
tracts, and compare the results to practices of code-switching in the same texts.
I finish the section with a discussion comparing my findings. In the third main
section, I discuss problems of linking script to language in code-switching in
manuscripts and have a closer look at the various functions of typeface-switch-
ing, and conclude in the final section by considering the broader implications
of the results of this study.

Code-Switching and Script- and Typeface-Switching

What is Meant by Code-Switching in This Study

This study embraces the “common-sense approach” to code-switching

described by Penelope Gardner-Chloros.1 In this approach, the term ‘code-
switching’ is used as a non-restrictive umbrella for all occurrences of foreign
words and passages within a text. Such a broad definition of code-switching
may be problematic for more nuanced studies, but my usage follows previous
work and conforms with the encoding schemes of the corpora studied (see the
section ‘Methodological note: how to count script- and code-switches’ below).2
1
P. GARDNER-CHLOROS, Code-Switching (Cambridge, 2009), pp. 7-9.
2
This study builds on S. KAISLANIEMI, “Encountering and appropriating the Other: East
India Company merchants and foreign terminology”, in: The Language of Daily Life in England
1450-1800, ed. A. NURMI, M. NEVALA, and M. PALANDER-COLLIN (Amsterdam, 2009: Prag-
matics and Beyond New Series 183), pp. 219-251; S. KAISLANIEMI, “The early English East India
Company as a community of practice: Evidence of multilingualism”, in: Merchants of Innova-
tion: The Language of Traders, ed. E.-M. WAGNER, B. BEINHOFF, and B. OUTHWAITE (Berlin,
2017: Studies in Language Change 15), pp. 132-157; A. NURMI and P. PAHTA, “Social
stratification and patterns of code-switching in early English letters”, Multilingua 23 (2004), pp.
417-456; P. PAHTA and A. NURMI, “Code-switching in the Helsinki Corpus: A thousand years of
multilingual practices”, in: Medieval English and its Heritage: Structure, Meaning and
Mechanisms of Change, ed. N. RITT, H. SCHENDL, C. DALTON-PUFFER, and D. KASTOVSKY
(Frankfurt, 2006), pp. 203-220; P. PAHTA and A. NURMI, “Multilingual discourse in the domain
of religion in medieval and early modern England: A corpus approach to research on historical
code-switching”, in: Code-Switching in Early English, ed. H. SCHENDL and L. WRIGHT (Berlin
and Boston, 2011: Topics in English Linguistics 76), pp. 219-251.
For recent work on historical code-switching, see a.o. Bilingualism in Ancient Society:
Language Contact and the Written Text, ed. J.N. ADAMS, M. JANSE, and S. SWAIN (Oxford,
2002), Oxford Scholarship Online <www.oxfordscholarship.com>; Code-Switching in Early
English, ed. SCHENDL and WRIGHT; Multilingual Practices in Language History: New Per-
Code-Switching, Script-Switching, and Typeface-Switching 167

And as one of the questions posed in this study is whether visual highlighting
can tell us something about multilingual writing in general and code-switching
practices in particular, an inclusive approach was deemed better for data col-
lection purposes.

Scripts, Typefaces and Languages in Early Modern England

In palaeography, the term ‘script’ is (confusingly) used for both ‘writing

systems’ and ‘types of writing’: examples of the former would be the Latin
alphabet and the Devaganari alphasyllabary; examples of the latter would be
the medieval Insular and Caroline Minuscule scripts.3 This study focusses on
this latter kind of distinction, between styles or types of writing, and ‘script’ is
used in this restricted sense.
Although a number of different scripts were in fact in use in early modern
England, only the most common scripts need discussing here, as many or most
of the scripts had restricted spheres of usage. The primary cursive script used
in Elizabethan England is known as secretary script.4 Despite its name, it was
actually a blend of two different late-medieval Gothic scripts, secretary and
Anglicana, and might better be called early modern English cursive script, for
they had evolved into a distinctly English script which could be written at
speed. But by 1600, secretary was beginning to lose ground to the italic script,
introduced into England by humanist scholars in the early sixteenth century,

spectives (Papers from the symposium Historical Code-switching: The Next Step, Tampere, 11-13
June 2014), ed. P. PAHTA, J. SKAFFARI, and L. WRIGHT (in preparation).
3
In P. BEAL, A Dictionary of English Manuscript Terminology (Oxford, 2008), s.v., ‘script’
is defined as “an identifiable type or system of writing, with all its distinctive forms and
characteristics, usually belonging to a particular historical period or location”. The difference to
‘hand’ is captured in Malcolm Parkes’ now classic summary: “A script is the model which the
scribe has in his mind’s eye when he writes, whereas a hand is what he actually puts down on the
page” (M.B. PARKES, English Cursive Book Hands, 1250-1500 (Oxford, 1969), p. XXVI; cf. BEAL,
Dictionary, s.v. ‘handwriting’). The terms ‘script’ and ‘hand’ are often used interchangeably.
4
For some basic palaeographical works, see S.A. TANNENBAUM, The Handwriting of the
Renaissance (London and New York, 1930; 1931), and G.E. DAWSON and L. KENNEDY-SKIPTON,
Elizabethan Handwriting 1500-1650: A Guide to the Reading of Documents and Manuscripts,
2nd edn. (Chichester, 1981); a recent article-length overview is G. IOPPOLO, “Early modern
handwriting”, in: A New Companion to English Renaissance Literature and Culture, ed. M.
HATTAWAY, 2 vols. (Chichester, 2010), 1, pp. 177-189. A concise description of ‘all’ scripts in
use in early modern England can be found in H. JENKINSON, “Elizabethan handwritings: A
preliminary sketch”, The Library, 4th ser., 3.1 (1922), pp. 1-50.
168 SAMULI KAISLANIEMI

which was clearer to read and write.5 ‘Digraphia’ became the norm, as edu-
cated people – gentry and merchants alike, but in particular those who made
their living through writing – learned to write in both scripts.6 This somewhat
5
This process is one of the central topics in J. GOLDBERG, Writing Matter: From the Hands
of the English Renaissance (Stanford, CA, 1990). Roman (with italic) was considered “the easiest
hand that is written with Pen”, which contemporaries (in)famously considered to mean it was easy
enough for women to learn (e.g. M. BILLINGSLEY, A Newe booke of copies containing divers
sortes of sundry hands, as the English and French secretarie, and bastard secretarie, Italian,
Roman, chancery, and court hands (London, 1620: STC 33638; Early English Books Online
(EEBO), Chadwyck-Healey, <eebo.chadwyck.com>: Harvard University Library), pp. C2r-v and
C4r; and J. DAVIES, The writing schoolemaster, or, The anatomie of faire writing wherein is
exactlie expressed each seuerall character, together with other rules & documents, coincident
to the art of faire and speedy writing, 6th enlarged edn. (London, 1631: STC 6344.5; EEBO:
Harvard University Library), p. Av, but Davies praises women’s ability on B2r-v. Cf. GOLDBERG,
Writing Matter, pp. 138-139).
Studies of typefaces and legibility include B. REIMER, B. MEHLER, and J.F. COUGHLIN,
“An evaluation of typeface design in a text-rich automotive user interface”, Massachusetts
Institute of Technology AgeLab white paper 2012-12 (2012), available from <agelab.mit.edu>;
and J. DOBRES, B. REIMER, L. PARIKHAL, E. WEAN, and N. CHAHINE, “The incredible shrinking
letter: How font size affects the legibility of text viewed in brief glances”, paper presented at the
8th International Driving Symposium on Human Factors in Driver Assessment, Training and
Vehicle Design, Salt Lake City, UT, 22-25 June 2015, available from <agelab.mit.edu>; see also
K. THOMAS, “The meaning of literacy in early modern England”, in: The Written Word: Literacy
in Transition: Wolfson College Lectures 1985, ed. G. BAUMANN (Oxford, 1986), pp. 97-131, at
pp. 99-100.
Although differentiated by contemporaries and palaeographers alike, this discussion
subsumes roman and italic under the latter name, since the focus is on the conceptual difference
between types of writing (Gothic secretary and humanist italic scripts). Naturally there was also
great variation within these scripts.
6
Since early modern English writing literacy was usually ‘biliteracy’ – schoolboys started
by learning Latin – it is difficult to imagine they would not have also been taught the scripts
appropriate for writing in each language (but see THOMAS, “The meaning of literacy”). Some
contemporary authors advocated the teaching of both scripts, e.g. R. MULCASTER, The first part
of the elementarie vvhich entreateth chefelie of the right writing of our English tung (London,
1582: STC 18250; EEBO: Huntington Library), pp. 56-57; see also H.C. SCHULZ, “The teaching
of handwriting in Tudor and Stuart times”, Huntington Library Quarterly 6.4 (1943), pp. 381-
425, at p. 403; A.G. PETTI, English Literary Hands from Chaucer to Dryden (London, 1977), p.
19; and GOLDBERG, Writing Matter, pp. 50-54.
I am not aware of an established word to describe situations where writing literacy in
practice means skill in two (or more) distinct scripts (types of writing). My use of the term
‘digraphia’ differs from sociolinguistics, where it has been used to describe the phenomenon of
a single language written in multiple scripts, and specifically in multiple writing systems (see S.
GRIVELET, “Introduction”, International Journal of the Sociology of Language 150 (2001), pp.
1-10, and P. UNSETH, “The sociolinguistics of script choice: An introduction”, International
Journal of the Sociology of Language 192 (2008), pp. 1-4).
A sophisticated typology of the use of multiple scripts for the same language was
Code-Switching, Script-Switching, and Typeface-Switching 169

inevitably led to the proliferation of mixed hands, containing features from

both secretary and italic scripts, and the eventual development of the cursive
Round hand towards the end of the seventeenth century.7 The key issue here is
that in Elizabethan and Jacobean England, secretary and italic scripts were
conceptually distinct entities, reserved for different purposes (in theory any-
way), and at the very least carried different connotations. Secretary script was
the primary script used by merchants, and it carried a sense of being the appro-
priate hand for business and general correspondence. In turn, italic script was
increasingly the preferred hand of the educated and the social elites.8
These scripts were also associated with certain languages: (English) secre-
tary with English, and italic with Italian and Latin. The association of ‘writing
systems’ with languages is familiar to us for instance from signs at airports,
where the English word ‘Arrivals’, written in the Latin alphabet, may be seen
next to the Chinese term , in logographic hanzi, or the Russian word
Ïðèë¸ò, in the Cyrillic alphabet. Less familiar today is the practice of different
languages being written in and associated with specific different ‘types of writ-
ing’ – but this was the case in the early modern period. Across Europe, there
were distinct ‘national’ scripts, such as French secretary and Spanish italic, and
contemporary handwriting copybooks contain examples of various national
scripts.9 Since all scripts were used to write essentially the same alphabet, it

published after this article was submitted: Biscriptality: A Sociolinguistic Typology, ed. D.
BUNÈIÆ, S.L. LIPPERT, and A. RABUS (Heidelberg, 2016). The practice I call digraphia is there
divided into three types of ‘glyphic variation’: ‘diglyphia’, ‘glyphic pluricentricity’, and ‘bi-
glyphism’ (D. BUNÈIÆ, “A heuristic model for typology”, in: Biscriptality, pp. 51-72, esp. pp.
63-68, and the table of types of biscriptal situations on p. 67). Most relevant to this article is a
long discussion of ‘biglyphism’ in German: J. SPITZMÜLLER and D. BUNÈIÆ, “German [bi-
glyphism]: Blackletter and roman”, in: Biscriptality, pp. 282-300. Biscriptality is a definitive
work for the sociolinguistics of writing, but not exhaustive, as for instance it does not discuss
script-switching as a phenomenon.
7
Mixed hands are described as an inevitable result of digraphia by PARKES, English
Cursive Book Hands, pp. XXIV-XXV.
8
For instance, BILLINGSLEY, Newe booke of copies, pp. C3v and C4r, writes that the
secretary “is the onely vsuall hand of England, for dispatching of all manner or businesses”, and
that the roman is a “hand of great account, and of much vse in this Realme, especially in the
Vniuersities”. Cf. DAVIES, The writing schoolemaster, pp. Bv-B3r, on the variety of hands used
in England and their purposes.
9
E.g. E.B., A nevv booke, containing all sorts of hands vsually written at this day in
Christendome as the English and French Secretary, the Roman, Italian, French, Spanish, high
and low Dutch, court and chancerie hands: with examples of each of them in their proper tongue
and letter (London, 1611: STC 3361.7; EEBO: Folger Shakespeare Library); and BILLINGSLEY,
Newe booke of copies; cf. D.P. BECKER, The Practice of Letters: The Hofer Collection of Writing
170 SAMULI KAISLANIEMI

was perfectly possible to use any script for any language, but it is nonetheless
rare to find examples of documents written in the ‘wrong’ script – for instance,
an English text in a French or Spanish script – and generally one can identify
the language a text is written in from the script used.10 The identification of
scripts with languages means that language learning involved learning the
appropriate script, and therefore the ability to write a foreign script implied
knowledge of the language it indicated and vice versa; and conversely, national
scripts were not readily legible to those not proficient in the language – scripts
could be very different, much as if they had been different writing systems
altogether.11 A partial if notable exception was italic, which became the pre-
ferred script of scholars and secretaries across Europe, who used it to write any
and all languages, Latin as well as the vernaculars. In digraphic England italic

Manuals 1514-1800 (Cambridge, MA, 1997).

The existence of national scripts is attested by the archival record, but also reflected by
the existence of language-specific palaeography manuals. See also S. MORISON, “The develop-
ment of hand-writing: An outline”, in: ID., Selected Essays on the History of Letter-Forms in
Manuscript and Print, ed. D. MCKITTERICK, 2 vols. (Cambridge, 1981), 1, pp. 161-176, esp. pp.
165-169 (first publ. as introduction to A. HEAL, The English Writing-Masters and Their Copy-
books, 1570-1800 (1931), pp. XXIII-XL); and M.P. BROWN, A Guide to Western Historical Scripts
from Antiquity to 1600 (London, 1993), esp. pp. 32-47.
10
Writing a language in the script of another is commonly called ‘allography’, or some-
times ‘aljamia(do)’. See Scripts Beyond Borders: A Survey of Allographic Traditions in the Euro-
Mediterranean World, ed. J. DEN HEIJER, T. PATARIDZE, and A.B. SCHMIDT (Louvain-la-Neuve,
2014: Publications de l’Institut Orientaliste de Louvain 62), and the discussion in J. DEN HEIJER
and A.B. SCHMIDT, “Scripts beyond borders: Allographic traditions and their social, cultural and
philological aspects: An analytical introduction”, in: Scripts Beyond Borders, pp. 1-64, esp. pp.
2-3.
11
It is difficult to find contemporary discussions of this, although oblique references occur,
such as, “fewe mens writ, (except those who write the universall ROMANE Character) prooue anie
more profitable for their use, than their natiue Language, when they resorte to anie forraine
Nation” (D. BROWNE, The New Invention, intitvled, calligraphia: or, the arte of faire writing (St
Andrews, 1622: STC 3905; EEBO: British Library), p. [¶¶7r]), and, “out of this kingdome there is
no use of our secretary hand” (DAVIES, The writing schoolemaster, p. B2v).
Examples of Englishmen writing foreign languages in the appropriate script are common.
J. GALLAGHER, “The Italian London of John North: Cultural contact and linguistic encounter in
early modern England”, Renaissance Quarterly 70.1 (2017), pp. 88-131, gives examples of
sixteenth-century English travellers to Italy keeping journals in Italian in italic script. In contrast,
writers who used italic to write English did not (need to) change their hand when changing
language: on his grand tour 1609-1610, William Cecil, Lord Cranborne kept a journal in French
in the same italic hand he used for English (MS Hatfield, Hatfield House Archives, Cecil Papers
(CP) 317/1; Cecil Papers, Hatfield House Archives and ProQuest, <cecilpapers.chadwyck.
co.uk>).
Code-Switching, Script-Switching, and Typeface-Switching 171

retained a connection to Latin, but it also became the default script used for
visual highlighting in texts written in secretary script.12
Developments in the realm of print were similar but different.13 The first
typefaces had been developed on the basis of local scripts, which at that point
in time were Gothic. Roman type was developed as early as the 1460s, fol-
lowed by italic in 1501, and by the time of Henry VIII, a basic trinity of black
letter, roman and italic had been established.14 Roman began to replace black
letter as the default typeface by the last decades of the sixteenth century, and
the process was broadly completed by about the mid-seventeenth century,
although black letter remained in use as a secondary typeface.15
This was a categorical change: although roman and black letter are repre-
sentations of the same alphabet (the primary letters are the same, even if the
graph inventories differ between languages, i.e. not all languages use the same
set of characters, for instance including or omitting <j>, <ä>, or <ç>), like
secretary and italic script, they are visually quite distinct, and knowledge of
one does not impart the ability to read or write the other.16 And literacy in a
typeface was linked to literacy in a language: roman typeface was originally
used to print Latin texts, with black letter used for vernacular works. The same
basic associations as developed in the world of scripts were repeated in print;
the long association of black letter with the vernacular is reflected in how the
typeface later came to be called ‘English’.17

12
R.B. MCKERROW, An Introduction to Bibliography for Literary Students (Oxford, 1928),
pp. 251-252.
13
This paragraph is based on MCKERROW, Introduction, pp. 288-298; S.H. STEINBERG, Five
Hundred Years of Printing, 2nd edn. (Harmondsworth, 1961), pp. 27-42, 176; and M. BLAND,
“The appearance of the text in early modern England”, Text 11 (1998), pp. 91-154.
14
Called “the three orders of Print Letters” by J. MOXON, Regulae Trium Ordinum literarum
typographicarum: or the Rules of the three orders of Print Letters: viz. The Roman, Italick,
English, Capitals and Small (London, 1676: WING M3019; EEBO: Huntington Library); this triad
survives in present-day typography as plain text, bold and italics.
15
Diachronically speaking, this narrative, if simplistic, is true. For a critique of the process,
see J.A. DANE, Out of Sorts: On Typography and Print Culture (Philadelphia, PA, 2011), pp. 57-
71; for a discussion of Gothic scripts and typefaces, see G. NOORDZIJ, “Broken scripts and the
classification of typefaces”, Journal of Typographic Research [Visible Language] 4.3 (1970), pp.
213-240; for developments in German, see SPITZMÜLLER and BUNÈIÆ, “German [biglyphism]”,
pp. 282-300.
16
For literacy and printed typefaces, see C.C. MISH, “Black letter as a social discriminant
in the seventeenth century”, Proceedings of the Modern Language Association 68.3 (1953), pp.
627-630; and THOMAS, “The meaning of literacy”, esp. pp. 99-100. See also SPITZMÜLLER and
BUNÈIÆ, “German [biglyphism]”, pp. 287-288.
17
MOXON, Regulae, t.p.
172 SAMULI KAISLANIEMI

From the start, there was also the desire and attempt to differentiate lan-
guages by using different typefaces – indeed, the practice has been called “a
well-established international and trans-lingual visual coding system”.18 The
idea of ‘national typefaces’ to match national scripts did emerge and such
founts were created, but these script types – what came to be called civilité
types – ultimately failed to become generic typefaces.19 Instead of language-
specific typefaces, printers relied on the basic trinity of black letter, roman and
italic, and accordingly the associations between typefaces and languages were
not as strong as those between scripts and languages.20 In multilingual texts, it
was normal to use black letter for the vernacular (for example English), and
roman and italic for all foreign languages (that shared the same alphabet) –
even when more languages were involved, the number of typefaces was usually
kept at three, alternating the typefaces to retain contrast between languages.21

18
G. ARMSTRONG, “Coding continental: Information design in sixteenth-century English
vernacular language manuals and translations”, Renaissance Studies 29.1 (2015), pp. 78-102, at
p. 92. Although quantitative surveys of typeface-switching are rare, book historians regularly
address the links between typefaces and languages: i.a. MCKERROW, Introduction; BLAND, “The
appearance of the text”; V. HOTCHKISS and F.C. ROBINSON, English in Print from Caxton to
Shakespeare to Milton (Urbana, IL, 2008); DANE, Out of Sorts.
19
They did survive in various niches well into the eighteenth century, however. See H.
JENKINSON, “English current writing and early printing”, The Library: Transactions of the
Bibliographical Society 13.1 (1913), pp. 273-295; MCKERROW, Introduction, pp. 296-297;
STEINBERG, Five Hundred Years of Printing, pp. 38-40; and S. MORISON, “On script types”, in:
ID., Selected Essays, 1, pp. 47-80 (first publ. in The Fleuron 4 (1925), pp. 1-39). The definitive
study of civilité types is H. CARTER and H.D.L. VERVLIET, Civilité Types (Oxford, 1966).
20
ARMSTRONG, “Coding continental”, p. 102; cf. C. SUHR, Publishing for the Masses: Early
Modern English Witchcraft Pamphlets (Helsinki, 2011: Mémoires de la Société Néophilologique
de Helsinki 83), p. 68.
21
A great illustration of this are polyglot glossaries and dialogue books: in N. DE BERLE-
MONT, Colloquia et dictionariolum octo linguarum, Latinae, Gallicae, Belgicae, Teutonicae,
Hispanicae, Italicae, Anglicae, et Portvgallicae (Delft, 1613: STC 1431.22; EEBO: Huntington
Library), an eight-language dialogic language manual, the languages and their respective type-
faces – presented in order across each opening – are Latin (roman), French (italic), Flemish (i.e.
Dutch, black letter), German (italic), Spanish (roman), Italian (italic), English (black letter), and
Portuguese (italic). Yet there are always exceptions, for there was (considerable) variation within
each of the ‘trinity’ typefaces (MCKERROW, Introduction, pp. 288-296), allowing for works such
as G. DE MONTENAY, A booke of armes, or remembrance wherein ar one hundered godly
emblemata, in péeces if brasse very fine graven, and adorned pleasant to bé séen; first by the
noble, and industrious minde Georgetta de Montenay, invented and only in the Frenchtongve
[sic] elabourated; bot [sic] now, in severall langvages, as; Latin, Spanish, Italian, Highdutch,
English, and Lovedutch, meetre or verse wys, of the same manner declared, and augmented
(Frankfurt, 1619: STC 18046; EEBO: British Library), which prints emblem texts in seven lan-
guages in four typefaces: French (roman), Latin (italic), Spanish (roman), Italian (italic), German
Code-Switching, Script-Switching, and Typeface-Switching 173

Summing up this section, I have shown and argued how in early modern
Europe different languages were written with different scripts, which were
conceptually and visually distinct. In turn, different scripts were associated
with different spheres of use, and with different languages. In print, the situa-
tion was broadly the same, except that instead of national scripts there were
only three primary typefaces in use. In the next section, I discuss methods of
visual differentiation in more detail.

Defining Script- and Typeface-Switching

Although this chapter sets out to look at script- and typeface-switching as

possible markers of code-switching, this should by no means be taken to sug-
gest that there are no other ways of flagging code-switching in texts. Moreover,
flagging code-switching is only one of many functions of visual highlighting in
texts.22
Visual highlighting can be accomplished in many different ways, of which
changing the type of writing (script or typeface) is only one. Others include
changing text size, colour, position, text direction, or aspects of the mise-en-
page such as paragraphing, indentation, or spacing, or using lines to underline
or surround the desired words of passages. Further possible methods of high-
lighting include capitalisation and enlarged initials, the use of abbreviations,
brevigraphs, and special characters, and the use of punctuation (this is not an
exhaustive list). The term ‘visual highlighting’ (or ‘visual flagging’) serves as
an umbrella term for all of these different things.23 Based on Grant-

(black letter), English (roman), and Dutch (a smaller, different black letter).
22
For visual aspects of code-switching in present-day texts, see Language Mixing and
Code-Switching in Writing: Approaches to Mixed-Language Written Discourse, ed. M. SEBBA,
S. MAHOOTIAN, and C. JONSSON (New York and London, 2012); and M. SEBBA, “Multilingualism
in written discourse: An approach to the analysis of multilingual texts”, International Journal of
Bilingualism 17.1 (2013), pp. 97-118. For historical texts, see T.W. MACHAN, “The visual prag-
matics of code-switching in late Middle English literature”, in: Code-Switching in Early English,
pp. 303-333; and Visual Text, ed. J. KENDALL, M. PORTELA, and G. WHITE (= European Journal
of English Studies 17.1 (2013)).
23
Linguistic work on visual aspects of historical texts include R. CARROLL, M. PEIKOLA, H.
SALMI, M.-L.VARILA, J. SKAFFARI, and R. HILTUNEN, “Pragmatics on the page: Visual text in late
medieval English books”, European Journal of English Studies 17.1 (2013), pp. 54-71; Principles
and Practices for the Digital Editing and Annotation of Diachronic Data, ed. A. MEURMAN-
SOLIN and J. TYRKKÖ (Helsinki, 2013: Studies in Variation, Contacts and Change in English 14),
<www.helsinki.fi/varieng/series/volumes/14>; Manuscript Studies and Codicology: Theory and
174 SAMULI KAISLANIEMI

Russell’s term ‘typographic flagging’ – by which I understand her to mean

most of the methods listed above – we can coin the term ‘graphic flagging’ (or
‘graphic highlighting’), using the specific sense of ‘graph’ meaning ‘character
/ letter’, to mean ‘visual highlighting based on changing the graphs’.24 But
where graphic flagging, then, would include for instance the use of majuscules
or changing text size, the kind of graphic flagging where the visual highlighting
is achieved through changing the type of writing (the script or typeface) is what
is meant by ‘script-switching’ and ‘typeface-switching’ in this study.25

Practice, ed. M. KYTÖ and M. PEIKOLA (= Studia Neophilologica 86 Supplement 1 (2014)); and
ARMSTRONG, “Coding continental”. See also the Visible Language journal, esp. the two special
issues on bilingual writing, Bi-Graphic Differences: Languages in Con(tact)(flict), ed. R.
HODGSON and R. SARKONAK (= Visible Language 21.1 (1987)); and Writing ... in Stereo:
Bilingualism in the Text, ed. R. HODGSON and R. SARKONAK (= Visible Language 27.1-2 (1993)).
24
P. GRANT-RUSSELL, “The influence of French on Quebec English: Motivation for lexical
borrowing and integration of loanwords”, LACUS Forum 25 (1998), pp. 473-486, at pp. 479-480;
BEAL, Dictionary, s.v. ‘graph’; OED, s.v. ‘graph’, n.3. Given that graphic flagging is by definition
visually prominent, it has no more escaped the notice of scholars than that of contemporaries (e.g.
MOXON, Regulae, pp. 10-11). Outside the field of typography, graphic flagging has attracted
interest from linguists, both in modern printed texts (R. WALLER, The Typographic Contribution
to Language: Towards a Model of Typographic Genres and Their Underlying Structures (un-
published doctoral thesis, University of Reading, 1987), available from <www.robwaller.org>;
E. MCATEER, “Typeface emphasis and information focus in written language”, Applied Cognitive
Psychology 6 (1992), pp. 345-359; S. WALKER, Typography & Language in Everyday Life: Pre-
scriptions and Practices (Harlow, 2001); D. CRYSTAL, Dictionary of Linguistics and Phonetics,
6th edn. (Oxford, 2009), s.v. ‘graphetics’, ‘graphology’), and also in early modern print and
manuscript (C. MOORE, Quoting Speech in Early English (Cambridge, 2011)).
25
‘Script-switching’ appears not to have been investigated in any detail in previous work.
The term is used in connection to language-switching by A. DANYLENKO, “The Uniate ‘Leksikon’
of 1722: A case of regional relapse from script- to language-switching”, Slavia Orientalis 56.4
(2007), pp. 543-557; GARDNER-CHLOROS, Code-switching, at pp. 67-68, 114-116; and G.Y.
LEUNG, “Hong Kong university students’ language use in blogs: Tensions between creativity and
prescriptivism (‘omg my grammar is rubbish’)”, in Sinographic Languages: The Past, Present,
and Future of Script Reform, ed. V.H. MAIR (= Sino-Platonic Papers 189 (2009)), pp. 30-62,
<sino-platonic.org>. In K. KATAOKA, “Affect and letter-writing: Unconventional conventions in
casual writing by young Japanese women”, Language in Society 26.1 (1997), pp. 103-136, the
term ‘script-switching’ is applied to both scribal and typographical highlighting. Other manu-
script scholars have used terms such as ‘script-mixing’ (E. OKASHA, “Script-mixing in Anglo-
Saxon inscriptions”, in: Writing and Texts in Anglo-Saxon England, ed. A.R. RUMBLE (Wood-
bridge, 2006: Publications of the Manchester Centre for Anglo-Saxon Studies 5), pp. 62-70) and
‘script alternation’ (P.S. ANGERMEYER, “Bilingualism meets digraphia: Script alternation and
hybridity in Russian-American writing and beyond”, in: Language Mixing and Code-Switching
in Writing, pp. 255-272), and print scholars have discussed ‘font shifts’ (A. WILSON, “Speech,
writing and discourse type”, in: English Language. Description, Variation and Context, ed. J.
CULPEPER, F. KATAMBA, P. KERSWILL, R. WODAK, and T. MCENERY (Basingstoke, 2009), pp.
Code-Switching, Script-Switching, and Typeface-Switching 175

Fig. 1 Script-switching in an abstract of a letter dated 28 January 1608 (TNA SP 94/15

f. 9r (detail)).26

If the methods of visual highlighting in texts are various, so are the pur-
poses to which it is put, from text ordering to flagging semantically important
words and passages – “Whatsoever words the Author laies any kind of stress
or force upon, these he either writes in a different character, or else prefixes a
Capital before them, or both”, wrote Elisha Coles.27 Some other functions of
script- and typeface-switching are given here below and in the section ‘Discus-

425-438) and ‘type switches’ (SUHR, Publishing for the Masses). All of these authors either look
at switching between different writing systems, rather than between types of writing, or treat
language briefly and as one of the many functions of visual highlighting.
26
Images of TNA manuscripts published with the permission of the UK National Archives.
Fig. 5 taken from State Papers Online, 1509-1714, published with permission of Gale/Cengage
Learning.
27
E. COLES, The compleat English schoolmaster or, the most natural and easie method of
spelling English (London, 1674: WING C5067A; EEBO: Bodleian Library), p. 107 [115].
176 SAMULI KAISLANIEMI

sion and further thoughts’, but Fig. 1 illustrates the use of script-switching for
several purposes.
The document reproduced in Fig. 1 is an abstract of a letter of news, with
marginal explanatory notes. As is the usual case in early modern English manu-
scripts, the base script is English secretary, and the script switched to is italic.
At the top and in the left margin, script-switching is used for textual organisa-
tion and metatexts: the header and marginal notes are in italic. Within the run-
ning text, script-switches are used to highlight names (l. 4 “Leoneret”, “Bast-
able”, l. 9 “Aspitia”; cf. Fig. 6) and important content words (l. 12 “Armatha”).
And finally, script-switching is used to mark change in language (l. 8 “Corigi-
dor”).28
Fig. 1 also shows how the direction of script-switching is not always to
italic. In the otherwise italic header the roman numerals are in secretary script
– although note how the foreign items, “Januarij, stilo nouo”, are not script-
switched. In early modern English-language texts, code-switches as a rule
would never be flagged by script-switching from italic to English secretary, but
it was possible to switch from italic to a more formal set italic script.29 This can
be difficult to spot, especially where the base hand is very neat, but script-
switching from cursive to formal italic can also be found in drafts, which
shows that the phenomenon of script-switching is not restricted to presentation
documents and neat copies.
Just as script-switching was used as a generic means of visual highlighting,
similar practices evolved in print. In black letter texts, words and passages in
foreign languages would be indicated using roman or italic type; and in texts
printed in roman typeface, italic was the primary differentiation type (see Fig.
4).30 Other purposes to which typeface-switching was put included flagging
proper names (people, places, titles, sometimes festivals, weekdays, and
months), indicating quotations (often from the Bible or other authorities, and
thus in Latin, or another foreign language; sometimes including reported
28
The scribe has also used other types of visual highlighting. For instance, the first words
in the header and the first paragraph have enlarged initials, and annotated words are underlined
and linked to their marginal glosses with superscript letters.
29
For an example of script-switching between two italic hands (although in a Latin text),
see PETTI, English Literary Hands, pp. 114-115. The claim that script-switching from italic to
secretary was avoided is of course difficult to prove without quantitative evidence, and it is based
rather on personal experience over ten years of working in British archives. I am more than
willing to believe that examples do exist.
30
MCKERROW, Introduction, pp. 296-298; this practice arose as soon as roman and italic
typefaces were introduced.
Code-Switching, Script-Switching, and Typeface-Switching 177

speech), and marking text structure (parts of the running text printed in differ-
ent typeface – such as dedications, or epistles to the reader, but also marginalia
and notes – as well as headers, section headings, and the like).31 More nuanced
uses included flagging ‘style switches’ or ‘register switches’, such as learned
words, but applicable to all jargon; and topical words, i.e. words particularly
relevant for the contents or message of the text.32

Two Case Studies of Script- / Typeface-Switching and Code-Switching

Having described and defined ‘code-switching’ and ‘script-’ / ‘typeface-

switching’ as they are used in this study, the next step is to investigate their
appearance in the chosen materials. The aim of the following case studies is to
look at the co-occurrence of code-switching and graphic flagging in early mod-
ern English manuscript letters and printed tracts, drawing out any similarities
and differences in practices. This study is but a pilot, scratching at the surface
of a broad topic, but as we have barely begun to look at script- and typeface-
switching, and given the lack of material suited for quantitative surveys, empir-
ical studies of seemingly disparate materials are not only unavoidable, but

31
See esp. SUHR, Publishing for the Masses, pp. 71-94; see also MCKERROW, Introduction,
pp. 292, 297-298; BLAND, “The appearance of the text”, pp. 97-100; C. CLARIDGE, “Life is ruled
and governed by opinion”: The Lampeter Corpus of Early Modern English Tracts Manual of
Information (1999; 2003), available at <www.tu-chemnitz.de/phil/english/sections/linguist/
real/independent/lampeter/manual/pages/contents.html> (see also <clu.uni.no/icame/manuals/
LAMPETER/LAMPHOME.HTM>), pp. 24-25; WILSON, “Speech, writing and discourse type”,
pp. 431-432; C. CLARIDGE, “From page to screen: The relevance of encoded visual features in the
Lampeter Corpus”, in: Principles and Practices for the Digital Editing and Annotation of
Diachronic Data, <www.helsinki.fi/varieng/series/volumes/14/claridge/>, sect. 3; and D.
CRYSTAL, Making a Point: The Pernickety Story of English Punctuation (London, 2015), pp.
318-326. In the index of the Chicago Manual of Style, 14th edn. (Chicago and London, 1993),
‘italics’ has 41 entries, and ‘roman type’ 13.
32
SUHR, Publishing for the Masses, pp. 78-90. The function of a typeface-switch can be
opaque, but this may be due to modern readers not having deciphered the intent of the author or
typesetter yet – although cases where typesetters run out of the base type are not unheard of
(SUHR, Publishing for the Masses, pp. 87-88; cf. some of the English texts in DE MONTENAY, A
booke of armes, e.g. pp. 60, 64, 152, 156). An even more nuanced practice is word-internal
typeface- or script-switching. Word-initial capitals can be borrowed from a different typeface, and
morphologically embedded borrowings can be script-switched (to italic) for the body, and back
(to the vernacular hand) for the inflected ending (see SPITZMÜLLER and BUNÈIÆ, “German
[biglyphism]”, p. 286-287).
178 SAMULI KAISLANIEMI

necessary. Only once we have baseline data can we start to investigate the
nuances of early modern English practices of script- and typeface-switching.33

Methodological Note: How to Count Script- and Code-Switches

Perhaps the primary reason why script- / typeface-switching remains un-

derstudied is because it is so very common that it takes inordinate effort to go
through the data to find interesting correlations. Manual quantitative investiga-
tion of both script- and code-switching is tedious and time-consuming. For a
text of any size, this soon becomes impractical. If we are to acquire enough
data to make generalisations, reading through the texts will not serve. Instead,
the material must be computer-readable, and the switches encoded in it. This
requirement guided the choice of materials used for this study.
However, the existence of suitable material cannot be taken as a given:
tagging code-switches and script- / typeface-switches is of course a laborious
task, and one which has not been performed on historical corpora as a default.
This restricted the scope of my study. For manuscript material, I edited the
letters of Richard Cocks from manuscript myself, in the process of which I
inserted tags to encode information about script and language. In the case of
printed material, the fact that both typeface and language are tagged in the

33
The use of multiple writing systems (e.g. the Roman alphabet and Chinese hanzi) is very
common in multilingual environments, and has attracted a fair bit of scholarly attention on
historical periods: e.g. ancient inscriptions in Italy (M. LEIWO, “From contact to mixture: Bilin-
gual inscriptions from Italy”, in: Bilingualism in Ancient Society; DOI: 10.1093/acprof:oso/
9780199245062.003.0007) or the British Isles (B.C. TILGHMAN, “Writing in tongues: Mixed
scripts and style in Insular art”, in: Insular & Anglo-Saxon Art and Thought in the Early Medieval
Period, ed. C. HOURIHANE (University Park, PA, 2011), pp. 93-108, available from <www.
academia.edu/1498997/Writing_in_Tongues_Mixed_Scripts_and_Style_in_Insular_Art>), Insular
scripts (OKASHA, “Script-mixing”), and script and print in early modern Vilnius (J. NIEDZW ì IEDZì,
“Cyrillic and Latin script in late medieval Vilnius”, in: Uses of the Written Word in Medieval
Towns: Medieval Urban Literacy II, ed. M. MOSTERT and A. ADAMSKA (Turnhout, 2014: Utrecht
Studies in Medieval Literacy 28), pp. 99-116, who discusses both writing systems and types of
writing). A range of case studies are included in Biscriptality, many of them historical. Research
on script-switching is more rare. Some examples of work on switching between different writing
systems include ANGERMEYER, “Bilingualism meets digraphia”, on both script and print in
present-day Russian-American writing; and as the focus of Scripts Beyond Borders is on
‘allography’ (the writing of one language in the script of another), several of its chapters mention
the process of switching back and forth.
Code-Switching, Script-Switching, and Typeface-Switching 179

Lampeter Corpus led me to use that corpus. Examples of how these are tagged
in both sources are shown in the following sections.
In practice, then, the data was gathered by searching for the tags. A switch
from one script or language to another and back again within the same text was
counted as one switch; text-initial or -terminal switches were considered to be
embedded within the base script and language, and counted as single switches
(there were no cases where this was ambiguous). No parts of the texts were
excluded: the manuscript corpus contains datelines, salutations and superscrip-
tions, just as the print corpus contains title pages, front matter and headings
(paratexts in manuscripts use visual highlighting just as paratexts in printed
texts; cf. Fig. 1). Naturally, any possible errors in the tagging are reflected in
the results (but see the sections ‘Typeface-switching and code switching in
early modern English printed tracts’ and ‘What are the functions of typeface-
switching in early modern English tracts?’ below).
Using two corpora compiled by different people raises questions about
comparability.34 Script- and typeface-switching is in principle easy to distin-
guish – after all, the whole point of graphic flagging is for there to be readily
discernible visual contrast. This is particularly the case in print, where there are
no fuzzy typefaces halfway between upright roman and slanted italic. In the
case of manuscripts, it is possible for the script-switch to be subtle, but editors
will have a trained eye to spot such minute variation. In other words, it is rea-
sonable to expect that where script- / typeface-switching is marked in a digit-
ised text, the tagging is reliably accurate.35
Tagging instances of foreign language use is potentially more problematic.
In the letters of Richard Cocks, I have applied the same principles as used in
the Corpus of Early English Correspondence and the Helsinki Corpus, where

34
Users of corpora and editions need to be aware that they are, in essence, outsourcing
philological and also palaeographical analysis (S. DOLLINGER, “‘Philological computing’ vs.
‘philological outsourcing’ and the compilation of historical corpora: A Late Modern English test
case”, Vienna English Working Papers (VIEWS) 13.2 (2004), pp. 3-23, <anglistik.univie.ac.at/
views>; R. LASS, “Ut custodiant litteras: Editions, corpora and witnesshood”, in: Methods and
Data in English Historical Dialectology, ed. M. DOSSENA and R. LASS (Bern, 2004: Linguistic
Insights 16), pp. 21-48). This directly affects the comparability of results: if a feature is not
captured in one of the sources used, naturally comparison is fruitless.
35
Some caveats are worth adding to these statements. First, editorial practices always reflect
the interests of the editors, and inevitably fail to comprehensively represent all aspects of the
manuscript or print artefacts. Second, one can in fact imagine cases where the contrast in printed
texts is more subtle – for instance, if different founts are used (by accident or intent), where the
letter-shapes fall on a cline somewhere between upright roman and slanted italic.
180 SAMULI KAISLANIEMI

in uncertain cases historical context and the OED were used to support editorial
decisions whether to tag things as ‘foreign’ or not.36 Much the same principles
were followed by the compilers of the Lampeter Corpus, and in both cases the
compilers note that there are bound to be many cases where not only present-
day opinions of ‘foreignness’ vary, but contemporary audiences might also
have disagreed with the editorial decisions.37
Ultimately, there is no satisfactory way to identify conclusively all in-
stances of ‘foreign’ language use in any text. Where we can say that the tag-
ging of visual highlighting is reliable and, thus, comparable, the best we can
conclude regarding the tagging of foreign items and passages is that the com-
pilers of both resources used in this study draw on the same tradition of corpus
compilation. Regardless of whether you agree with the tagging of foreign items
in these corpora or not, the results, I argue, are comparable.

Script-Switching and Code-Switching in Early Modern English Manu-

script Letters

The aim of this first case study is to look at the correlation of script-switch-
ing and code-switching in early modern English manuscript letters. Given that
contemporaries associated languages with certain scripts, I wanted to see
whether and to what extent this association was retained in code-switching
practices. In present-day English it is still normal to flag foreign words by
italics, so I expected the correlation to be fairly strong.38 One of my initial
questions was to see whether script-switching could reveal the level of integra-
tion of ‘foreign’ items in a language. That is to say, would loanwords and
borrowings be in the running script, and code-switches always script-switch-
ed?39

36
See A. NURMI, Manual for the Corpus of Early English Correspondence Sampler CEECS
(Helsinki, 1998), available at <clu.uni.no/icame/manuals/CEECS/INDEX.HTM>, ch. 2; but given
in more detail in M. Kytö, Manual to the Diachronic Part of The Helsinki Corpus of English
Texts: Coding Conventions and Lists of Source Texts, 3rd edn. (Helsinki, 1996), available at
<clu.uni.no/icame/manuals/HC/INDEX.HTM>, pp. 30-31, section 3.2 (2b).
37
CLARIDGE, “Life is ruled”, p. 25.
38
Cf. Chicago Manual of Style; CRYSTAL, Making a Point, pp. 323-324.
39
The same question is raised by ANGERMEYER, “Bilingualism meets digraphia”, pp. 262-
263, 269.
Code-Switching, Script-Switching, and Typeface-Switching 181

My material for this case study is a collection of 105 letters from 1603-
1609, containing 76,761 words. They are a part of the British diplomatic corre-
spondence, and were written by an English merchant called Richard Cocks
(1566-1624), who lived in Bayonne. Cocks’s letters surviving from this period
are written in English, and the primary script used in them is English secretary
script (see Figs. 2, 3, 5, 6). Cocks was making a living trading in France and
Spain, and he could speak and write French and Spanish. Accordingly, Cocks
could also write in their respective ‘national’ scripts – namely in French secre-
tary script and in Spanish italic script.
As noted in the previous section, in the process of editing the letters of
Richard Cocks, I inserted tags to mark changes in script and language. To illus-
trate the encoding which allows quick retrieval of script- and language-switch-
es, Example 1 shows an extract from the unformatted transcriptions. The base
transcript is in plain text. I have used pseudo-XML to encode changes in script
and language: foreign language is marked inside <foreign />-tags, with the
language marked with an attribute (=sp for ‘Spanish’). Script-switches are
marked with milestone tags (which have no closing tag), <sc/>, with attrib-
utes indicating the script switched to (=sp for ‘Spanish’, =en for ‘English
[secretary]’).40 Further, I have used ASCII-based tags to encode textual features
such as allographs, superscripts, and punctuation: long <s> is marked as +s,
superscripts like “St” for “Saint” as S=t=, and <·> (raised punctus) as !.41

Example 1. Extract from the base transcription of a letter of Richard Cocks show-
ing the encoding scheme (SP 94/12 f. 150v; see Fig. 6 below)

the <foreign=sp>veador</foreign> at <sc=sp/> S=t=ebastians

<sc=en/>was put out of <lb/> his place by meanes of +lettrs
+sent from don <lb/> <sc=sp/>Juan de Ta+s+ses ! <sc=en/>only
becau+se he mi+sv+sed engli+shmen <lb/> but now <sc=sp/>don

40
“Pseudo-XML” means that the XML encoding (Example 1) is not well-formed: the scheme
was developed ad hoc in the editing process, where graphemic accuracy was primary and the
result needed to be human-readable. The tags are designed to be replaceable by well-formed TEI
XML using search and replace, e.g. <sc=en/> into <handShift script="english-
secretary"/>.
41
These practices are derived from conventions used in the Corpus of Early English
Correspondence and Helsinki Corpus: see H. RAUMOLIN-BRUNBERG and T. NEVALAINEN,
“Historical sociolinguistics: The Corpus of Early English Correspondence”, in: Creating and
digitizing language corpora, 2: Diachronic databases, ed. J.C. BEAL, K.P. CORRIGAN and H.L.
MOISL (Houndsmills, 2007), pp. 148-171, pre-print available at <www.helsinki.fi/varieng/
CoRD/corpora/CEEC/generalintro.html>; and KYTÖ, Manual, pp. 18-40.
182 SAMULI KAISLANIEMI

Fig. 2 Code-switch with script-switch (letter from Cocks to Thomas Wilson, 5 May
1605; TNA SP 94/11 f. 38v (detail)).

Juan d<sc=en/>e <sc=sp/>Ta+s+ses <sc=en/>is retorned . the

<lb/> <sc=sp/><foreign=sp>veador </foreign> <sc=en/>is
e+stabli+shed againe in his place & begyneth <lb/> to kepe a
greater rut then eaver / I lyke <lb/> not the Spani+sh
fations /

Cocks was what his contemporaries called a Spanish merchant, i.e. his
primary trade was into Spain. Not surprisingly, then, Spanish is the most fre-
quent foreign language found in his letters. To show some examples, Fig. 2
reproduces a part of a letter where Cocks quotes a Spanish proverb, and script-
switches into Spanish italic as he does so. The following is a transcription of
the text shown in the image (in this and all transcripts from manuscripts below,
italics are used for italic script, and bold emphasis is added):

yor selfe . I would writ more then I doe / but I

remember the ould Spanish proverbe hablan cartas . y .
callian barbas / I can say noe more, but that it is pyttie
yow should be put by the place, & that the other is not fytt

In present-day Spanish, the proverb is “hablan cartas y callan barbas”, liter-

ally “letters speak, so beards should keep quiet”, the meaning being that words
are no use against black on white.
But the correlation of script- and code-switching turns out not to be rigid.
In contrast to this example, where the proverb is clearly written in a different
script, Cocks often does not script-switch when using a single Spanish word.
In the excerpt reproduced in Fig. 3 he writes, starting after the long blank space
(insertion marked):
Code-Switching, Script-Switching, and Typeface-Switching 183

Fig. 3 Code-switch without script-switch (letter from Cocks to Thomas Wilson, 8

February 1607; TNA SP 94/13 f. 150r (detail))

Alsoe now of late

ther had =\lick to haue/ byn a great rebuelto at Bourdeaulx . by

In this case, the word “rebuelto” (present-day Spanish revuelto, a ‘revolt’ or

‘riot’) is not script-switched.
Tallying the number of script- and code-switches from all of Cocks’s let-
ters, the results are shown in Table 1. We can see that script-switching is much
more frequent than code-switching, being almost exactly three times as com-
mon. But to look at the results we were after: out of 267 code-switches in
Cocks’s letters, in 108 cases the switch of language is accompanied by a switch
of script, or about 40% of the time. More than half of the code-switches in
Cocks’s letters are not visually highlighted. As for graphic flagging, 86% of the
script-switches in Cocks’s letters mark other things than code-switches (see
Fig. 6 and adjacent discussion below).

Table 1 Script-switches and code-switches in the letters of Richard Cocks, absolute

frequencies

Script-switch No script-switch TOTAL

Code-switch 108 159 267

No code-switch 685

TOTAL 793

Since we do not have a control corpus, it is impossible to tell whether

Cocks script-switches more or less often than his contemporaries. Elsewhere I
have shown that Cocks’s letters contain more code-switches than early modern
English letters on average, although not more than the letters of his merchant
184 SAMULI KAISLANIEMI

peers.42 But is 40% of the time more or less than what we can expect Cocks to
script-switch to mark code-switches? The figure is large enough to suggest that
the variation is meaningful. To return to the question of script-switching poten-
tially marking the level of integration of foreign items, Fig. 3 shows us that
even if this were the case generally, it cannot be a rigid rule, since the obvi-
ously foreign “rebuelto” is not marked by script-switching.43 I pursue this point
below in the section ‘Is script-switching in manuscripts a marker of language?’.
It is worth stressing that although the examples in this chapter are of
Cocks’s script-switches from English secretary into italic – and code-switches
from English to Spanish – Cocks’s letters also contain words and phrases in
other scripts and languages. The third most common script is French secretary;
the other languages found in Cocks’s letters are Latin and Italian.
Although they do not affect the results of this study, it is worth noting
briefly that ambiguous cases are not uncommon in manuscript material. In a
few places in Cocks’s letters, a word or phrase cannot be 100% attributed to a
specific language (usually it is a case of clashing cognates: “per mer” (‘by
sea’) is technically not the French par mer). And occasionally, script identifi-
cation is technically impossible since there are so many polyvalent letter
shapes, which are the same in two or more of the scripts involved (like
<a>,<b>, <t>, and <u> in Cocks’s secretary and italic hands).

Typeface-Switching and Code-Switching in Early Modern English Printed

Tracts

The distribution of script- and code-switching in Cocks’s letters raises

many questions. The one pursued in this study is, what was the situation in
contemporary printed texts? To answer this, I turned to the Lampeter Corpus
of Early English Tracts, which happily has encoded both typeface- and code-
switches – called “foreign elements” in the manual – in SGML.

42
S. KAISLANIEMI, “The linguistic world of the early English East India Company”, Journal
for Early Modern Cultural Studies17.3 (2017), in press.
43
Note that in describing an event that happened at Bordeaux, Cocks uses a Spanish word.
He was perfectly capable of using an English term instead (rebellion occurs in Cocks’s letters 5
times; rebel 6 times; revolt once; revuelto this once). Code-switches like this are probably cases
of reported speech: Cocks may have been quoting a source who spoke in Spanish.
Code-Switching, Script-Switching, and Typeface-Switching 185

Fig. 4 H. HODY, A Letter from Mr. Humphry Hody, to a Friend, Concerning a Col-
lection of Canons said to be Deceitfully omitted in his Edition of the Oxford
Treatise against Schism (Oxford, 1692: ESTC R35437; EEBO: RB 34342, The
Huntington Library, San Marino, California), p. 33), demonstrating typeface-
switches encoded in the Lampeter Corpus.44 Image published with the permis-
sion of ProQuest and the Huntington Library.

44
The Lampeter Corpus of Early Modern English Tracts, comp. J. SCHMIED, C. CLARIDGE,
and R. SIEMUND (1999), available from the Oxford Text Archive <ota.ox.ac.uk>. Information on
the Lampeter Corpus given in this section is drawn from CLARIDGE, “Life is ruled” (for a list of
the texts used with their authors and titles see p. 38).
186 SAMULI KAISLANIEMI

The Lampeter Corpus consists of non-literary prose texts from 1640-1740,

and contains 120 texts (12 per decade), totalling 1.2m words. These texts are
divided equally into 6 “domains” – more of less the same as ‘fields’ in register
studies: “a domain comprises texts that belong to the same area of life or know-
ledge, and thus share similar topics to some extent”. The domains are called
Religion, Politics, Economy, Science, Law, and Miscellaneous (a “‘catch-all’
category” for texts which do not fit the other domains).
For this case study, I only used the first subperiod of the Lampeter Corpus,
being closest in time to my manuscript material. This subperiod contains 12
texts (two per domain) from 1641-1649, totalling 129,431 words.
Fig. 4 reproduces a page from the source of Lampeter Corpus text
RelB1692 (domain Religion, text sample B, dating from 1692). The running
text in this tract is in roman type, with italic as the secondary typeface – even
though on this particular page, half the text is in italic. There are some
typeface-switches within the roman-type running text, mostly marking proper
names, but also used to flag topical words (l. 7, “Province”) and code-switches.
Most of the switched typeface occurs as marginal commentary, being long
quotations from Latin sources.
The first half of the text shown in Fig. 4 is reproduced below in Example
2. As can be seen, typeface-switching can be marked in different ways in the
Lampeter Corpus (here by <IT />-tags on the one hand, and by a
REND="it" attribute within an element on the other; IT / it denotes
‘italic’). The same applies for language, and when given as attributes, both
language and typeface can be included as attributes of several different ele-
ments (for instance <Q /> for ‘quotation’).

Example 2 Extract from Lampeter Corpus (text RelB1692) showing the encoding
scheme45

<PB N="33">of his Province, were fled to <IT>Rome</IT> to

have their Cause heard by <IT>Cornelius</IT>, when (as he
says) it was contrary to the Constitutions of the Church, and
likewise to reason and equity, that a Cause should be try'd
in any other <IT>Province</IT> but that where the Crime was
committed; and that the Bishops of that Province are to give
an account of their Actions to God, and not to other Bishops.

45
The Lampeter Corpus is in plain text, and special characters are expressed as SGML
entities, for instance connecting hyphens as &rehy; and <æ> as &aelig;. These are omitted
or converted in the examples for the reader’s convenience.
Code-Switching, Script-Switching, and Typeface-Switching 187

<NOTE TYPE="marginalia" PLACE="inline" ANCHORED="NO"><Q

REND="it" LANG="LAT">Nam cum statutum sit ab omnibus nobis,
& æquum sit pariter ac justum, ut uniuscujusq causa illic
audiatur, ubi est crimen admissum, & singulis Pastoribus
portio gregis sit adscripta, quam regat unusquisque &
gubernet, rationem sui actûs Domino redditurus; oportet
utique eos, quibus præsumus, non circumcursare, nec
Episcoporum concordiam cohærentem suâ subdolâ & fallaci
temeritate collidere, sed agere illic causam suam, ubi &
accusatores habere & testes sui criminis possint; nisi si
paucis desperatis & perditis minor videtur esse auctoritas
Episcoporum in Africâ constitutorum, qui jam de illis
judicaverunt, & eorem conscientiam multis delictorum laqueis
vinctam judicii sui nuper gravitate damnarunt.</Q> Ep.
59.</NOTE></P>

The number of typeface-switches and code-switches in the 1640s part of the

Lampeter Corpus can be seen in Table 2.

Table 2 Typeface-switches and code-switches in the Lampeter Corpus (1640s only),

absolute frequencies.

Typeface-switch No typeface-switch TOTAL

Code-switch 389 7 396

No code-switch 5214

TOTAL 5603

There are by far more typeface-switches in these printed texts than there were
script-switches in the manuscripts. Further, nearly all of the code-switches in
the material are marked by typeface-switching. This latter discovery is particu-
larly striking, and investigation further reveals that, in fact, all seven instances
of code-switch unmarked by typeface-switch appear not to be meaningful,
being mistakes by the typesetter or the encoder of the text.46 That is to say, for
46
For instance, in the text LawB1649, on p. 3 the running text is in roman typeface. A
passage therein is encoded as:

conferre with HIM (<FOREIGN LANG="LAT">NOBISCUM</FOREIGN>,

not his Heirs and Successors) of the weighty and urgent
affairs that concerned (<FOREIGN LANG="LAT">NOS</FOREIGN>)
HIM and HIS KINGDOME of England
188 SAMULI KAISLANIEMI

Table 3 Word count, and typeface-switches and code-switches in the Lampeter Corpus
(1640s only) by domain, normalised frequencies.

Lampeter 1640s Word count Typeface-switches Code-switches

subcorpus domain per 10k words per 10k words

Economy 12,599 240 2.4

Miscellaneous 23,424 237 4.3

Politics 15,759 173 10.2

Law 17,951 287 12.3

Science 27,406 714 51.4

Religion 32,292 620 63.2

TOTAL 129,431 433 30.6

all practical purposes, we can consider code-switches always to be marked with

a typeface-switch in this material.
The domains allow us to break down the results to look at genre or register
variation. Table 3 gives normalised frequencies for typeface- and code-switch-
ing in the material (subcorpus word counts vary, but normalised figures are
comparable). As can be seen, there is in fact considerable variation between the
domains in the numbers of both typeface-switches and code-switches. The
domains Science and Religion contain the most of each type of switches, which

That is, neither “NOBISCUM” or “NOS” is encoded as italic typeface. A look at the printed work
shows that the first word is in roman capitals, and the second in italic capitals (W. PRYNNE, A
legall vindication of the liberties of England, against illegall taxes and pretended Acts of
Parliament lately enforced on tre people (London, 1649: WING P3996A; EEBO: British Library),
p. 3). In the case of the former, I can only allocate blame on the typesetter through speculation,
but lapses like this, when foreign words are otherwise systematically in italic typeface, are
suggestive. As for the latter, since “NOS” is inarguably in italic typeface in the printed text, this
is a clear mistake on the part of the encoder. (The caveat must be added that the copy of the tract
used for the Lampeter Corpus is not in EEBO; but the three copies that are in EEBO agree on this
point).
Incidentally, on the topic of encoding mistakes, the whole second half of the text
PolA1646 is mistakenly encoded as being in italic typeface. Happily, the typeface-switches within
the text are all marked so this does not affect my results (although they are of course encoded as
switches from italic to roman, rather than the other way around).
Code-Switching, Script-Switching, and Typeface-Switching 189

Table 4 Script- / typeface-switches and code-switches in the letters of Richard Cocks

and the Lampeter Corpus (1640s only) compared, absolute frequencies with
normalised frequencies and percentages.

Corpus Cocks Lampeter

Time span 1603-1609 1641-1649

Word count 76,761 129,431

Script- / typeface-switches 793 5,603

Script- / typeface-switches per 10k words 103.3 432.9

Code-switches 267 396

Code-switches per 10k words 34.8 30.6

% of code-switches with script- / typeface-switch 40.4% 98.2%

% of script- / typeface-switches used to flag code-switch 13.6% 6.9%

is not surprising considering their conventions of citing sources and authorities,

most of which would be in foreign languages.
A more detailed breakdown of these results is beyond this survey, but just
as Cocks’s letters contain more than two languages and scripts, Latin is of
course not the only foreign language found in the Lampeter Corpus, and nei-
ther is italic the only alternate typeface to roman.47 In the subcorpus used in
this study, of 396 instances of foreign language, 288 are Latin, 95 Greek, 8
Hebrew, 3 French and 2 Arabic. Of 5603 typeface-switches, 5236 are to italic,
194 roman (within for example italic passages), 91 Greek, 69 small capitals, 8
Hebrew, and 5 bold typeface.48

47
For other studies of the Lampeter Corpus, see CLARIDGE, “From page to screen”; A.
NURMI and P. PAHTA, “Multilingual practices in the language of the law: Evidence from the
Lampeter corpus”, in: Ex Philologia Lux: Essays in Honour of Leena Kahlas-Tarkka, ed. J.
TYRKKÖ, O. TIMOFEEVA, and M. SALENIUS (Helsinki, 2013: Mémoires de la Société Néophilo-
logique de Helsinki 90), pp. 187-205; and J. TUOMINEN, “‘Trifling Shews of Learning’? Patterns
of code-switching in English sermons 1640-1740”, in: Multilingual Practices in Language His-
tory (in preparation).
48
For a full list of languages and typefaces found in the entire Lampeter Corpus, see
CLARIDGE, “Life is ruled”, pp. 24-25.
190 SAMULI KAISLANIEMI

Comparison of Results

As can be seen from comparing Tables 1 and 2, script- / typeface-switching

and code-switching practices are quite different in the two sets of material.
Table 4 directly compares the results with normalised figures and percentages.
Although we have no data to give an answer about general contemporary prac-
tice, to which these figures should of course be first compared, we can use
Table 4 to try to answer several questions.
While we cannot tell whether the number of script- or typeface-switches in
these materials is generally high or low, in comparing the two we can in any
case see that typeface-switching is more than four times as frequent as script-
switching. Indeed, Table 3 shows that typeface-switching is much more fre-
quent across all domains in the Lampeter Corpus than script-switching in
Cocks’s letters.
For the frequency of code-switches in the materials, on the other hand, the
relative number of foreign words and phrases in Cocks’s letters and in the
Lampeter Corpus excerpt are more or less par. Other studies suggest that the
figures for both materials are on the high side, although as already seen in
Table 3, there is great variation between domains within the Lampeter Cor-
pus.49
Turning to the co-occurrence of script- / typeface-switching and code-
switching, how often are code-switches visually marked? Here the practices in
manuscript and print diverge greatly: in Cocks’s letters, code-switches are
marked less than half the time. In sharp contrast, in the Lampeter Corpus ex-
cerpt, code-switches are marked as a rule.
Finally, what about the inverse point: to what degree is graphic flagging
used particularly to mark code-switching? Interestingly, the correlation appears
to be stronger in manuscripts than in printed texts. In other words, typeface-
switching is not only used more in printed texts, but used more for other things
than marking foreign words and phrases.

49
See a.o. NURMI and PAHTA, “Social stratification”; ID., “Multilingual practices”; PAHTA
and NURMI, “Code-switching”; ID., “Multilingual discourse”; KAISLANIEMI, “The early English
East India Company”.
Code-Switching, Script-Switching, and Typeface-Switching 191

Fig. 5. To script-switch, or not to script-switch 1 (letter from Cocks to Thomas Wil-

son, 2 June 1603; TNA SP 94/9 f. 30v (detail))

holland or Sealand ./ the flemyng is made good prise /

but the Corigedor . would not p[er]myt them to mack
prise of the Englshman . soe that the haue Apealed
to the Court of Spaine . And the Embassetor · wth
the Corigedor haue written in the Englishmans favour
sayinge that p[ro]clemation is made in Spaine that

Discussion and Further Thoughts

In this section I briefly discuss two avenues of investigation which stem

from the case studies above. First, I look more closely at whether script can
function as a marker of language in code-switching in Cocks’s letters. And
second, I look at the variety of functions of typeface-switching in one tract in
the Lampeter Corpus.

Is Script-Switching in Manuscripts a Marker of Language?

In Fig. 3 above, the word “rebuelto” in a letter by Richard Cocks was not
script-switched, despite being in a foreign language. It is possible that writers
simply fail to script-switch even though they mean to – but then, the figures in
Table 1 suggest that this is a simplistic explanation. If the same words are
sometimes marked by switching script and sometimes not, can we tell if there
192 SAMULI KAISLANIEMI

is any meaning in this variation? To return to my earlier question, is there

something we can say about non-script-switched code-switches? In this section
I give two further examples of where Cocks does not script-switch.
The first example is seen in Fig. 5 (transcription follows the image; brevi-
graphs are expanded in square brackets). This image shows a passage from a
letter by Cocks in which, within the same paragraph, the Spanish term for the
provincial chief judge, corregidor (Cocks spells it “Corigedor”), occurs first
script-switched (to Spanish italic), and then, three lines down, unswitched. If
we take script-switching to be a possible indicator of the degree of integration
of foreign elements, does a case like Fig. 5 show a code-switch in the process
of becoming established as part of the ‘native’ (English) lexicon of the writer?
Does the need to mark such words visually erode over time?
One possible explanation for the scribal behaviour seen in Fig. 5 is that
only the first instances of a foreign item need to be marked (as new informa-
tion). However, there are two easy objections to this explanation. First, almost
all of Cocks’s letters are to the same recipient. The word corregidor occurs
very frequently in them, and is usually, but not always, marked. If it was a case
of flagging new information, then corregidor would only have had to have
been marked in the first occurrence in a long correspondence. Secondly, we
have cases where the opposite is true, as seen in Fig. 6.
In this example, again within the same paragraph, Cocks first does not
script-switch the word veador – the overseer of the local port or market – and
then, four lines down, switches into Spanish italic when writing the same word.
In other words, Cocks has not highlighted the first instances of a foreign word
systematically. And indeed, there are numerous examples of such variation in
Cocks’s letters.
The conclusion based on this evidence is, then, that script-switching is not
a reliable measure for determining language boundaries in manuscripts (or in
the texts written by Cocks, at any rate), nor of the degree of integration of
foreign items. Script-switches may be of use in a detailed long-term study, for
instance looking at whether, as I said above, writers highlight foreign items less
and less over time.
The caveat for all the evidence drawn from Cocks’s early letters is that I
have been exploring and describing a scribal idiolect. Without further data, we
cannot say whether Cocks’s script-switching practices are idiosyncratic or
typical. The general characteristics nonetheless probably hold: that script-
Code-Switching, Script-Switching, and Typeface-Switching 193

Fig. 6 To script-switch, or not to script-switch 2 (letter from Cocks to Thomas Wil-

son, 6 January 1606; TNA SP 94/12 f. 150v (detail))

the veador at Stebastians was put out of

his place by meanes of lettrs sent from don
Juan de Tasses · only because he misvsed englishmen
but now don Juan de Tasses is retorned . the
veador is established againe in his place & begyneth
to kepe a greater rut then eaver / I lyke
not the Spanish fations /

switching is primarily a device of textual highlighting, used for a variety of

purposes, of which flagging foreign items is only one.

What Are the Functions of Typeface-Switching in Early Modern English Tracts?

To get a better sense of what kinds of purposes typeface-switching is put

to in printed texts, this section looks at one text in the Lampeter Corpus in
more detail. The text chosen was EcA1641, Richard Kilvert’s A reply to the
most untrue relation made and set forth in print, by certaine vintners, in excuse
of their wine project, printed in 1641, some 3637 words long. This was
194 SAMULI KAISLANIEMI

Table 5 Typeface-switches in the EcA1641 text in the Lampeter Corpus, by switch

type, absolute frequencies.50

Switch type Subtype N Examples

Proper names Places 30 Canary, French, London

People 28 Abel, Griffith, Kilvert

Organisations 5 Vintners

Dates* Months 15 February, November

Days, years 5 13, Michaelmas, 1637

Text structuring Title page 2

Work title 2

Section headers 8

Lists 5

Quoted letter 1

Letter sections 3

Style-switches Accounting terms 14 per, provizo, viz.

TOTAL 118

* There are only two foreign items in EcA1641, “Die Mercurij” and “Iulij”, here counted as dates
(although counted as code-switches in the figures given in section ‘Typeface-switching and code-
switching in Early Modern English printed tracts’).

checked against a copy of the same work on EEBO.51 The results of the survey
are seen in Tables 5, 6 and 7.

50
There is a slight discrepancy between the figures given in Table 5 and those incorporated
for the entire Lampeter Corpus 1640s extract. According to the encoding, there are 119 typeface-
switches in EcA1641, but a close analysis reveals that two tags are redundant (e.g. a <QUOTE
rend=it> within a <DIV rend=it>), and one switch should be divided in two (discussed
below). If the net effect in the entire corpus excerpt is a similar +<1% false positives, this issue
can be safely disregarded: see the discussion below.
51
R. KILVERT, A reply to a most untrue relation made and set forth in print by certaine
vintners in excuse of their wine proiect (London, 1641: WING K478; EEBO: Thomason Collection,
British Library); this may or may not be from the same print run.
Code-Switching, Script-Switching, and Typeface-Switching 195

In Tables 5 and 6, the typeface-switched words in EcA1641 are grouped by

categories (which emerged in the process: other categories are of course possi-
ble). Table 5 shows a detailed breakdown, while the figures are summarised in
Table 6. Finally, Table 7 records whether the typeface-switched words also
occur without graphic flagging.
As seen in Tables 5 and 6, typeface-switching in EcA1641 is used for four
primary purposes: to flag proper names, dates, jargon, and then to organise or
structure the text. The largest category are proper names: the names of people
and places are always graphically flagged.52 Similarly, dates, when written out
(see Table 7), are always highlighted. ‘Style-switches’ indicate words from
different registers, or jargon, here being words common in mercantile texts that
come from legal or accounting practices, such as per, proviso, and viz. (i.e.
videlicet). Note that while these words have long been established in English
by this time, they still retain a foreign flavour.53 These words have not been
encoded inside <foreign />-tags, which shows that typeface-switching
alone has not been used by the compilers of the Lampeter Corpus as an indica-
tor of foreign items.
The last category is text structuring or text organisation. This includes
typeface-switching on the title page, in section headings, in lists, and other
similar places. Noteworthy here is a letter quoted in its entirety (pp. 11-12),
printed in italic typeface, but with the salutation, closing formula and super-
scription in roman typeface.54 Without a control corpus or data of contempo-
rary practice, we cannot extrapolate from these figures whether this printer’s
practices of typeface-switching are typical or unusual.55 We can say with some
confidence, however, that the main function of typeface-switching is highlight-
ing names. Tables 6 and 7 show that names of people and places are always
typeface-switched in EcA1641, and indeed that over half of the instances of
typeface-switching are names. This practice appears to have been followed in
manuscripts too (see Fig. 1). In the letters of Richard Cocks, although names

52
CLARIDGE, “Life is ruled”, p. 24, notes this is the case in the entire Lampeter Corpus.
53
OED, s.v.; CRYSTAL, Making a Point, pp. 323-324.
54
The base text in EcA1641 is in roman type, with most switches to italic, although
(particularly in headings) some switches are to small capitals. In passages in italic typeface
(headings, the quoted letter), the typeface-switches are usually to roman, sometimes to small
capitals. The typefaces are layered and embedded, e.g. the base roman text can have a passage in
italic, containing a switch to roman, with a further switch to small capitals.
55
A survey of the rest of the texts in the 1640s part of the Lampeter Corpus would reveal
more in this regard.
196 SAMULI KAISLANIEMI

Table 6 Typeface-switches in the EcA1641 text in the Lampeter Corpus, by switch

type, absolute and relative frequencies.

Switch type N % of typeface-switches

Proper names 63 53.4%

Dates 20 16.9%

Text structuring 21 17.8%

Style-switches 14 11.9%

TOTAL 118 100.0 %

are not always script-switched, the proportion of all script-switches used to

highlight names is as high as 88% (cf. Figs. 5 and 6).

Table 7 Typeface-switch vs. no typeface-switch: comparison of some switch types in

the EcA1641 text in the Lampeter Corpus, absolute frequencies.

Typeface-switch No switch

Names: people, places 58 0

Names: organisations 5 30

Dates: words 18 0

Dates: numerals 2 23

Style-switches 14 1

Yet not all names and dates are typeface-switched. Table 7 takes a part of
the list of words that have been typeface-switched in EcA1641, and charts
whether they have been switched throughout the text. Place names and per-
sonal names are consistently switched, but organisation names are not: only the
first five instances of Vintners have been graphically flagged. As the tract is
directed against the members of the Vintners’ Company, presumably these first
five instances are highlighted as topical words for rhetorical emphasis – to
underline the target of the accusations.
As for dates, numerals are never given in italic typeface (the switches
listed in Tables 5 and 7 are to roman). It is not possible to tell if this is by con-
Code-Switching, Script-Switching, and Typeface-Switching 197

vention, for effect, or if this simply means the printer lacked italic numerals.
Interestingly, one per (top p. 16) is in roman, but I suspect this was missed by
the typesetter since the word only occurs in a set phrase, for example “40 shil-
lings per Tunne”.56
Discussing these aspects raises questions about what constitutes a type-
face-switch and what does not. First, what constitutes a typeface? The compil-
ers of the Lampeter Corpus have tagged small capitals as a separate typeface;
while this makes sense visually, it does not follow the division of typefaces
argued in the section ‘Code-switching and script- and typeface-switching’.
Comparative studies need to make sure they count the same phenomena.57
Secondly, how should we count joint instances? For example, on p. 17 in
EcA1641 a passage runs,

the 13. of <IT>February, Griffith</IT> writes to.

Since this study relied on the tags in the source materials, this was counted as
one switch. But, although “February” and “Griffith” are both given in italic
typeface throughout the text, arguably they do not (here) belong to the same act
of typeface-switching, and should be treated as two separate switches.58
Another example is provided by the only code-switches in this text: a sec-
tion heading on p. 7 reads

<foreign rend=it LANG="LAT">Die Mercurij</foreign> 21.

<IT><FOREIGN LANG="LAT">Iulij </FOREIGN></IT> 1641.

Although encoded and, thus, counted as two separate code-switches, arguably

the whole date should be counted as a single Latin phrase.
Thirdly, can we be certain we can accurately attribute functions to in-
stances of typeface-switching? Is the section heading just described typeface-
switched because it is a date, or because it is in Latin, or because it is used as
a section heading – that is, as a text organisation device? Or is there a further
reason for printing (most of) it in italic typeface? The answer must be that there

56
Per is not mandatory: it occurs 9 times, and “a Tun(ne)” occurs 14 times.
57
Only 69 out of 5603 typeface-switches in the 1640s part of the Lampeter Corpus are to
small capitals, so while these strictly speaking may not be comparable to script-switches in
Cocks’s letters, their small number does not affect the results given above.
58
Comparison with KILVERT, A reply, p. 17, actually shows that this is another encoding
error, for the printed comma is in roman typeface. But the principle and question stand.
198 SAMULI KAISLANIEMI

is of course overlap in functions, and it is not uncommon for there to be several

reasons to highlight a word or passage in a text (cf. “Corigidor” in Fig. 1).
Naturally, this further complicates any attempts to establish causality between
graphic flagging and the use of foreign items in a text.
Finally, it should be noted that the distribution of functions of typeface-
switching varies between texts. EcA1641 is just one out of the twelve tracts in
the Lampeter Corpus from the 1640s, and its typeface-switching practices may
or may not be typical. But while the figures in this section cannot be taken to
reflect general contemporary trends, they are certainly indicative of the range
of things typeface-switching is used for.

Conclusions

Can we draw some overall conclusions from this study? In Cocks’s letters,
script- and code-switching correlate, but the variation suggests that the link is
not very strong, or that other factors, not looked at in this study, are involved.
In the Lampeter Corpus, all instances of foreign language are typeface-
switched, but graphic flagging overall is used for a wide range of purposes, of
which marking foreign language is but a small part. In both materials, there is
a connection between the visual appearance of the text and the language, which
suggests that with further studies, trends may reveal themselves.
The differences between script- and typeface-switching practices seen
above raise new questions. Do they reflect the different authorship of the
switches? The script-switches in Cocks’s letters are authorial, but the typeface-
switches in the Lampeter Corpus were most likely inserted by the typesetters.59
Am I comparing a scribal idiolect of a single writer to the practices of several
different printing-houses? Are the genres – letters and six “domains” – compa-
rable? Can we really compare textual practices in the 1600s and 1640s? In
manuscript and print?60 I can only hope that these potential objections can for
the moment be ignored, as we are attempting to understand contemporary prac-

59
SUHR, Publishing for the Masses, pp. 64-68.
60
The relationship between manuscript and print and the development of script- and
typeface-switching appears to be very complex (JENKINSON, “English current writing”; cf. DEN
HEIJER and SCHMIDT, “Scripts beyond borders”). Certainly there was influence both ways, so that
while manuscript practices were replicated in print, conventions developed in print came to be
copied in manuscripts (M.B. PARKES, Pause and Effect: An Introduction to the History of
Punctuation in the West (Aldershot, 1992), p. 56).
Code-Switching, Script-Switching, and Typeface-Switching 199

tices in general. We need much more data before we can even begin to see the
whole picture of these practices in manuscript and print in early modern Eng-
land.
Considering the historical developments of scripts and typefaces outlined
in the section ‘Scripts, typefaces and languages in early modern England’, we
could posit that these case studies do show that in the early seventeenth cen-
tury, graphic flagging was in the process of changing from a method to mark
texts in different languages to becoming a generic means of visual highlighting,
and that this process was being led by print. Thus in Cocks’s letters there are
two systems at play: the (old) association of scripts and languages, and the
(newer) practice of graphic flagging where the form is not tied to a specific
meaning. In comparison, in the early tracts in the Lampeter Corpus typographic
flagging is used for a wide range of purposes, and typeface-switching has be-
come a tool for differentiation with little or no regard to the idea of language-
appropriate typefaces.
This chapter has argued that script- and typeface-switching in early modern
English texts is not the same as switching between different writing systems.
This point is worth stressing, because the multilingualism visible in early mod-
ern manuscript texts manifested in a digraphia which does not exist in English
today.61 Societies where different writing systems are in use at the same time
are of course common, and it is typical for writing systems to carry different
connotations. Studying these will help us to understand early modern English
script-switching better.
Today, we have for the most part lost differentiation between types of
writing as something that gives meaning. Yet different typefaces – or rather, to
use the term familiar from computers, different fonts, still carry different asso-
ciations, even if these are to domains and spheres of usage rather than to lan-
guages (for instance, Courier New is often used for code, as in this chapter).62
Likewise, despite being used for a wide range of purposes, different kinds of

61
The Japanese writing system presents one possible parallel. The hiragana and katakana
syllabaries are allographic (graphemically the same), but used for different functions; loanwords
and foreign items are conventionally written in katakana (cf. C.K. SCHMIDT, “Loanwords in
Japanese”, in: Loanwords in the World’s Languages: A Comparative Handbook, ed. U. TADMOR
and M. HASPELMATH (Berlin, 2009), pp. 545-574, at pp. 559-561). On the other hand these
purposes developed gradually, and unlike italic script in England, katakana was not introduced
to Japan from outside as the script used to write a foreign language.
62
Associations vary between communities: the US National Security Agency appears not
to share the same bias against Comic Sans as, say, the academic community.
200 SAMULI KAISLANIEMI

typographic flagging carry different meanings (at least in different contexts),

for instance writing in all capitals in social media carries the sense of SHOUT-
ING. On the other hand, the transformation of italic typeface into a generic
differentiation type has led to most fonts today having an italic version. From
a sixteenth-century viewpoint, this makes as much sense as a situation where
all fonts would have a black letter version.
In any case, the main conclusion of this study is that further research is
warranted – on the connections of visual highlighting and language, as well as
on switching scripts and typefaces (kinds of writing). For instance, writers on
present-day English usage note that ‘unfamiliar’ and ‘foreign-looking’ words
are often italicised – but when does a loanword “stop ‘looking foreign’?”.63
Although the connection of script- and code-switching is neither strong nor
straightforward in Cocks’s letters, this question is worth pursuing in other
material. As script-switching is not an obligatory practice for any writer, any
and all visual highlighting can reveal more about their multilingual practices.
At present, however, there are few suitable corpora of early modern Eng-
lish manuscript texts – script not being something editors necessarily record in
editions or encode in corpora. The situation is slightly better for print, for in
addition to the Lampeter Corpus, EEBO-TCP has encoded typeface-switches –
but not the language of words and passages. For the moment, then, it is possi-
ble to pursue a better understanding of typeface-switching, but studies of
script-switching remain a laborious undertaking.

63
CRYSTAL, Making a Point, p. 323.
 Seeing is Reading:
Typography in Some Early Modern Dictionaries*

R.W. MCCONCHIE

Introduction

part from the last quarter-century, the history of research on dictionar-

A ies has been largely that of complacent neglect. Perhaps dictionaries are
rather like computers – we all use them often, understand almost noth-
ing of their fundamental make-up, and only get annoyed or curious when they
fail us. We fail to see their inherent biases, and even take their errors for truth.
It was perhaps only with digital access to the Oxford English Dictionary
through the CD-ROM version in 1987 that many scholars realised that here was
a huge, largely untapped research resource which was of interest in its own
right, and that other dictionaries might also be of intrinsic interest.1
Dictionaries have always been complex in nature, needing to convey a
relatively large amount of disparate information in each of the thousands of

*
My thanks to an anonymous reviewer for comments which greatly improved this chapter.
1
Pioneers in this field included Thomas Finkenstaedt, Jürgen Schäfer and Richard W.
Bailey.

......................................................................................................................................
Verbal and Visual Communication in Early English Texts, ed. Matti PEIKOLA, Aleksi
MÄKILÄHDE, Hanna SALMI, Mari-Liisa VARILA, and Janne SKAFFARI, Utrecht Studies in Medieval
Literacy, 37 (Turnhout: Brepols, 2017), pp. 201-218.

DOI 10.1484/M.USML-EB.5.114136
202 R.W. MCCONCHIE

smaller parallel chunks into which the text is broken, but their graphics have
not always reflected this basic fact. Some of this information is best conveyed
through the visual implications of variations in the text itself, not simply what
the text signifies linguistically. A change of typeface or size or the use of some
symbol is sufficient to tell the reader at a glance what he or she needs to know
about entry structure without recourse to verbal explanations. This streamlining
has immediate advantages in comprehension, and lexicographers and printers
have exploited and standardised them since the first printed dictionaries. Some
have even gained significance beyond their function as simple signposts.
The distinction between the various typefaces was not as trivial as it may
seem – it has been argued that at the less sophisticated level, typefaces may
indicate degrees of literacy and kinds of readership.2 Generally speaking, an-
other view is that the use of roman and italic typefaces as against black letter
represents a manifestation of what Rhodes has called “status anxiety” in his
discussion of Renaissance translation, encapsulating fears over the degradation
and vulgarisation of original texts.3 The particular question in this chapter is
about how dictionaries micro-manage the position, size, shape, and weight of
the type itself in ways that we now take for granted, but was not always done.
This chapter introduces some of the ways in which this was achieved.
We first need to say a little about the various types of dictionary we are
concerned with. There is no single, completely satisfactory definition of a
dictionary, but they do share at least some of the following characteristics.
They are usually monographs which list and define the words of one or more
languages. They may also provide a linguistic apparatus, such as pronuncia-
tion, grammar, marks for accentuation and syllabification, as well as providing
derivations, cognates, and etymologies for the headwords. The words listed are
individually defined, and account will often be taken of the polysemy of the
headwords. They may also incorporate citations to illustrate the use of words
and to show their place in the lexical history of a language. Dictionaries may
also provide some degree of encyclopaedic information about headwords; in
the more specialist rather than general dictionaries, this may be extensive.
Some exhibit many or indeed most of these characteristics, but rarely will all
of this be done in any given dictionary. Finally, a dictionary may cover a single
2
P.M. JONES, “Medical literacies and medical culture in early modern England”, in: Medi-
cal Writing in Early Modern English, ed. I. TAAVITSAINEN and P. PAHTA (Cambridge, 2011), pp.
30-43, at pp. 32-33.
3
N. RHODES, “Status anxiety and English Renaissance translation”, in: Renaissance
Paratexts, ed. H. SMITH and L. WILSON (Cambridge, 2011), pp. 107-120, especially pp. 110-111.
Seeing is Reading 203

language (monolingual), two languages (bilingual), or more than two (polylin-

gual). The latter two usually have definitions which are translations or glosses
rather than full semantic explanations of what a word means; that is, the com-
piler seeks interlanguage equivalence rather than semantic explication in the
form of definitions.4 Some or even most of these elements may be distin-
guished from each other by typographical means. A glossary, by contrast, is
usually incorporated into another work, often to explain the specialist terms
used in that work, and is often restricted in its lexicographical features. Glossa-
ries frequently list translation equivalents or synonyms rather than providing
full definitions.
Dictionaries became increasingly complex throughout the early modern
period. Typography was thus called upon not only to reflect these changes as
more diverse information was incorporated into entries and the nature of the
dictionary itself changed, but also to reflect the distinctions already internal to
lexicons which could aid the reader in identifying relevant textual elements and
their systematic implications. Although the move towards more nuanced typog-
raphy was slow and very inconsistent, progression through various kinds of
dictionary is observable. This paper offers a brief introductory survey of these
changes from the late fifteenth century until the middle of the eighteenth. Al-
though the works mentioned are primarily English, the survey is by no means
confined to the products of the English printing industry, which tended to lag
behind the continent in a number of important respects. The fundamental prob-
lem from the modern and more theoretical point of view is, as Paul Luna puts
it, the extent to which the typography was systematically mapped onto the
information contained in the dictionary.5
Early printed dictionaries show little differentiation between textual ele-
ments, even such elementary components as headword and text. By the mid
sixteenth century, however, sophistication is increasing, especially under the
influence of humanist classical scholarship. Capitals, small capitals, roman,
italic, and black letter and other founts are increasingly used for particular
functions, and systems eventually emerge for such linguistic distinctions as
grammar, pronunciation, and stress. In general, the more scholarly the dictio-

4
I leave out of this account the many reference works which are called dictionaries, but are
in fact specialist encyclopaedias containing no linguistic information, usually confined to
restricted subject areas.
5
P. LUNA, “Not just a pretty face: The contribution of typography to lexicography”, in:
Proceedings of the Eleventh EURALEX International Congress, EURALEX 2004, ed. G. WILLIAMS
and S. VESSIER, 3 vols. (Lorient, 2004), pp.847-857, at p. 849.
204 R.W. MCCONCHIE

nary, the more likely it was to show typographical sophistication. This growth
is also linked in a rather complex way to the emergence of monolingual dictio-
naries as distinct from glossaries, although not at first in England, and to the
widespread appearance of various types of polylingual dictionaries. The acad-
emy dictionaries, the first of which appeared in the early seventeenth century,
espoused typological gravitas, or as John Considine puts it, a “marked typo-
graphical dignity”.6 The ultimate beneficiary was the user, as readability and
accessibility improved.
The change in readability and differentiation of discrete textual elements
cannot, however, be necessarily ascribed to the lexicographer, whose input
must be thought of as merely one among several. The primary locus of deci-
sions about the graphic elements of a page is the printing house, in which the
printer in conjunction with the author / compiler and, ultimately, the composi-
tor will have the final responsibility, as Joseph Moxon makes clear in his
seventeenth-century account of the art of printing.7 This will also be deter-
mined by the established practice of the printing-house. Fundamental charac-
teristics such as the disposition of the text block on the page and the relative
sizes of header and footer will not have been in question, since they were es-
tablished much earlier. Other characteristics of dictionary layout appear to
move from being experimental and thus perceived characteristics to generally
well-understood and thus subliminal or received ones, to use a distinction from
the work of John Bateman.8
There are also several dimensions to be considered in terms of dictionary
type, given that not all types will necessarily employ exactly the same solu-
tions: first, the learned-popular distinction and, secondly, the Latin / Greek-
vernacular one. A third dimension is the distinction between English and conti-
nental works, while a fourth is bi- and polylingual as against monolingual.
How these were handled will emerge through consideration of some examples.

6
J. CONSIDINE, Academy Dictionaries 1600-1800 (Cambridge, 2014), p. 3.
7
J. MOXON , Mechanick exercises, or, the doctrine of handy-works ... (London, 1677),
2, No. 22, pp. 220-228.
8
See J. BATEMAN, Text and Image: A Critical Introduction to the Visual / Verbal Divide
(Abingdon, 2014).
Seeing is Reading 205

Some Examples

The examples chosen for this section should not be taken as a representa-
tive sample, but only as a diagnostic indication of what might be available for
further research. Most have been selected simply because they were conve-
niently available. Such works as the dictionaries from the Estienne presses are
largely passed over, since the most advanced of them would require highly
technical discussion of such matters as the use of Greek digraphs and brevi-
graphs. Similarly passed over are the bilingual Latin-English dictionaries pub-
lished in England during the seventeenth century.
Very much at the learned end of this spectrum and the earliest dictionary
we will consider is by Ambrosius Calepinus (Calepine, c.1435-1509): a purely
Latin, humanist dictionary which first appeared in Reggio Emilia in 1502 and
was subsequently printed in a number of cities before 1550, including Venice,
Hagenau, Paris, and Basel.9 This work, constantly re-edited and expanded,
ultimately became a polylingual and largely vernacular dictionary which en-
dured well into the eighteenth century.

Fig. 1a-b A typical page of Calepinus 1502 (Diiiiv) (Fig. 1b, on the next page),
preceded on this page by a detail (Fig. 1a) <https://de.wikipedia.org/wiki/
Ambrogio_Calepino#/media/File:Calepino-1502-483.jpg>.

9
A. CALEPINUS, Ambrosii Calepini Bergomatis Dictionarium (Reggio Emilia, 1502).
206 R.W. MCCONCHIE
Seeing is Reading 207

This dictionary is typographically unelaborated, as was typical of its time.

We see that the text itself shows only a few basic means of distinguishing
entries. Each lemma is introduced by a hanging indent, thus treating each as a
paragraph, but beyond that little distinguishes one entry from another. The
linguistic information about Latin endings, where it appears, is simply a visu-
ally undistinguished list incorporated into the text, as can be seen in the detail
of the illustration above. A single typeface is used throughout. This basic prin-
ciple of minimal differentiation was followed in England in early works such
as the Ortus Vocabulorum published by Wynkyn de Worde in 1509, which
even lacks the distinguishing feature of an indent to introduce the headwords.10
There is no variation in typeface at all, and no attempt to distinguish either the
grammatical information which is sometimes supplied or definitions from
quotations. Hence we find mactus macta mactum (‘worshipped’, ADJ.), and
macero maceras (‘make wet’ V) inserted without punctuation, the entry then
moving directly to the definition and quotes; macies (‘poverty’), however, does
not include the genitive maciei. A further familiar convention however, that of
not leaving very short lines at the ends of paragraphs, seems to have been
largely abandoned, as this is unavoidable in a dictionary. The white space thus
left is in fact an advantage for the reader, both in identifying the end of an entry
quickly and in allowing space for making short hand-written notes.
Robert Estienne’s Dictionarium, seu Latinæ linguæ thesaurus of 1531 set
a new standard for such publications.11 Quite apart from the high-quality press
work, this shows careful attention to the differentiation of the components of
the entries (see Fig. 2). Headwords are done in capitals with extra spacing
between the letters. Grammatical information is punctuated and derived forms
and quotations distinguished by a further level of indentation, each on a new
line. The source data for quotes and parallel quotes introduced by idem are
marked by extra spaces after the preceding full stop. Second letter alphabet-
isation (as in LE under L) is marked by a centred text line. It did not follow,
however, despite the desirability of these innovations from the standpoint of
the modern reader and user, that all Estienne dictionaries subsequently retained

10
ANON., [Ortus vocabulorum] (Londini, 1509). These and other early English dictionaries
were recently discussed in fuller detail in a paper at the OX-LEX (HEL-LEX) 4 conference in Oxford,
March 2015, by Gabriele Stein. Stein has written extensively on sixteenth-century dictionaries.
11
R. ESTIENNE, Dictionarium, seu Latinæ linguæ thesaurus (Parisiis, 1531). For Robert
Estienne’s dictionaries and influence, see E. ARMSTRONG, Robert Estienne, Royal Printer: An
Historical Study of the Elder Stephanus (Cambridge, 2011), pp. 87-94; on his types, pp. 53-54.
208 R.W. MCCONCHIE

Fig. 2 A page of Robert Estienne’s Dictionarium, seu Latinæ linguæ thesaurus, 1531
(bir) Les Bibliothèques Virtuelles Humaniste <http://www.bvh.univ-tours.fr/
Consult/consult.asp?numtable=B372612102_FB1859&numfiche=237&mod
e=3&ecran=0&offset=68>.

them.12 Sophistication and comprehensiveness did not always mean good

sales. The massive and immensely learned dictionary of Greek and Latin by
Robert’s eldest son Henri, the Thesaurus grecae linguae, was the high point of
humanist lexicographical scholarship, but the work derived from it, the
abridgement by Johan Scapula, ironically proved far more popular, going

12
I am not aware of any contemporaneous discussion of the typographical features of
dictionaries.
Seeing is Reading 209

through many editions and becoming a standard reference text.13 Generally

speaking, however, the typographical standards set on the Continent would
remain higher than those in England.
The first edition of Thomas Elyot’s dictionary (1538), the first humanist
dictionary in England, uses straightforward typography.14 There are two col-
umns per page with hanging indents. All headwords are left-justified and in
roman, while the definitions are in black letter, following the familiar conven-
tion of using black letter for English. Black letter, as Edmund Coote explains
in The English School-Maister is for words “meerely English, or from some
other vulgar tongue”.15 This dictionary is a distinct step forward in humanist
lexicography in England, and to some extent in typography as well. Grammati-
cal information is incorporated but not distinguished in type. Elyot’s dictionary
of 1548, the first version edited by Thomas Cooper, again shows the two-col-
umn layout and some sophistication in founts, employing roman for the head-
words and black letter for the definitions.16 The white space on the page is
more generous than in 1538. By 1565, the first edition of this work under Coo-
per’s own name, not a great deal has changed.17
While Elyot’s work continued to be published and re-edited, notably by
Thomas Cooper, who became Bishop of Winchester, two more dictionaries
appeared in the middle of the sixteenth century in England. Vernacular head-
words in black letter were employed by both John Withals in his modest no-
menclator of 1553, and in Richard Howlet’s impressive folio of 1552, although
in many respects these two works could hardly be more different.18 Richard
Howlet, about whom little is known, produced a large and probably overly
expensive folio.19 His Abcedarium anglico latinum came out adorned with the
13
J. SCAPULA, Lexicon Graecolatinum novvm in qvo ex primitivorvm & simplicivm fontibus
derivata atque composit (Basileæ, 1580). See J. CONSIDINE, Dictionaries in Early Modern
Europe: Lexicography and the Making of Heritage (Cambridge, 2008), pp. 83-84, for a dis-
cussion of Estienne’s typography in this dictionary.
14
T. ELYOT, The dictionary of syr Thomas Eliot knyght (Londini, 1538). For a recent and
very full account of Elyot as a lexicographer, see G. STEIN, Sir Thomas Elyot as Lexicographer
(Oxford, 2014).
15
E. COOTE, The English school-maister: Teaching all his scholers, the order of distinct
reading, and true writing our English tongue (London, 1596), p. 73.
16
T. ELYOT, Bibliotheca Eliotæ = Eliotis librarie. This dictionarie novv nevvly imprinted
... is augmented and inriched with aboue. xxxiij. thousande wordes and phrases (Londini, 1548).
17
T. COOPER, Thesaurus linguae Romanæ & Britannicæ (Excusum Londini, 1565).
18
J. WITHALS, A shorte dictionarie for yonge begynners (London, 1553); R. HOWLET,
Abcedarium anglico latinum, pro tyrunculis Richardo Huloeto exscriptore (Londini, 1552).
19
For fuller details of Howlet’s life and dictionary see R.W. MCCONCHIE, “The real Richard
210 R.W. MCCONCHIE

arms of Thomas Goodrich, Bishop of Ely, who became Lord High Chancellor
of England in the year the dictionary was published. It also had a fine, elabo-
rate title border. The intended market was obviously schools, but it was expen-
sive and cumbersome, and was almost immediately superseded by Withals’s A
shorte dictionarie for yonge begynners, also intended for school use. John
Withals is equally obscure as a person. The success of his dictionary may how-
ever be measured by the large number of editions which appeared by the end
of the sixteenth century – a striking contrast with Howlet, whose dictionary
was only re-edited once. Withals’s dictionary was a nomenclator, organised by
topic rather than alphabetically, a useful pedagogical feature in a school text.
Howlet’s large and relatively sophisticated dictionary employs a two-col-
umn layout, hanging indents, and black letter and roman. The Latin glosses
provide grammatical data. Howlet also makes use of historiated drop capitals
to begin letters and smaller drop capitals to distinguish second-letter alphabet-
isation, and occasionally pilcrows and manicules – inconsistently, but appar-
ently for some quotations, synonyms, and derived forms. Withals’s typography
is relatively unremarkable. A modest quarto in comparison to Howlet’s work,
Withals’s is neatly printed, with the usual blackletter-roman distinction, large
type for the centred sectional headings, a pilcrow for the first entry of each
topical section and hanging indents, although not consistently used, making the
text seem rather cluttered. A historiated capital adorns the beginning of the
prologue, but they do not appear elsewhere. This less expensive format obvi-
ously found a ready market.
John Baret’s Alvearie (“the beehive”), an English-Latin work first pub-
lished by Henry Denham in 1574 and then again in 1580, shows a great in-
crease in sophistication over most of what had preceded it, page layout and
indexing both setting a high standard.20 The book itself is a smallish folio
bound in sixes. This work is meant to be visually impressive in all respects.
The title page (Fig. 3) is an attractive floral image incorporating the beehive in

Howlet”, in: Words and Dictionaries from the British Isles in Historical Perspective, ed. J.
CONSIDINE and G. IAMARTINO (Newcastle upon Tyne, 2007), pp. 39-49.
20
J. BARET, An aluearie or triple dictionarie in Englishe, Latin, and French: very profitable
for all such as be desirous of any of those three languages. Also by the two tables in the ende of
this booke, they may contrariwise, finde the most necessary Latin or French wordes, placed after
the order of an alphabet, whatsoeuer are to be founde in any other dictionarie: and so to turne
them backwardes againe into Englishe when they reade any Latin or French aucthors, & doubt
of any harde worde therein (London, 1574).
Seeing is Reading 211

Fig. 3 Baret’s Alvearie, title page (image from the second edition, 1580).

Fig. 4 Baret’s Alvearie, dedication drop capital with Cecil’s coat of arms (image
from the second edition, 1580).
212 R.W. MCCONCHIE

Fig. 5a-b. A typical page of the Alvearie, with detail of another page

the lower centre with the bees issuing forth in search of the flowers. This im-
age is surmounted by the royal arms, while the dedication is introduced by a
huge drop capital D (Fig. 4).
This is backed by the arms of the dedicatee, William Cecil, Lord Burgh-
ley.21 Similar drop capitals are used to introduce each letter as well. The pres-
ence of the beehive image indicates that this title border was designed and cut
specifically for this work – a considerable investment. It is also likely to be an
allusion to the familiar image of the student as an industrious bee gathering
honey from a variety of flowers.
The text itself employs black letter, roman, and italic in each entry, facili-
tating searching by making the various elements of the entry easy to distin-

21
See R.B. MCKERROW and F.S. FERGUSON, Title-page Borders used in England &
Scotland 1485-1640 (London, 1932), No. 140. This title-border was subsequently used for other
works, a common practice.
Seeing is Reading 213

guish. It also makes use of such signs as asterisks and pilcrows, as well as
considerable variations in text size, particularly in the main lemmas. Pilcrows
are used to introduce main entries, while asterisks mark out the individual
meanings and applications of a term.
Some features are immediately obvious, such as the employment of double
rules to enclose the text, which is in two columns, and a header indicating the
position in the alphabet for that page. The Alvearie also incorporates a sophisti-
cated numbered indexing system which allows the reader to navigate from the
index to a term in a definition; thus canthus in the index refers the reader to
“s.957” (that is, entry 957 in the letter S), where we find it as the gloss for
strake, the iron ring around a carriage wheel. While many headwords are not
the first in the entry, often being preceded by an article, larger point size clear-
ly distinguishes them. Where there is a French gloss, this is in italics. The
index numbers, which restart with each new letter, appear in the marginal rules.
The asterisks introduce a gallimaufry of minor elements deriving from the
headword, including polysemous senses, particular phrases usually determined
by the necessity of explaining the Latin source rather than the English, what
the OED would call combining forms, idioms, and various derived grammatical
forms. Much of this suggests that the origin of the material for the dictionary
was Latin texts, as Baret mentions in the preface to the reader in explaining its
origin in the collecting activities of his own students.
The degree of care and sophistication taken by Baret and Denham is cer-
tainly not normal in an English production, and they appear to have taken a
number of conscious decisions about how the entries were to appear rather than
relying on established printing-house custom for the production of dictionaries
in England. The page looks spacious and easy to access, yet contains much
information, carefully graded in style, size, and typeface to facilitate use. To
take an example, the headword chafing is embedded in the phrase a chafing
dishe, in which only chafing is picked out in large bold black letter, the rest
being in a much smaller size. Nevertheless, it seems that the effective lemma
is chafing dish, since the gloss, in roman, is calefactorium. Even by modern
standards, this dictionary is eminently lookable-up.
Elsewhere in Europe in the early seventeenth century, a patently learned
bilingual dictionary of German and Latin by Georg Henisch (1549-1618), pub-
lished in 1616, shows considerable variation in type size, intended to further
accentuate the distinction between the elements of the entries, the use of a
slash to mark the end of the headword and the German synonyms which come
214 R.W. MCCONCHIE

immediately after it, as well as hanging indents and variation between fractura
and roman types.22 German is in fractura and Latin in roman. Only the letters
A-G of this dictionary were ever published. Henisch’s is a relatively scholarly
production, although, unlike early humanist dictionaries, Henisch’s agenda was
to promote German as a language.23
The first English monolingual English dictionary is generally agreed to be
Robert Cawdrey’s modest Table Alphabetical of 1604, a work designed to
assist less learned readers with the more difficult words of English rather than
to describe the English lexicon.24 Cawdrey uses roman for the headwords
which, being English, may be the first use of the vernacular headword in roman
and black letter for the definition. Otherwise, it is a relatively simple two-col-
umn layout, and thus harks back to an earlier graphic system rather than look-
ing forward. By contrast, the 1613 edition adopts the single column layout.25
This is however understandable given the nature of this rather straightforward
little glossary, targeted quite specifically at a vacant niche in the market for
books, which falls well short of the lexicographical standards set during the
sixteenth century by the learned bilingual dictionaries.
Turning briefly to a seventeenth-century specialist dictionary, the antiquar-
ian and legal historian Sir Henry Spelman published the first part of his dictio-
nary of ancient terms used in the law, especially Old English and Latin, in
1626.26 The complete dictionary was published posthumously, Spelman having
died in 1641, with the assistance of a fellow historian, William Dugdale, in
1664.27 The format changes only in minor details between 1626 and later edi-
tions. Entries contain much information, especially glosses and synonymy
along with examples of use (see Fig. 6). They are also very clearly marked with

22
G. HENISCH, Teutsche Sprach und Weiszheit. Thesaurus linguae et sapientiae germanicae
(Augusta Vindelicorum (Augsburg, 1616).
23
For comment on the typographical details of the learned Italian dictionary of the
Accademia della Crusca, published in 1612, see CONSIDINE, Academy Dictionaries, pp. 13, 21.
24
R. CAWDREY, A table alphabeticall, conteyning and teaching the true vvriting, and
vnderstanding of hard vsuall English wordes, borrowed from the Hebrew, Greeke, Latine, or
French. &c. (London, 1604).
25
R. CAWDREY, A table alphabeticall contayning and teaching the true writing and
vnderstanding of hard vsuall English words, borrowed from the Hebrevv, Greeke, Latine, or
French, &c. With the interpretation thereof by plaine English words, gathered for the benefit and
help of all vnskilfull persons (London, 1613).
26
H. SPELMAN, Sir, Glossarium archaiologicum, Part 1 (Londini, 1626).
27
H. SPELMAN, Sir, Glossarium archaiologicum: continens latino-barbara, peregrina,
obsoleta, & novatæ significationis vocabula (Londini, 1664).
Seeing is Reading 215

Fig. 6 Detail of page 83 of Spelman’s Glossarium archaiologicum, 1687.

a pilcrow, and the headword is marked by a closing square bracket at the end.
The headword itself is printed in italics, black letter appearing only within
entries for English, Old English or law French words (occasionally German, as
under ganerbii). Within the matrix Latin text of the definition, Spelman also
uses marginal notes for references. Given the high-quality press work, this
makes Spelman’s Glossarium particularly clean and accessible despite its com-
plexity.28
Cocker’s English Dictionary, a monolingual work, appeared precisely a
century later than Cawdrey’s Table alphabetical, possibly the project of a
group of publishers who simply exploited the name of the mathematical writer,
teacher, and calligrapher Edward Cocker (1631-1675).29 It is probably the least

28
For a brief comment on the typography of the pioneering dictionary of Anglo-Saxon by
William Somner published in 1659, see CONSIDINE, Making, p. 211; W. SOMNER, Dictionarium
Saxonico-Latino-Anglicum voces, phrasesque præcipuas Anglo-Saxonicas, e libris, sive
manuscriptis, sive typis excusis, aliisque monumentis tum publicis tum privatis, magna diligentia
collectas; cum Latina et Anglica vocum interpretatione complectens (Oxonii, 1659).
29
E. COCKER, Cocker’s English dictionary: interpreting the most refined and difficult words
216 R.W. MCCONCHIE

well-regarded of the seventeenth-century monolingual English dictionaries, as

it is generally agreed to have been a piece of hackwork by a conger of book-
sellers. The putative editor was John Hawkins, who, it is claimed on the title
page, worked from Cocker’s manuscript, but he had died in 1692 – it is entirely
possible that neither had an actual hand in this work.30 The layout is straight-
forward. Exploiting the fact that the entries are short on the whole, a three-
column page is used rather than two. The headword is in roman, while the
definition is in italics, and white space is at a minimum. Conventional rather
than hanging indentation is used, and no further differentiation is made, pre-
sumably because this dictionary lacks grammatical and other linguistic infor-
mation, has no illustrative citations, and records no polysemy. In these re-
spects, it follows the hard-word dictionary by J.K., possibly John Kersey, pub-
lished in 1702.31
Kersey, however, who became one of the most important monolingual
lexicographers of the early eighteenth century, gained in skill and sophistica-
tion as his career as a lexicographer progressed. By the time of the 1721 edition
of his Dictionarium Anglo-Britannicum, much has been gained in complexity
to match the content.32 Pages have a double rule at the top, and a single central
rule dividing the two columns. Indentation is still conventional. Black letter in
a larger size is now used for the headwords, and roman for the text – almost as
if harking back to Baret. Derivational information appears as a capital letter in
square brackets for Latin, Greek, and so on, while the same convention is used
for specialised terminology, such as heraldry, geometry, and so on. Key words
in the definition are sometimes italicised as well, not simply confined to the
well-known convention of italicising names.
A successful formula is unlikely to change very rapidly. The second edition
of Thomas Dyche and William Pardon’s A New General English Dictionary,
destined to become one of the most popular of the eighteenth-century English

... By Edward Cocker ... Perused and published from the authors correct copy, by John Hawkins
(London, 1704).
30
J. KERLING, Chaucer in Early English Dictionaries: The Old-Word Tradition in English
Lexicography down to 1721 and Speght’s Chaucer Glossaries (Leiden, 1979), p. 206.
31
J.K., A new English dictionary: or, compleat collection of the most proper and significant
words, commonly used in the language; with a short and clear exposition of difficult words and
terms of art (London, 1702).
32
J. KERSEY, Dictionarium Anglo-Britannicum: or, a general English dictionary,
comprehending a brief, but emphatical and clear explication of all sorts of difficult words
(London, 1721).
Seeing is Reading 217

dictionaries, came out in 1737 (first edition 1735).33 The layout is compact,
employing a smallish font, but with fully capitalised headwords, a feature seen
later in works such as Robert James’ A Medicinal Dictionary (1743).34 There
are hanging indents, and word class is indicated by abbreviations in brackets.
Subject areas are in italics, as are new sub-senses and derived forms of the
headword. This worked well: by the time the eighteenth edition of this very
durable publication appeared in 1781, the only really obvious change in the
typography is that the round brackets have become square, and the indents are
now conventional, not hanging.35
Samuel Johnson’s 1755 dictionary is a model of exemplary dictionary
typography, but it was certainly not the first to reach this standard in England.36
Benjamin Martin’s Lingua Britannica reformata of 1749 (reprinted 1754)
makes use of several of the conventions Johnson used, such as capitalised
headwords with accent marks (orthoepy), bracketed derivation data, indication
of special terms, and polysemy.37 Both also use hanging indents, and Johnson
has further indents for some quotes, a feature found in Estienne 1531. Johnson
uses small capitals for the headwords, and right-justifies the italicised source-
names for his quotations, a practice which neatly separates them from the text
in many cases. Where there are multiple meanings each is placed on a new line
and numbered, again providing a neat visual separator. Martin, who may have
been the model for this arrangement, also puts the number at the position of the
first indent, and leaves a double space before the text, thus making separation
of the various sub-senses particularly clear. Johnson is slightly different, fully
left-justifying the numbers and putting a full-stop after each. This of course
allows for the insertion of more text, but Martin’s arrangement is visually a
little more reader-friendly.

33
T. DYCHE, A new general English dictionary; peculiarly calculated for the use and
improvement of such as are unacquainted with the learned languages.... now finish’d by William
Pardon, Gent. The second edition, with additions (London, 1737).
34
R. JAMES, A medicinal dictionary; including physic, surgery, anatomy, chymistry, and
botany, in all their branches relative to medicine, 2 vols. (London, 1743).
35
T. DYCHE and W. PARDON, A new general English dictionary (London, 1781).
36
S. JOHNSON, A dictionary of the English language; in which the words are deduced from
their originals and illustrated in their different significations by examples from the best writers,
2 vols. (London, 1755).
37
B. MARTIN, Lingua Britannica reformata: or, a New English Dictionary (London, 1749).
218 R.W. MCCONCHIE

Conclusion

This article was not intended as a thorough survey of dictionary typology,

and has not attempted a coherent narrative history of the subject, which would
take a great deal more space. It is simply an introduction based on a few exam-
ples meant to show the range of types and stimulate interest and further re-
search – the hard yards on this subject remain to be done. A detailed examina-
tion of some of these early modern dictionaries reveals a multitude of graphic
techniques intended to convey information. The continental works are well in
advance of the English ones in terms of press work and innovation, while the
monolingual dictionaries are slow to achieve the same standard as the earlier
bilinguals. Dictionaries are also products of the printing industry, which dic-
tates their layout to a considerable degree. Estienne, as a printer himself as well
as a scholar, would have had complete control of the publishing process, and
others in the early sixteenth century worked closely with their printers and / or
publishers, often lodging with them during the process. Later lexicographers
would have had less say.
There is a gradual and halting introduction and acceptance of refinements
in dictionary formatting. Although there are obvious advantages in accessibility
for the reader in the two-column format, it does not seem to have correlated
more with the length of entries than with the size of the book on the whole, and
we see it used inconsistently. The use of symbols such as manicules and pil-
crows comes and goes. The distinction between roman for Latin and black
letter in England or fractura in Germany for vernaculars was established early
and persisted long. The widespread use of the contrast between roman and
italic for such purposes comes later, although it is anticipated by Spelman.
Although there is a natural tendency to seek the most obvious advances in
graphic presentation in the most learned works, in a way it is just as instructive
to look at those appearing in the least sophisticated dictionaries, on the assump-
tion that if such works find them acceptable, all will do so, and the innovation
in question will have become a permanent feature.
 Whose Letters Are They Anyway?
Addressing the Issue of Scribal Writing in
Bess of Hardwick’s Early Modern English Letters

I.J. MARCUS

Introduction

lthough often associated with the medieval period, scribes were still

A widely employed between 1500 and 1700, and grammatical, lexical,

orthographic and palaeographical variation was still widespread. This
chapter presents a scribal profiling technique that was designed to distinguish
between groups of letters written by different scribes in a corpus of Early Mod-
ern English (henceforth EModE) personal correspondence to and from Eliza-
beth Talbot, Countess of Shrewsbury, c. 1527-1608, also known as Bess of
Hardwick and henceforth referred to in this chapter as Bess.1 It argues that
scribal composition had a greater influence upon the language used within
Early Modern English letters than has previously been assumed, and suggests
1
To date, 241 extant letters to and from Bess have been identified. These letters are located
in nineteen different repositories across the UK and USA and have been digitised in a recent web
edition, to be found at <www.bessofhardwick.org>.

......................................................................................................................................
Verbal and Visual Communication in Early English Texts, ed. Matti PEIKOLA, Aleksi
MÄKILÄHDE, Hanna SALMI, Mari-Liisa VARILA, and Janne SKAFFARI, Utrecht Studies in Medieval
Literacy, 37 (Turnhout: Brepols, 2017), pp. 219-250.

DOI 10.1484/M.USML-EB.5.114137
220 I.J. MARCUS

that scribal profiling can be a way to take such influence into account before
and during linguistic analyses.
Previous studies on the influence of scribes on epistolary language have
suggested that manuscript letters from this period were copied down verbatim.2
Terttu Nevalainen and Helena Raumolin-Brunberg make this claim in relation
to the Paston letters, highlighting the use and distribution of the relative pro-
noun ‘which’ to demonstrate the point.3 Meanwhile Alexander Bergs also looks
at morpho-syntactic variation in the Paston letters, and concludes that “scribes
may have had some influence on morpho-syntactic items, but in general took
down faithfully what was dictated to them”.4 Finally, Muriel St. Clare Byrne,
who looks at the Lisle letters, which date from 1533 and 1540, notes a “consis-
tency of style” in Honor Lisle’s letters, which does not change depending on
which scribe she used. 5
However, more recent work on the distribution and use of linguistic fea-
tures of spoken communication such as discourse connectives in the letters of
Bess of Hardwick has highlighted clear differences in the use of these features
in Elizabeth’s scribal and holograph letters.6 This research is supported by that
of Alison Truelove on Elizabeth Stonor’s letters, and Graham Williams on the
letters of the Thynne family, dating from 1575-1611.7 Williams notes that in
addition to orthographic variation, which other scholars within the fields of
palaeography and historical linguistics have observed, scribal composition in

2
“Copied down verbatim” also covers transmission through dictation.
3
T. NEVALAINEN, and H. RAUMOLIN-BRUNBERG, Historical Sociolinguistics (London,
2003), pp. 197-198.
4
A. BERGS, Social Networks and Historical Sociolinguistics: Studies in Morphosyntactic
Variation in the Paston Letters (1421-1503) (Berlin, 2005), p. 80.
5
The Lisle Letters, ed. M. ST. CLARE-BYRNE, 6 vols. (Chicago, 1981), ,<vol.?>, pp. 228-
230.
6
See I.J. MARCUS, The Linguistics of Spoken Communication in Early Modern English
Writing: Exploring Bess of Hardwick’s Manuscript Letters (London, forthcoming 2017).
7
A. TRUELOVE, “Commanding communications: The fifteenth-century letters of the Stonor
women”, in: Early Modern Women’s Letter Writing, 1450-1700, ed. J. DAYBELL (Basingstoke,
2001), pp. 42-58; G. WILLIAMS, “The language of early modern letters: A reader’s guide”, in:
Bess of Hardwick’s Letters: The Complete Correspondence, c.1550-1608, ed. A. WIGGINS, A.
BRYSON, D. STARZA SMITH, A. TIMMERMANN, and G. WILLIAMS (Glasgow, 2013), web develop-
ment by K. ROGERS, University of Sheffield Humanities Research Institute (25 May 2016),
<https://www.bessofhardwick.org/background.jsp?id=174>; G. WILLIAMS, “‘yr Scribe Can
proove no nessecarye Consiquence for you’?: The social and linguistic implications of Joan
Thynne’s using a scribe in letters to her son, 1607-1611”, in: Women and Writing, c.1340-c.1650:
The Domestication of Print Culture, ed. P. HARDMAN and A. LAWRENCE-MATHERS (Woodbridge,
2010), pp. 131-145.
Whose Letters Are They Anyway? 221

the Thynne correspondence corpus produces clear differences between authori-

al and scribal productions with respect to anaphoric legalisms.8 It is also indi-
rectly supported by Kathleen L. Doty, who has looked at the use of contextual
commentary and evaluative adjectives or adverbs in the records of examina-
tions that survive from the Salem witchcraft trials of 1692 and 1693 and has
found evidence of scribal variation.9
Collectively, these studies have significant implications for corpus-based
historical linguistic research that does not distinguish between holograph and
scribal material. They suggest very strongly that we need to take the holograph
versus scribal variable into account. This chapter argues that scribal profiling
provides researchers with a way of taking this variable into account by allow-
ing them to identify the holograph handwriting of the ‘signer’, i.e. the person
whose signature is on the letter, and the handwriting of their scribe(s). Follow-
ing the introduction, it will briefly describe the corpus of personal correspon-
dence used, and the scribal profiling methodology that was devised in relation
to it. The chapter will then discuss how this particular profiling method can be
applied in relation to other relatively small correspondence corpora, before
discussing how it could be potentially applied to studies that use much larger
corpora in future research.

Potential Involvement of Scribes in the Compositional process of EModE

Letters

The potential involvement of scribes in the compositional process of

EModE) is an important factor for historical linguists to consider. A scribe,
also known as an amanuensis or secretary, is a writer who copies a document
for someone else. Anyone with a relatively high level of writing proficiency
could act as a scribe, including family members and household servants, al-
though towards the end of the EModE period there was an increasing number
of professional scribes. These professional scribes would potentially have had
a degree of training in how to write a variety of scripts. As defined by Malcolm
8
See WILLIAMS, “The language of early modern letters”, p. 57, on “Anaphoric legalisms”
that are the “compound adverbs and other anaphoric reference terms (e.g. thereto, the said, etc.)”
that occurred more frequently in legal statutes than in other types of text during the EModE
period.
9
K.L. DOTY, “Telling tales: The role of scribes in constructing the discourse of the Salem
witchcraft trials”, Journal of Historical Pragmatics 8.1 (2007), pp. 25-41.
222 I.J. MARCUS

Parkes, “a script is the model which the scribe has in his mind’s eye when he
writes”.10 The two main models of relevance for these letters are Secretary (an
offshoot of the court hands of the beginning of the sixteenth century (early-late
Tudor, Jacobean) and italic (created in Italy c. 1400 and popular with English
humanists from the early sixteenth century on).
Furthermore, each scribe had their own unique ‘hand’. A hand is a repre-
sentation of a particular script that the writer actually puts down on the page.11
A hand will bear a generic resemblance to the writing of other scribes adopting
the same script, but is also likely to bear the personal idiosyncrasies and distin-
guishing features of the individual writer’s handwriting. There are various
ways in which messages could have been transmitted to scribes. It was com-
mon for people to orally dictate messages verbatim to their scribes or secretar-
ies. However, other potential methods of transmission include oral outlines or
written notes detailing what the originator of the message wanted the written
missive to contain. The potential involvement of scribes in the compositional
process of an individual’s letters is a major evidential issue that has to be ad-
dressed by any researcher who is using an individual’s early modern manu-
script letters as sources of their language use. In relation to Bess’s letters for
example, it is not enough to write about ‘the letters of Bess of Hardwick’; it is
essential to classify letters from Bess as holograph, scribal, or something in
between.
However, before scribal elements can be identified and any potential scri-
bal influence over grammatical and lexical choices can be taken into account,
it is essential to unpack the concepts of ‘scribal’ and ‘holograph’. Common
consensus has it that a scribal letter is a letter copied by a scribe on behalf of
the person who conceives of and signs the letter, whereas a holograph letter is
one in which the body of the letter is copied by the person who conceives of
and signs the letter.12 But it is not enough to say that the EModE manuscript
letters from Bess are either scribal or holograph, because the situation is more
complex than that. An EModE manuscript letter can be scribal or holograph to
different degrees. An EModE letter is made up of different parts, and each of
these parts can be either scribal or holograph, and can be either present or not
present.

10
M.B. PARKES, English Cursive Book Hands, 1250-1500 (London, 1979), p. XXVI.
11
PARKES, English Cursive Book Hands, p. XXVI.
12
Some scholars use the word ‘autograph’ rather than ‘holograph’, but ‘holograph’ is the
term adopted in this chapter.
Whose Letters Are They Anyway? 223

The different parts of an EModE manuscript letter include: address, body,

endorsement, margin, marginalia, postscript, signature, subscription, and super-
scription.13 These different elements constitute the ‘layout’ of a letter, i.e. the
way the text of a letter and accompanying features such as margins are physi-
cally arranged on the page. Each of these elements can be either scribal or
holograph, and can either be present or not present. When it comes to the cate-
gory of ‘scribal letters’ in particular, there are a number of possible variations.
The address and body might be in one Secretary scribal hand, while the sub-
scription and signature might be in a second, italic scribal hand, for example.
Another possible combination might be that the address and body are in one
italic scribal hand, whilst the subscription and signature are in the signer’s
hand. The body, subscription and signature are relatively constant elements of
early modern letters, whilst the presence of addresses, postscripts, superscrip-
tions and endorsements is more variable. Superscriptions are sometimes written
by the scribe who writes the body, although they are sometimes copied by the
signer. Endorsements are usually added to a letter by a clerk or scribe at the
receiving end, although some endorsements are added at a much later date, as
part of the archival process. There can also be more than one endorsement. In
this chapter, a letter is classed as ‘scribal’ if a minimum of its body has been
written by a scribe.14
There is also the question of how many scribal hands there are in a single
letter, and whether these hands are representations of the same script. Further-
more, there is a possibility that rather than there being two separate scribes
writing two separate scripts, there is one scribe who copies the whole letter, but
switches scripts in the middle of the letter as in ID 158 (see Fig. 1).
In this image of a letter from Bess to Henry Bebington and Alexander
Whyte dating from 1565, it is possible to see that the scribe in question writes
the body of the letter in a Secretary hand, before completing the subscription
and signature in an italic hand. Due to the fact that an EModE manuscript letter
can be scribal or holograph to different degrees, it is important, whenever pos-

13
For more information about these parts, see Women Letter-Writers in Tudor England, ed.
J. DAYBELL (Oxford, 2006) and WILLIAMS, “The language of early modern letters”.
14
Bess has a “scribal profile” like her scribes, but in this instance ‘scribal’ refers to the fact
that Bess is thought to have copied these letters in her own hand.
224 I.J. MARCUS

Fig. 1. Image of letter ID 158

sible, to examine each text individually when undertaking qualitative, data-

sensitive analyses of this kind.
Another evidential issue related to the use of scribes to copy letters is the
difficulty of distinguishing between scribal and holograph writing, and of dis-
tinguishing between individual scribal hands, in the Bess of Hardwick letter
corpus as well as more generally with EModE manuscripts. For example, Bess
consistently writes in a distinctive italic hand, but there are a number of letters
in which a remarkably similar italic hand is used. The question of whether or
not letters such as these were copied by Bess herself needs to be addressed in
an empirically rigorous manner. For example, it needs to be established
whether the similar italic hand in question contains some distinctive features of
its own. Furthermore, a number of Bess’s scribes employ Secretary script. As
a result of this, their handwriting looks very similar, because they are all at-
tempting to conform to the same ideal model of handwriting. Therefore, some
kind of systematic method is needed to identify diagnostic palaeographic and
Whose Letters Are They Anyway? 225

orthographic features characteristic of individual hands that are manifestations

of the same script.

Scribal Profiling Methodology

Now that the introduction has outlined the scribal issue, this chapter will
focus more particularly on the group of letters from Bess used for the purposes
of linguistic analysis in this chapter. In order to develop a fuller and more nu-
anced appreciation of the linguistic data, it is important, as far as possible, to
track the scribes Bess regularly used and to identify Bess’s holograph hand-
writing. This process needs to be carried out before any linguistic analysis can
begin. The scribal profiling methodology described in this chapter is ideal for
a corpus that contains both scribal and holograph material, because it allows
the researcher to identify both the holograph and the scribal handwriting. It is
best suited to differentiating between holograph and scribal letters from the
same individual, i.e. the person whose signature is at the bottom of each letter.
It could, however, potentially be applied to a group of letters from different
individuals, or to a set of corresponding letters, i.e. sent letters and their re-
sponses. In the rest of the chapter individual letters are referred to by the iden-
tification number system that was implemented by the AHRC Letters of Bess of
Hardwick Project team. For example, the letter that was assigned the identifica-
tion number 179 is referred to in the chapter as ‘ID 179’.15

Stage 1

Stage 1 of the scribal profiling analysis involves deciding which letters to

include in the initial data set. The ‘initial data set’ is a study-specific term that
refers to the group of letters from Bess that will be broken down into separate
categories at Stage 2 of the scribal profiling analysis. Once this decision has
been made it is possible to carry out Stage 2 of the analysis, i.e. the formulation
of a hypothesis about the various hands found within these letters. To exem-
plify using the Bess corpus, there are 88 letters ‘from’ (as opposed to ‘to’) Bess
in the corpus. This number is inclusive of four potential drafts, eight jointly

15
See <www.bessofhardwick.org> for more information.
226 I.J. MARCUS

sent, co-signed letters and six later scribal copies.16 Despite the relatively large
number of different hands found within them, it was decided that the majority
of these 88 letters from and signed by Bess would be included in the initial data
set. For example, it was decided that letters thought to be potential drafts rather
than sent copies (namely ID 115, 143, 175, and 177) would be included in the
initial data set.
The jointly sent, co-signed letters and the later copies were deliberately
omitted at this stage. The eight jointly sent letters, co-signed by both Bess and
her fourth husband Shrewsbury, were excluded because the bodies of these
letters are all copied either in what was thought to be Shrewsbury’s holograph
hand or in a scribal hand. These letters were: letter ID 115, 188, 189, 190, 193,
194, 195, and 197. The six later scribal copies of letters known to be from Bess
were also omitted. After these omissions, 74 letters remained within the initial
data set.

Stage 2

The second stage of the process involves splitting up the letters contained
within the initial data set into categories that separate out the letters according
to scribal structure. This is done so that the researcher can see how many hands
are thought to be contained within each letter in the dataset, and also what kind
of scripts are being used across the set. In order to do this, it is necessary to
become acquainted with all of the letters within the initial data set, as well as
with the various different scribal hands present within these letters. A close
acquaintance with the 74 letters taken through to the second stage from the
Bess corpus was developed through a comprehensive examination and re-ex-
amination of the letters. Each letter within the initial data set was read and re-
read several times. This was done until each letter, and each of the hands pres-
ent within the initial data set as a whole, became relatively familiar and recog-
nisable. The letters were then split into thirteen separate categories. These
categories, and the letters that belong to them (represented by their ID num-
bers), are presented below.

16
Nota bene: ID 115 is both a potential draft and a jointly sent letter.
Whose Letters Are They Anyway? 227

1) Address, body, subscription, signature, (postscript): in one italic hand.

ID 099, 100, 101, 107, 109, 110, 111, 112, 116, 120, 122, 123, 177, 178, 182, 183,
184, 186, 198, 200, 233

2) Address, body, subscription: in one italic scribal hand, signature in another italic
hand.
ID 001, 124, 125, 126, 140, 145, 148, 149, 150, 152, 153, 156, 159, 160, 163, 179,
180, 181, 185, 187, 204, 210, 229, 230, 231, 234

3) Address, body: in one italic scribal hand, subscription and signature: in another
italic hand.
ID 121, 202

4) Address, body: in one italic scribal hand, subscription in a second italic hand,
signature in a third italic hand.
ID 128

5) Address, body, subscription: in one Secretary scribal hand, signature in an italic

hand.
ID 157, 102

6) Address, body, subscription: in one Secretary scribal hand, signature and postscript
in an italic hand.
ID 002

7) Address, body: in one Secretary scribal hand, subscription: in a second italic scri-
bal hand, signature in a third italic hand.
ID 127, 129, 130, 131, 134, 135, 139, 161

8) Address, body: in one Secretary scribal hand, subscription and signature: in a sec-
ond italic scribal hand.
ID 103, 104, 105, 144, 158, 162

9) Address, body, postscript: in one Secretary scribal hand, subscription: in a second

italic scribal hand, signature in third italic hand.
ID 132

10) Address, body: in one Secretary scribal hand, subscription, signature and post-
script: in an italic hand.
ID 113
228 I.J. MARCUS

11) Body: in one Secretary scribal hand, subscription, signature and postscript: in an
italic hand.
ID 114

12) Body in one italic scribal hand, no address, subscription or signature (therefore a
possible draft)
ID 143, 176

13) Body in one italic scribal hand, no address, subscription or signature (therefore a
possible draft), but with a postscript in an italic hand.
ID 115, 175

A working hypothesis regarding the individual scribal hands across the

initial data set can then be formulated.17 The working hypothesis formulated at
this stage of the analysis in relation to the letters in the Bess corpus was that
six major scribes had contributed to the composition of the 74 letters in the
initial data set. These ‘major’ scribes had copied down the bodies of at least
two letters from Bess in the initial data set. It was further hypothesised that at
least thirteen other, more minor scribes had contributed to the composition of
the 74 letters in the initial data set. These minor scribes had only copied down
the body of a single letter from Bess. Some letters within the larger group of
letters thought to have been copied by these minor scribes were placed into
loose smaller groups, based on handwriting and the place of composition.
Seven letters were written from Sheffield, and there were five possibly distinct
scribal hands in this group. ID 144 and ID 162 appear to be in the same hand, as
do ID 002 and 103. Three other letters from Sheffield, ID 104, 105 and 114, are
in miscellaneous hands. So this group of minor scribes was loosely donned the
Sheffield scribes. There are also three letters in different hands written from
Chatsworth, so this group became known as the Chatsworth scribes.

Stage 3

Stage 3 of the scribal profiling analysis involves making a decision, based

on the working hypothesis made in Stage 2, about which of the letters included
within the initial data set to include within the eventual data set. The ‘eventual

17
The methodology assumes that one scribe produces one scribal hand, but could be easily
adapted so that it allows for attributing more than one scribal hand to a given scribe.
Whose Letters Are They Anyway? 229

data set’ is a study-specific term which refers to the data set analysed in the
scribal profiles completed. It involves the completion of the scribal profiles
themselves. In the case of this corpus of letters from Bess of Hardwick, it was
decided that the only hands that would be profiled would be the six ‘major’
scribal hands. This decision was made because it was considered desirable to
have relatively large word counts for each writer. Therefore, the 15 letters
whose bodies were thought to have been copied in ‘minor’ scribal hands were
omitted from the eventual data set. This left a total of 59 letters from Bess in
the eventual data set.
So, whilst the initial data set contained 74 letters, the eventual data set
contained 59 letters. It was hypothesised that 21 of these were copied by a
scribe dubbed Scribe 1, and that the remaining 38 were copied in the five other
‘major’ scribal hands that had been provisionally identified at Stage 2. Each of
these five ‘major’ scribal hands were assigned to individual scribes, who were
labelled Scribe 2, Scribe 3, Scribe 4, Scribe 5 and Scribe 6. The total word
counts for each of the writers, and the overall word count for the eventual data
set as a whole are as follows:

Scribe 1: 5996 words.

Scribe 2: 7109 words.
Scribe 3: 661 words.
Scribe 4: 3498 words.
Scribe 5: 526 words.
Scribe 6: 1193 words.
Overall word count: 18983 words.

It is worth briefly mentioning the fact that some letters are included in
more than one scribal profile, because more than one scribe is thought to have
contributed to these particular letters. So, Scribe 1’s profile includes 21 letters,
Scribe 2’s includes 20, Scribe 3’s includes 13, Scribe 4’s includes 8, Scribe 5’s
includes 3, and Scribe 6’s also includes 3. However, due to the fact that it was
hypothesised that Scribe 3 potentially contributed subscriptions to nine letters
in which the address and body were thought to have been potentially copied by
either Scribe 4 and Scribe 5, these nine letters are included in three different
scribal profiles; i.e. the profiles of Scribes 3, 4 and 5. This kind of overlap is
not surprising when profiling an EModE letter corpus, because it was common
for more than one scribe to contribute to individual EModE letters.
230 I.J. MARCUS

The dates of some of the letters in the eventual data set are unknown. How-
ever, of the letters that are dated, the earliest one is ID 099, which dates from
1553. The latest is ID 179, which dates from 1607, the year before Bess’s death
in 1608. The range of topics and correspondents represented by the letters
included in the eventual data set is various and wide-ranging. For example, the
recipients of the 21 letters provisionally thought to be copied by Scribe 1 come
from across the social spectrum of early modern English society. They include
monarchs (for example Queen Elizabeth I), aristocrats (for example Robert
Dudley, Earl of Leicester), courtiers (for example Francis Walsingham and
William Cecil, first Baron Burghley), members of parliament (for example
John Thynne), family members (for example George Talbot, Earl of Shrews-
bury, henceforth referred to as Shrewsbury) and servants (for example Francis
Whitfield). These letters also cover a range of topics and perform a variety of
communicative functions. For example, among the group of letters provision-
ally thought to be copied by Scribe 1 there are letters of petition, instruction,
and reportage and letters of request, as well as letters of miscellaneous content.
After decisions have been made about which letters to include in the even-
tual data set, Stage 3 involves the production of scribal profiles in order to test
and refine the hypotheses made at Stage 2. The scribal profile template devised
has both palaeographical and linguistic elements (the linguistic element being
the inclusion of a spelling questionnaire). Over the past thirty years, the meth-
odology of scribal profiling has been developed in particular by Middle Eng-
lish palaeographers such Malcolm Parkes and Ian Doyle and historical dialect-
ologists such as Angus McIntosh, Michael Samuels, and Michael Benskin, who
created A Linguistic Atlas of Late Mediaeval English, or LALME, with the assis-
tance of Margaret Laing and Keith Williamson.18 Other scholars such as Simon
Horobin, who discusses the benefits of combining spelling questionnaires with
palaeographic profiling, have contributed valuable work on scribal attribution
in late medieval English texts.19
However, these scholars work on medieval as opposed to early modern
texts. This chapter applies and adapts the techniques developed by these Mid-
dle English scholars to early modern textual material, specifically the sixteenth

18
A.I. DOYLE and M.B. PARKES, ”A palaeographical introduction”, in: A Variorum Edition
of the Works of Geoffrey Chaucer, 1, ed. P.G. RUGGIERS and D.C. BAKER (Folkestone, 1979), pp.
XIX-XLIV; A. MCINTOSH, M.L. SAMUELS and M. BENSKIN, A Linguistic Atlas of Late Mediaeval
English, 4 vols. (Aberdeen, 1986).
19
S. HOROBIN, “The criteria for scribal attribution: Dublin, Trinity College MS 244 recon-
sidered”, Review of English Studies 60 (2009), pp. 371-381.
Whose Letters Are They Anyway? 231

century private letters of Bess of Hardwick. Although the scribal profiling

methodologies and typologies developed by scholars such as McIntosh have
yet to be taken up by a large number of EModE scholars, there is an increasing
amount of sensitivity towards the compositional complexities of the texts under
investigation within the study of early modern language and literature as a
whole.20 From a literary perspective, Henry Woudhuysen, in a chapter on
Queen Elizabeth I’s handwriting, writes that the queen’s handwriting shows
“the considerable importance of bearing in mind or trying to reconstruct the
circumstances under which a document of any kind was written”.21 Further-
more, this approach has been endorsed both by scholars working on medieval
manuscripts and scholars such as Risto Hiltunen and Matti Peikola, working on
the later, 1692 Salem witchcraft trial documents, who note that “scribal pro-
files can be reconstructed by means of linguistic and palaeographic analysis”.22
The scribal profiling method identifies consistencies and inconsistencies across
the group of letters in each profile in order to single out any letters that are not
by the scribe. Below is a scribal profile template. Each of the individual sec-
tions of the profile will be discussed in more detail in the remainder of this
section.
As can be seen from the template, there are seven core elements of the
profile: the use of abbreviation, capitalisation, and punctuation, graphemes
(including the scribe’s inventory of graphic representations of these gra-
phemes), the ductus characteristic of the hand, and the penmanship of the
writer in question. These six parts constitute the palaeographical component of
the profile. The spelling questionnaire is the sole linguistic component.

20
See for example A. JUCKER and P. PAHTA, Communicating Early English Manuscripts
(Cambridge, 2011).
21
H.R. WOUDHUYSEN,“The Queen’s own hand: A preliminary account”, in: Elizabeth I and
the Culture of Writing, ed. P. BEAL and G. IOPPOLO (London, 2007), pp. 1-27.
22
R. HILTUNEN and M. PEIKOLA, “Trial discourse and manuscript context: Scribal profiles
in the Salem witchcraft records”, Journal of Historical Pragmatics 8.1 (2007), pp. 43-68, at p.
1<check>. For other work that addresses issues of scribal identification in relation to early
modern handwriting, see, e.g., J. GIBSON, “The Queen’s two hands”, in: Representations of
Elizabeth I in Early Modern Culture, ed. A. PETRINA and L. TOSI (New York, 2011), pp. 47-65.
232 I.J. MARCUS

Scribal Profile Template

Scribe 1
Letters:

Archive reference, date, place of composition, recipient and content:

Abbreviations:

Punctuation:

Capitalisation:

Ductus:

Penmanship:

Graphemes:

Spelling Questionnaire and summary:

Conclusion:

Image of sample letter:

Letters

The seven core elements referred to above are preceded by contextual

information about the letters included in the profile, namely a short, one-line
description of the scribal structure of each individual letter. Below is an exam-
ple from Scribe 1’s profile:

Letters: 21

Address, body, subscription, signature, postscript: in one italic hand:

ID 099, 100, 101, 107, 109, 110, 111, 112, 116 120, 122, 123, 177, 178, 182, 183,
184, 186, 198, 200, 233
Whose Letters Are They Anyway? 233

The list above means that in each of the 21 letters listed, all of these struc-
tural elements referred to, for example the body or main text of the letter, are
written in what is thought to be the same scribal hand.

Archive Reference, Date, Place of Composition, Recipient and Content

The information about the scribal structure of each letter is followed by

contextual information relating to place and date of composition, recipient, and
content of each epistolary text. To take another example from Scribe 1’s pro-
file, in relation to letter ID 099:

ID 099: Folger, Cavendish-Talbot MSS, X.d.428 (82). 14 November 1553. London.

To Francis Whitfield. Letter instructing Whitfield to look after things at Chatsworth
until her aunt Marcella Linacre arrives, and to cause Bronshawe to look to things
at Pentrich. Lists specific tasks for Whitfield to complete.

Abbreviations

The purpose of this section is to identify similarities across the letters in

relation to abbreviation practice. The key question here is: are there particular
words that are continually being abbreviated and, if so, does this suggest that
these letters have been written by the same scribe?
Key terminology used in this section includes ‘grapheme’, ‘allograph’, and
‘macron’. ‘Grapheme’ refers to the smallest unit of a writing system, which in
relation to the Greco-Latin alphabet usually refers to individual letter forms in
the alphabet. ‘Allograph’ refers to a variant form of a grapheme which is either
in complementary distribution or free variation with another form of the same
grapheme, as in f and F. A ‘macron’ is a diacritical mark placed above a letter
form, usually a vowel. The method involved in completing the ‘Abbreviations’
section is a qualitative, close reading one. It involves reading through each of
the letters in question and noting which words, if any, are abbreviated. All the
words that are frequently abbreviated are listed, followed by examples of how
they are abbreviated, in addition to how many occurrences there are in each
letter and which lines they appear on. So, for example, Scribe 1 abbreviates a
number of words, including ‘forthwith’. This word is therefore listed in the
profile as:
234 I.J. MARCUS

FORTHWITH: <fourthewt>. 1 occurrence – line 17 of ID 099.

Once the different abbreviations have been listed, there is an ‘Abbreviation

Summary’, which helps the researcher to assess any overall patterns in relation
to particular abbreviations, such as the most commonly abbreviated word. It
also allows an assessment of whether elements of different scripts have been
incorporated into the dominant script. For example, in letter ID 107, Scribe 1
uses two abbreviations, <pseruacyon> for ‘preservation’ and <psent> for ‘pres-
ent’, abbreviations which are features of Secretary, rather than italic script.
Abbreviating words that begin ‘pre’, ‘per’ and ‘par’ by replacing the ‘pre’,
‘per’ or ‘par’ with corresponding brevigraphs based on the letter p was com-
mon practice when writing Secretary script. Scribe 1 therefore occasionally
incorporates elements of Secretary script into the dominant italic script.

Punctuation

Like abbreviation and capitalisation, the use of punctuation can be a useful

feature to investigate when conducting scribal identification. And as with these
other sections, the primary reason for conducting a survey of the punctuation
used is to identify patterns, i.e. similarities and / or differences in the group of
letters under consideration. The researcher may choose to use the medieval
terms for the different punctuation marks, such as punctus for the full stop
mark. They may also find it useful to use terms such as medial, to refer to
marks that occur above the line. The punctuation section is similar in structure
to the ‘Abbreviations’ section of the profile in that it consists of a list of the
punctuation marks used in the group of letters in question followed by a sum-
mary that assesses any overall patterns. Below are examples that refer to the
use of two punctuation marks in the group of letters thought to be copied by the
scribe dubbed Scribe 1 in the Bess corpus, a colon and a semi-colon:

<:> - ID 112, 183

<;> - ID 183

The punctuation summary allows for an assessment of which punctuation

marks occur most frequently, i.e. which are found in a larger number of letters
than the other marks. It also allows for a close analysis of the employment of
particular punctuation marks. For example, in relation to Scribe 1, some of the
Whose Letters Are They Anyway? 235

seventeen letters thought to be copied by this scribe will employ one particular
mark repeatedly. This can be seen in ID 110, in which there are several slashes,
or single virgules used. Whereas in others, such as ID 183, there are several
different types of punctuation mark. The summary is also a chance for an as-
sessment of the stability of the punctuation practice across a given set of let-
ters. So in relation to Scribe 1, it is possible to say that there is a relatively high
level of variation, both in terms of the amount and type of punctuation em-
ployed throughout the letters.

Capitalisation

The capitalisation section of the profile differs slightly from the abbrevia-
tion and punctuation sections because it only consists of a descriptive summary
of the use of capitalisation, rather than a comprehensive list of distinguishing
marks followed by a summary. However, its purpose is similar in that its aim
is to help establish how systematic the scribe’s capitalisation practice is, and /
or whether there is anything unique or unusual about it. As with punctuation,
the researcher may choose to use the older terms related to capitalisation, such
as majuscule for capital letter, or they may prefer to stick to modern terminol-
ogy.
Questions that can be asked when investigating the use of capitalisation
include: is capitalisation restricted to words in sentence-initial position? Are
proper nouns and / or address forms capitalised? Or is there a practice of capi-
talising words beginning with particular letter forms? For example, in the case
of Scribe 1, all words beginning with the letters l, c, and i are capitalised.
Words beginning with g, such as “Gracyous” in ID 120), are also occasionally
capitalised. Collectively, these queries form the basis of a description of any
patterns present in the group of letters being investigated, which in turn can
potentially provide clues about the capitalisation practice of the scribe in ques-
tion.
236 I.J. MARCUS

Ductus

As regards the palaeographical element of the profile template, it was de-

cided that the ductus of the hand in question as well as the penmanship of the
writer should both be looked at in addition to abbreviations, punctuation,
capitalisation practice, and letterforms, or as they are referred to in this chapter,
graphemes. The ductus (or ‘duct’), of a hand refers to what John describes as
“the overall, general ‘nature’ of the production of a given script, defined in
terms of the ‘number, sequence, and direction of the strokes used in forming
each letter of the script’s alphabet”.23 Ductus is more related to the flow of the
writing, as opposed to the static graphemes themselves. Some of the key termi-
nology related to ductus includes ‘stroke’, which means “a single sweep of the
pen; maybe a single ‘movement’ (that is, with no change of direction)”.24
Simon Horobin, who works on late Middle English texts, emphasises that
when undertaking scribal attribution, one should not overlook the ductus of the
hand. Through questioning Alan J. Fletcher’s attribution of the copying of a
manuscript known as MS Dublin, Trinity College, 244 to professional London
scribe Adam Pynkhurst, Horobin demonstrated why it is not just enough to
focus on graphemes in a palaeographical profile. To quote,

by basing his analysis on individual letter forms rather than on ductus and aspect,
Fletcher fails to notice the considerable disparity between the characteristics of
Pynkhurst’s hand and that of Trinity 244.25

The method of compiling this part of the profile involves observing the
handwriting in question and describing its key characteristics, in addition to
assessing the overall stability of the ductus. The first thing to establish is which
script the scribe’s hand maps onto. It is worth asking whether the scribe is
writing in italic script, Secretary script, or whether his hand prototypically
contains aspects of both scripts. For example, Scribe 1’s hand is an italic hand,
but there are three letters, namely ID 107, 109 and 123, where there are several
abbreviations and graphemes more typical of Secretary than italic script (such

23
J.J. JOHN, “Latin paleography”, in: Medieval Studies: An Introduction, ed. J.M. POWELL
(Syracuse, 1992), pp. 1-68, at p. 8.
24
JOHN, “Latin paleography”, p. 8. For an introduction to the terminology used in relation
to duct in palaeography see PARKES, English Cursive Book Hands, p. XXVI.
25
HOROBIN, “The criteria for scribal attribution”, p. 373. See also DOYLE and PARKES, “A
palaeographical introduction”, p. XXXV.
Whose Letters Are They Anyway? 237

as a Secretary s in terminal position). The reason for these anomalies could be

that Scribe 1 was copying out earlier scribal drafts of these three letters in his
or her own hand.
Once the script (or mixture of scripts) has been established, it is useful to
assess the overall look of the hand. Are the graphemes large or small, distinc-
tive, or uniform, confident or shaky, legible, or illegible? Is the hand functional
or a stylish, calligraphic hand, exemplified by a use of flourishes? Is it cursive,
free, and sloped? For example, in Scribe 1’s distinctive italic hand, long verti-
cal strokes frequently appear curved if executed quickly, and his or her
descenders are characteristically curved. There are few pen lifts, although
initial y, m, g, and f are not joined up.
Other questions to ask include: is the handwriting compressed? Does the
actual size of the handwriting vary from letter to letter, or does the size, slant
and spread of the graphs vary? And what does the surviving body of handwrit-
ing contained within the set of letters under consideration say, if anything,
about the overall stability of the hand’s ductus? To refer to Scribe 1’s profile
again, it was established that although several of the letter shapes are not that
stable, in the sense that Scribe 1 forms them in a different way, this does not
detract from the scribe’s ability to write well-written, legible letters.

Penmanship

The reason why penmanship is included in the profile is because it is simi-

lar to ductus and it is productive to view these two aspects of the profile in
relation to the other. Penmanship refers to the mode in which a particular script
is executed. So, as Parkes points out, different modes of penmanship are “char-
acterized by different patterns of repetitive strokes”.26 Some key terminology
relating to penmanship includes ‘constant pen angle’, which refers to how the
writer holds the pen, and the ‘module of the handwriting’, which refers to “the
distance between the bottom of the minims in one line and the bottom of those
in the next”.27 The module, combined with the constant pen angle, contributes
to the overall chiaroscuro effect, which is essentially how the writing appears
graphically when committed to the writing surface.

26
M.B. PARKES, Their Hands Before Our Eyes: A Closer Look at Scribes. The Lyell
Lectures Delivered in the University of Oxford 1999 (Aldershot, 2008), p. 108.
27
PARKES, Their Hands Before Our Eyes, p. 60.
238 I.J. MARCUS

The method involved in assessing penmanship is similar to the assessment

of ductus in that it requires the researcher to observe and then describe this
particular aspect of the scribal practice rather than cataloguing it with a list or
questionnaire. Questions that can be asked include: is a constant pen angle
employed? Is there a regular module? Is there is a distinction between thick and
thin strokes? Is the spacing between the words and / or letterforms stable or
unstable? In the group of 17 letters hypothesised to have been copied by Scribe
1, the spacing of the words and letters is unstable. In some letters, such as ID
198, the words are very spaced out, whereas in others such as ID 107 they are
very close together. The amount of space between this scribe’s individual letter
shapes also varies. In letters where the writing is larger and looser, there tends
to be more space between the individual graphs, such as in ID 099, 183, and
110. However, the spacing between the individual letter shapes is never ex-
treme. There is never an excessive amount of space, or so little that the letter
shapes become illegible.
Another aspect that can be considered in relation to penmanship is layout.
Is the letter layout consistent or inconsistent across the group being investi-
gated? For example, how close together are the signature and the subscrip-
tion?28 Considered together, ductus and penmanship can provide useful infor-
mation about whether a set of letters is copied by the same individual writer, or
whether there are some in the set that may not have been copied in the same
hand.

Graphemes

The ‘Graphemes’ section is a vital part of the profile template and comple-
ments the ‘Penmanship’ and ‘Ductus’ sections. Detailed observation of each
scribe’s allographs is a key way of identifying different scribal hands. It can be
particularly useful when a relatively unstable hand is being profiled. For exam-
ple, the ductus and penmanship of Scribe 1’s hand vary considerably. ID 200
and ID 099, two letters that are now, post analysis, thought to be by Scribe 1,
appear to be in the same hand. However, their ductus and penmanship differ.
28
This would also potentially depend on the social distance between the writer and the
addressee; see M. NEVALA, Address in Early English Correspondence: Its Forms and Socio-
Pragmatic Functions (Helsinki, 2004: Mémoires de la Société Néophilologique de Helsinki 64),
p. 233; M. NEVALA, “Accessing politeness axes: Forms of address and terms of reference in early
English correspondence”, Journal of Pragmatics 36 (2004), pp. 2125-2160.
Whose Letters Are They Anyway? 239

ID 200 has much thicker pen strokes compared to ID 099. But by looking at the
allographs contained within ID 200 and ID 099, alongside other features such as
spelling and punctuation, it was possible to identify the hand of Scribe 1. For
example, Scribe 1’s characteristic allographs, such as << b >>, were present in
both letters. The key word in the last sentence is ‘characteristic’. It is the char-
acteristic, or ‘distinctive’ features of a hand that really help to distinguish it. As
Horobin points out, “the identification of scribes in more than one manuscript
depends on a combination of certain distinctive palaeographical and linguistic
factors”.29 In isolation, a particular palaeographical feature is not sufficiently
diagnostic, but a combination of several features “helps to identify the scribe’s
hand”.30 Thus the profile template is aimed at identifying the distinctive fea-
tures of each scribal hand.
The key terminology used in this section includes ‘grapheme’ and ‘allo-
graph’.31 Other terminology used overlaps with the terminology used in the
‘Ductus’ section of the profile, and is useful when describing particular allo-
graphs. For example, ‘ascender’ refers to the vertical parts of particular letter
forms that extend above the x-height, such as b, d, f, h, k, l, and t. The vertical
lines are referred to as ‘stems’.
In order to complete this section of the profile, it is useful for the re-
searcher to use a close reading technique to identify between six to twelve
particularly distinctive or characteristic graphemes (which may have more than
one allograph) in a given scribal hand. This method is necessarily impressionis-
tic, but is productive when attempting to identify individual scribal hands in a
corpus of letters containing several different hands. Once these characteristic
graphemes have been identified, this part of the profile involves compiling a
descriptive list detailing how each grapheme is realised in the hand in question,
for example how many allographs of the grapheme exist. Within each descrip-
tion, it is useful to refer to the image of a sample letter from the set being in-
vestigated that is included at the end of the profile. This provides visual evi-
dence of what is being discussed, and allows the reader to corroborate, or not,
the researcher’s judgement about a given grapheme. The example below is
from Scribe 1’s profile, and describes the letter b:

29
HOROBIN, “The criteria for scribal attribution”, p. 379.
30
HOROBIN, “The criteria for scribal attribution”, p. 376.
31
See the subsection ‘Abbreviations’ above for definitions of these terms.
240 I.J. MARCUS

b There is one allograph of the grapheme b in Scribe 1’s hand. The shape of Scribe
1’s << b >> is very distinctive. It has an exaggerated, curved approach stroke and
flat bottom that is often also concave in appearance. Its lobe is spade-like, and
there is often a gap between the lobe and the stem of the ascender. This gap can
sometimes be quite large, depending on the size of the letter shape. See the initial
<< b >> of <bere> on line 16 of ID 099 below.

Spelling Questionnaire and Summary

The spelling section of the profile is the sole linguistic section of the pro-
file, but it is as important as the palaeographic sections. As McIntosh points
out, “a scribe’s graphological system is part of the system of his written lan-
guage, whether or not it reflects something in his phonological habits as a
speaker of the language”.32 Furthermore, it is methodologically constructive to
view the palaeographic and linguistic choices of a particular scribe in the con-
text of each other. The combination of both aspects of scribal practice can
strengthen the case, built on palaeographical phenomena, that a group of letters
is by the same scribe. Indeed, there are times when palaeographical profiling
alone is not enough. For example, many scribes could write in more than one
kind of script, and could potentially write “in quite dissimilar ways according
to various conventions of formality and elaborateness”.33 If the same scribe
writes different letters in different scripts, a spelling profile can highlight the
linguistic similarities between two texts that are palaeographically dissimilar.
LALME was an initial point of reference when thinking about how to incor-
porate a linguistic element into the profile template. LALME employs a fixed
spelling questionnaire, and although it was designed with a different objective
in mind and had to be adapted to my data, it acted as a suitable point of depar-
ture. A fixed, standardised spelling questionnaire was therefore included in the
profile template. The questionnaire is standardised in that it contains a set of 44
lexical items that are looked for in each of the scribal hands being profiled. The
reason why these specific 44 lexical items were chosen was that they were all
words that appeared relatively frequently in the group of 21 letters thought to
have been copied by Scribe 1, who was in turn hypothesised to be Bess of

32
A. MC INTOSH, “Scribal profiles from Middle English texts”, in: Middle English Dia-
lectology: Essays on Some Principles and Problems, ed. M. LAING (Aberdeen, 1989), pp. 32-45,
at p. 35.
33
MCINTOSH, “Scribal profiles from Middle English texts”, p. 37.
Whose Letters Are They Anyway? 241

Hardwick herself. It was then applied to the other five groups of letters thought
to have been copied in the hands of five of Bess’s scribes, in order to see how
the spelling practices of the individual scribes compared. The words in the
spelling questionnaire used for this particular corpus of letters range from
function words, including modal verb ‘should’, singular third person pronoun
‘her’, a small number of verbs such as ‘suffer’, proper nouns such as ‘Chelsea’
and ‘Shrewsbury’, and nouns such as ‘friendship’, and ‘heart’. Adjectives such
as ‘whole’ and ‘little’ were also included. However, the contents of the ques-
tionnaire can be tailored to reflect the particular corpus that is being used for
the purposes of research. The spelling questionnaire involves listing the lexical
item in its modern, Standard English form, such as ‘whole’ before listing the
orthographic variants of each form. So the example of ‘whole’, from Scribe 1’s
profile, appears like this in the questionnaire:

wolle. 2 examples – line 54 of ID 099, line 17 of ID 120.

43 WHOLE
wholle. 1 example – line 11 of ID 186.

The purpose of the questionnaire is to gauge how consistent and stable the
spelling practice is across a given set of letters thought to be copied in a partic-
ular hand. In accordance with the earlier research into spelling variation within
Queen Elizabeth I’s hand conducted by Vivian Salmon and Mel Evans,34 this
chapter emphasises how useful it is to focus on the consistency (i.e. the extent
to which there is a repeated use of particular spellings) and stability of a partic-
ular individual’s spelling practice.35 A relatively high amount of orthographic
variation is to be expected when using a corpus of texts from the sixteenth
century, due to the lack of a standardised spelling practice in English at this
time. As Donald Scragg points out, a standardised spelling system only started

34
V. SALMON, “Orthography and punctuation”, in: The Cambridge History of the English
Language, ed. R. LASS, 5 vols. (Cambridge, 1999), 3, pp. 13-55; M. EVANS, “A sociolinguistics
of early modern spelling: An account of Queen Elizabeth I’s correspondence”, in: Outposts of
Historical Corpus Linguistics: From the Helsinki Corpus to a Proliferation of Resources. ed. J.
TYRKKÖ, M. KILPIÖ, T. NEVALAINEN, and M. RISSANEN (Helsinki, 2012: Studies in Variation,
Contacts and Change in English 10) <http://www.helsinki.fi/varieng/series/volumes/10/evans/>.
35
For other research into EModE spelling variation within a single hand, see M. SÖNMEZ,
“Perceived and real differences between men’s and women’s spellings of the early to mid-
seventeenth century”, in: The History of English in a Social Context: A contribution to historical
sociolinguistics, ed. D. KASTOVSKY and A. METTINGER (Berlin, 2000: Trends in Linguistics:
Studies and Monographs 129), pp. 405-440.
242 I.J. MARCUS

to emerge at the close of the sixteenth century.36 Furthermore, as Evans notes,

this definition of ‘consistent’ is presently “undefined and impressionistic”, due
to the lack of “macro-level data” to compare with a particular set of letters
thought to be copied in a particular hand, to see if the spelling practice is more
consistent than that of other writers during the period.37 However, it is still
possible to assess the level of stability, particularly if comparing spelling varia-
tion in one set of texts with another set from the same corpus, and the benefit
of the questionnaire approach is that it provides an empirical basis for observa-
tions about the relative consistency of spelling practice, rather than being pure-
ly impressionistic.
In the spelling summary that follows the questionnaire, level of consistency
is assessed and described. For example, in Scribe 1’s hand, 26 out of the 44
lexical items included in the spelling questionnaire, so over 50% of the items,
have more than one variant spelling form across the 21 letters in the dataset.
However, Scribe 1’s spelling practice is relatively stable in the sense that there
are generally only one or two forms of a particular item in this hand, as in the
example of ‘whole’ given above, which appears as ‘wolle’ and ‘wholle’ in the
data set. Only eleven out of the 44 items have more than two variant forms in
this group of letters (although this does not include plural forms). The sum-
mary is also a chance for the researcher to expand on any diagnostic features
of a particular scribe’s spelling practice that may or may not (depending on the
data) be present. For example, there are several diagnostic features of Scribe
1’s spelling practice that can be gleaned from the questionnaire. One is the use
of w for u, in spellings such as <dowter> for item DAUGHTER and <bownd> for
item BOUND. This is an interesting feature of this scribe’s spelling practice,
because it shows that he or she prefers to spell in a more phonetic way. For
example, according to the OED online’s entry for DAUGHTER, <dowter> is the
Northern English form of <doughter>. <Doughter> was a form of ME
<do¥ter> that was still being used in the sixteenth century.
It is also possible to use the summary to highlight any other distinctive
spelling variants that were not included in the fixed questionnaire because they
did not occur as frequently as the other items. In the case of Scribe 1, these
included: <mouste> for item MUST, <oup> for item UP, <onbende> for item
HUSBAND, forms <faythfoul>, <faythfoull>, <faythfall> and <faythfull> for
item FATHFUL, <howse> for item HOUSE, <onse> for item ONCE, <shewder> for

36
D.G. SCRAGG, A History of English Spelling (Manchester, 1974), p.70.
37
EVANS, “A sociolinguistics of early modern spelling”, section 8.
Whose Letters Are They Anyway? 243

item SHOULDER, <vary> for item VERY, <dongorus> for item DANGEROUS,
<ynfettyone> for item INFECTION, and <yeldod> for item YIELDED. Observa-
tions about for example the extent to which words are being spelt in a particu-
larly phonetic way in a given set of letters can then be made.

Profile Conclusion

The purpose of the profile conclusion is to summarise the profile as a

whole in order to establish if a group of letters is written in the same scribal
hand. It allows the researcher to highlight any palaeographical and linguistic
trends, before noting whether or not they are corroborated by any distinctive
features. For example, it is possible to say that Scribe 1 is not systematic in his
or her punctuation practice. Rather, there is variation in both the amount and
type of punctuation employed across this set of letters. The same can be said of
this scribe’s capitalisation practice, although he or she does tend to capitalise
a word if it has one of several particular graphemes (l, c, i, and g) in word-
initial position. The hand that these letters are copied in is consistently large,
distinctive, confident, and legible, though not calligraphic. Most of these letters
have a reasonably regular module, although they were clearly not written using
ruled lines. The spacing of the words and graphemes that make them up is not
consistently stable, but there is never an excessive amount of space between
them, nor is there so little that the writing becomes illegible.
The layout of the letters is consistent; a top and left margin are always left,
and in the majority of the letters the subscription and signature are very close
together, just below the main text. There is a larger number of palaeographical
revisions in this set of 21 letters than in any of the other sets thought to be by
individual scribes. For example, the majority of the letters have crossings out,
and over half have insertions. This helps to distinguish these letters from those
copied by the other major scribes in the Bess corpus. The spelling practice
across these letters is relatively stable, although variant spelling forms are
employed, more so for particular items such as the pronoun <her>.
These palaeographical and linguistic trends are corroborated by several
distinctive features. Outstanding palaeographical features include the use of the
wt abbreviation of ‘with’ in word medial and word final position, and macrons
above various abbreviated words to signify missing u, m, n and e. Several of
Scribe 1’s allographs are also very distinctive. For example, this scribe’s
244 I.J. MARCUS

<<b>> allograph is notably idiosyncratic in comparison with the other scribes’

allographs of the b grapheme. There are also several distinctive spelling fea-
tures, such as w for u, in spellings such as <dowter> for item <daughter> and
double tt for single t, such as in the form <yett>, for item <it>. In relation to
content, there is no particular trend that unites the set of 21 letters, as they
cover a wide range of issues, have a wide range of functions, and are addressed
to a variety of different recipients. However, it is possible to say that Scribe 1
frequently copies letters addressed to a particular group of recipients, namely
Robert Dudley, Earl of Leicester, Shrewsbury, and Sir John Thynne, who are
all addressees of separate letters.
The group of 21 letters that were initially included in Scribe 1’s profile was
reduced to a smaller group of 17 letters because four letters from the original
group, namely ID 177, 100, 116 and 233 did not contain the characteristic
palaeographic and spelling features found in the other 17.38
Once the palaeographical and linguistic trends and distinctive features have
been noted, and particular letters excluded, it is possible for the researcher to
view the remaining set of letters collectively, and to potentially identify a scri-
bal hand. For example, the fact that the 17 remaining letters in Scribe 1’s pro-
file exhibit strikingly similar palaeographical and linguistic features that are
consistently used across this group of letters provides convincing evidence that
(a minimum of) each letter’s body is written in Scribe 1’s italic hand.

Image of Sample Letter

The image of a sample letter is an important part of the profile, because it

provides the reader with visual evidence of much of what is being discussed in
the scribal profile. To exemplify, Fig. 2 (see the end of this chapter) is an im-
age of letter ID 099: Folger, Cavendish-Talbot MSS, X.d.428 (82), which is one
of the letters in the group of 21 included in Scribe 1’s profile. Line numbers are
added to the image to facilitate its discussion in the text.

38
It should be noted that the spelling features were not used for corpus selection as the
corpus was already selected prior to the design of the fixed spelling questionnaire. The features
were chosen because they appeared relatively frequently in the eventual data set (please see
section ‘Stage 3’ above).
Whose Letters Are They Anyway? 245

Discussion and Conclusions

As can be seen from the description of the scribal profiling methodology

above, the main function of the profile is to test a working hypothesis that a
group of letters is copied by the same individual scribe. Its aim therefore is to
identify a scribal hand, or a scribe’s palaeographic and linguistic fingerprint.
Viewed as a whole process, this scribal identification method has various po-
tential applications. It is a qualitative process, and can inform and enable quali-
tative studies of the language of particular individual, or comparative sociolin-
guistic / socio-pragmatic studies focusing on the different linguistic repertoires
of writers across a particular corpus of manuscript texts. It can also potentially
allow researchers to isolate holograph writing from scribal writing in medieval
and early modern letters, although this is only really possible when a holograph
signature is present.
However, in addition to qualitative linguistic studies, the methodology
could potentially be adapted and applied to larger correspondence corpora, and
by extension be used for both qualitative and quantitative studies that use
larger bodies of linguistic data. As Tyrkkö shows in this volume, it is possible
to develop computational methods for quantifying amounts and distribution of
printed content on a page.39 A potential direction for future research in relation
to the scribal identification method presented in this chapter could be to de-
velop it along similar, computational lines. Could the digitised graphemic iden-
tification software that currently exists be adapted so that it can be applied to
manuscript letter data? There have been some recent studies investigating
similar ideas. For example, Aitor Arronte Álvarez, Tal Hassneret al., and Lior
Wolf et al. have carried out research into computational analysis of digitised
historical handwritten documents in relation to palaeography specifically.40
Tara Gilliam et al. have investigated the potential for automated scribal identi-

39
J. TYRKKÖ, “Quantifying contrasts: A method of computational analysis of text and
paratext on early medical title pages”, in this volume.
40
A.A. ÁLVAREZ, “Enriching digitized medieval manuscripts: Linking image, text and
lexical knowledge”, Proceedings of the 9th SIGHUM Workshop on Language Technology for
Cultural Heritage, Social Sciences, and Humanities (LATECH) (2015), pp 73-77
<http://aclweb.org/anthology/sighum.html>; T. HASSNER, M. REHBEIN, P.A. STOKES and L.
WOLF, “Computation and palaeography: Potentials and limits”, Dagstuhl Manifestos 2 (2012),
pp. 14-35; L. WOLF, L. POTIKHA, N. DERSHOWITZ, R. SHWEKA, and Y. CHOUEKA, “Computerized
paleography: Tools for historical manuscripts”, in: 18th IEEE International Conference on Image
Processing (2011), pp. 3545-3548 <http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=
6116481>.
246 I.J. MARCUS

fication, and Thomas Pilz et al. have looked into the possibility of developing
a generic spelling detection tool for Indo-European languages.41 Applying
graphemic identification software to manuscript letter data would of course
require the programme to identify visual as opposed to, or as well as, lexical
features, and it would still require a certain amount of preliminary qualitative
work, for example to establish which allographs the computer programme
needs to recognise. It would also require digitised copies of the manuscript
texts so that there are digital images of the texts available for the software
programme to read. However, it is potentially possible, and would potentially
enable researchers to take the holograph / scribal variable into account when
creating corpora of manuscript letter texts in the future.
To conclude, there is an increasing amount of evidence to suggest that
there are often clear linguistic differences between the holograph and scribal
writing contained within early modern letters, i.e. letters written between 1500
and 1700. This chapter argues that historical linguists and other scholars work-
ing with early modern textual material need to take these differences into ac-
count. Although it acknowledges that this can be challenging, it presents a
scribal identification methodology that has been specifically designed to be
applied to a corpus of sixteenth and early seventeenth century manuscript let-
ters from Elizabeth Talbot, commonly known as Bess of Hardwick. The chap-
ter focuses on the central part of the methodology, namely the completion of a
scribal profile that combines palaeographic analysis with a standardised spell-
ing questionnaire, and provides detailed descriptions of the individual sections
of the profile. It is suggested that once a researcher has completed this initial,
scribal identification process, he or she is in a better position to observe and
analyse any linguistic variation that may be present according to whether the
writing is holograph or scribal.
Furthermore, it is suggested that this methodology could potentially be
adapted to be applied to larger bodies of data. Specifically, that it might be
possible to design software that identifies and then counts particular spellings
and potentially also allographs present in digital images of individual manu-
script letters, and potentially other kinds of text. This would make the initial

41
T. GILLIAM, R.C. WILSON, and J.A. CLARK, “Scribe identification in medieval English
manuscripts”, in: Pattern Recognition (ICPR), 2010 20th International Conference (2010), pp.
1880-1883 <http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5597227>; T. PILZ, A.
ERNST-GERLACH, S. KEMPKEN, P. RAYSON and D. ARCHER, “The identification of spelling vari-
ants in English and German historical texts: Manual or automatic?”, Literary and Linguistic
Computing 23 (2008), pp. 65-72.
Whose Letters Are They Anyway? 247

scribal identification process faster and more efficient, and would enable re-
searchers working with large corpora to acknowledge the collaborative produc-
tion of early modern manuscript texts. It is hoped that this chapter may form
part of a chain leading to future work and methodologies which accommodate
large bodies of data and investigate a variety of linguistic features and aspects
of language, whilst simultaneously maintaining analytical integrity by taking
the nature of the data sources being used into account.

On the next pages:

Fig. 2a-c Letter ID 099: Folger, Cavendish-Talbot MSS, X.d.428 (82). By Permis-
sion of Folger Shakespeare Library, link to Folger Digital Image
Collection: <http://luna.folger.edu/>.
248 I.J. MARCUS

10

15

20
Whose Letters Are They Anyway? 249

25

30

35

40

45
250 I.J. MARCUS

50

55

60

65
 Bibliography

Primary Sources

Manuscripts

Cambridge, Magdalene College, Pepys 2344

Cambridge, Trinity College, R.3.25
Cambridge, University Library, Additional 3039
Cambridge, University Library, Additional 8335
Cambridge, University Library, Dd.1.1
Cambridge, University Library, Ff.6.33
Cambridge, University Library, Gg.5.31
Dublin, Trinity College, 244
Edinburgh, National Library of Scotland, 185
Edinburgh, National Library of Scotland, 3147
Edinburgh, National Library of Scotland, Acc. 9769 84/1/1 1/2 (‘Crawford MS I’)
Edinburgh, National Library of Scotland, Acc. 9769 84/1/1 2/2 (‘Crawford MS II’)
Edinburgh, National Library of Scotland, Adv. 35.4.11
Edinburgh, National Library of Scotland, Wodrow Folio XLVIII (‘Wodrow Folio’)
Edinburgh, University of Edinburgh Library, La.III.198
Edinburgh, University of Edinburgh Library, La.III.218
Edinburgh, University of Edinburgh Library, La.III.583
Exeter, Cathedral, Dean and Chapter Library 3501 (‘The Exeter Book’)
Glasgow, University Library, Hunterian T.6.18
Hatfield, Hatfield House Archives, Cecil Papers (CP) 317/1
London, British Library, Additional 22121
London, British Library, Additional 22283 (‘Simeon’)
London, British Library, Additional 38010
London, British Library, Cotton Julius D IX
London, British Library, Cotton Tiberius E VII
252 Bibliography

London, British Library, Cotton Vitellius E XVIII

London, British Library, Egerton 1993
London, British Library, Harley 2277
London, British Library, Harley 2391
London, British Library, Harley 4196
London, British Library, Stowe 949
London, Lambeth Palace Library, 260
London, The National Archives, State Papers 94/7
London, The National Archives, State Papers 94/9
London, The National Archives, State Papers 94/11
London, The National Archives, State Papers 94/12
London, The National Archives, State Papers 94/13
London, The National Archives, State Papers 94/15
Oxford, Bodleian Library, Ashmole 42
Oxford, Bodleian Library, Ashmole 43
Oxford, Bodleian Library, Bodley 779
Oxford, Bodleian Library, Eng. Poet. a.1 (‘Vernon’)
Oxford, Bodleian Library, Eng. Poet. c.3
Oxford, Bodleian Library, Laud Misc. 38
Oxford, Bodleian Library, Laud Misc. 463
Oxford, Bodleian Library, Laud Misc. 517
Oxford, Bodleian Library, Selden Supra 52
Oxford, Trinity College, 57
San Marino, CA, Huntington Library, HM 129
San Marino, CA, Huntington Library, HM 149
Washington, DC, Folger Shakespeare Library, Cavendish-Talbot MSS, X.d.428 (82)

Printed Primary Sources (from the Hand Press Period)

ANON., Aristoteles Master-piece, or the Secrets of Generation displayed in all the parts
thereof (London, 1684). WING A3689.
ANON., The Dictes or Sayengis of the Philosophhres ([Westminster, 1477]). STC 6826.
ANON., [Ortus vocabulorum] (Londini, 1509). STC 13830.7.
ANON., The plagues approved physitian (London, 1665). WING P2337.
ANON., The prophecies, and predictions, for London’s deliverance ([London], 1665).
WING P3674.
ANON., The shutting up infected houses as it is practised in England soberly debated.
([London], 1665). WING S3717.
ANON., A treatise of the plague (London, 1630). STC 19192.
Bibliography 253

BALMFORD, J., A short dialogve concerning the plagves infection (London, 1603). STC
1338.
BARET, J., An aluearie or triple dictionarie in Englishe, Latin, and French: very profit-
able for all such as be desirous of any of those three languages. Also by the two
tables in the ende of this booke, they may contrariwise, finde the most necessary
Latin or French wordes, placed after the order of an alphabet, whatsoeuer are to
be founde in any other dictionarie: and so to turne them backwardes againe into
Englishe when they reade any Latin or French aucthors, & doubt of any harde
worde therein (London, 1574). STC 1410.
BARKER, R., Sir., Consilium anti-pestilentiale, or, Seasonable advice concerning sure,
safe, specifick, and experimented medicines both for the preservation from, and
cure of, this present plague (London, 1665). WING B778.
BILLINGSLEY, M., A Newe booke of copies containing divers sortes of sundry hands, as
the English and French secretarie, and bastard secretarie, Italian, Roman, chan-
cery, and court hands (London, 1620). STC 33638.
BORASTON, W., A necessarie and briefe treatise of the contagious disease of the pesti-
lence (London, 1630). STC 3372.
BRADWELL, S., A vvatch-man for the pest (London, 1625). STC 3537.
BRADWELL, S., Physick for the sicknesse, commonly called the plague (London, 1636).
STC 3536.
BRIGHT, T., A treatise of melancholy (London, 1586). STC 3748.
BROWNE, D., The New Invention, intitvled, calligraphia: or, the arte of faire writing (St
Andrews, 1622). STC 3905.
CALEPINUS, A., Ambrosii Calepini Bergomatis Dictionarium (Reggio Emilia, 1502).
CAWDREY, R., A table alphabeticall, conteyning and teaching the true vvriting, and
vnderstanding of hard vsuall English wordes, borrowed from the Hebrew, Greeke,
Latine, or French. &c. (London, 1604). STC 4884.
CAWDREY, R., A table alphabeticall contayning and teaching the true writing and
vnderstanding of hard vsuall English words, borrowed from the Hebrevv, Greeke,
Latine, or French, &c. With the interpretation thereof by plaine English words,
gathered for the benefit and help of all vnskilfull persons (London, 1613). STC
4885.
CHAUCER, G., The Canterbury Tales ([Westminster, 1477]). STC 5080.
COCKER, E., Cocker’s English dictionary: interpreting the most refined and difficult
words ... By Edward Cocker ... Perused and published from the authors correct
copy, by John Hawkins (London, 1704).
COLES, E., The compleat English schoolmaster or, the most natural and easie method
of spelling English (London, 1674). WING C5067A.
COOPER, T., Thesaurus linguae Romanæ & Britannicæ (Londini, 1565). STC 5686.
COOTE, E., The English school-maister: teaching all his scholers, the order of distinct
reading, and true writing our English tongue (London, 1596). STC 5711.
254 Bibliography

DAVIES, J., The writing schoolemaster, or, The anatomie of faire writing wherein is
exactlie expressed each seuerall character, together with other rules & documents,
coincident to the art of faire and speedy writing, 6th enlarged edn. (London, 1631).
STC 6344.5.
DE BERLEMONT, N., Colloquia et dictionariolum octo linguarum, Latinae, Gallicae,
Belgicae, Teutonicae, Hispanicae, Italicae, Anglicae, et Portvgallicae (Delft,
1613). STC 1431.22.
DE BÈZE, T., A learned treatise of the plague (London, 1665). WING B2196.
DE CESSOLIS, J., The Game and Playe of Chesse ([Bruges, 1473-1474?]). STC 4920.
DE MONTENAY, G., A booke of armes, or remembrance wherein ar one hundered godly
emblemata, in péeces of brasse very fine graven, and adorned pleasant to bé séen;
first by the noble, and industrious minde Georgetta de Montenay, invented and
only in the Frenchtongve [sic] elabourated; bot [sic] now, in severall langvages,
as; Latin, Spanish, Italian, Highdutch, English, and Lovedutch, meetre or verse
wys, of the same manner declared, and augmented (Frankfurt, 1619). STC 18046.
DE VIGO, J., The Most Excellent Workes of Chirurgerye (London, 1543). STC 24720.
DURANT, J., Art and Nature Joyn Hand in Hand, or the Poor Mans Daily Companion
(London, 1697). WING D2681A.
DYCHE, T., A new general English dictionary; peculiarly calculated for the use and
improvement of such as are unacquainted with the learned languages. … now
finish’d by William Pardon, Gent. The second edition, with additions (London,
1737).
DYCHE, T., and W. PARDON, A new general English dictionary (London, 1781).
E.B., A nevv booke, containing all sorts of hands vsually written at this day in
Christendome as the English and French Secretary, the Roman, Italian, French,
Spanish, high and low Dutch, court and chancerie hands: with examples of each
of them in their proper tongue and letter (London, 1611). STC 3361.7.
ELYOT, T., The dictionary of syr Thomas Eliot knyght (Londini, 1538). STC 7659.
ELYOT, T., Bibliotheca Eliotæ = Eliotis librarie. This dictionarie novv nevvly imprinted
… is augmented and inriched with aboue. XXXIIJ. thousande wordes and phrases
(Londini, 1548). STC 7661.
ESTIENNE, R., Dictionarium, seu Latinæ linguæ thesaurus (Parisiis, 1531).
GADBURY, J. London’s deliverance predicted in a short discourse shewing the causes
of plagues in general (London, 1665). WING G86.
GERARDE, J., The Herball, or Generall historie of plantes (London, 1633). STC 11751.
HARVEY, G., Morbus Anglicus (London, 1666). WING H1070.
HENISCH, G., Teutsche Sprach und Weiszheit. Thesaurus linguae et sapientiae germa-
nicae (Augusta Vindelicorum (Augsburg, 1616).
HER(R)ING, F., Certaine rules, directions, or aduertisments for this time of pestilentiall
contagion (London, 1603). STC 13239.5.
Bibliography 255

HER(R)ING, F., A modest defence of the caueat giuen to the wearers of impoisoned
amulets, as preseruatiues from the plague wherein that point it somewhat more
largely reasoned and debated with an ancient physician, who hath mainteined
them by publicke writing: as likewise that vnlearned and dangerous opinion, that
the plague is not infectious, lately broched in London, is briefly glansed at, and
refuted by way of preface, by Fr. Hering D. in Physicke (London, 1604). STC
13248.
HOBBES, S., A nevv treatise of the pestilence, containing the causes, signes, preserua-
tiues and cure thereof (London, 1603). STC 12577.
HODY, H., A Letter from Mr. Humphry Hody, to a Friend, Concerning a Collection of
Canons said to be Deceitfully omitted in his Edition of the Oxford Treatise against
Schism (Oxford, 1692). WING H2342.
HOWLET, R., Abcedarium anglico latinum, pro tyrunculis Richardo Huloeto exscriptore
(Londini, 1552). STC 13940.
I.W., A Briefe treatise of the plague (London, 1603). STC 24905.7.
(JAMES I)., Orders, thought meete by his Maiestie, and his Priuie Counsell, to be exe-
cuted throughout the counties of this realme, in such townes, villages, and other
places, as are, or may be hereafter infected with the plague, for the stay of further
increase of the same (London, 1603). STC 9209.
JAMES, R., A medicinal dictionary; including physic, surgery, anatomy, chymistry, and
botany, in all their branches relative to medicine, 2 vols. (London, 1743).
J.K., A new English dictionary: or, compleat collection of the most proper and signifi-
cant words, commonly used in the language; with a short and clear exposition of
difficult words and terms of art (London, 1702).
JOHNSON, S., A dictionary of the English language; in which the words are deduced
from their originals and illustrated in their different significations by examples
from the best writers, 2 vols. (London, 1755).
KEMP, W., A brief treatise of the nature, causes, signes, preservation from, and cure of
the pestilence (London, 1665). WING K260.
[KEPHALE, R.], Medela pestilentiae wherein is contained several theological queries
concerning the plague (London, 1665). WING K330.
KERSEY, J., Dictionarium Anglo-Britannicum: or, a general English dictionary, com-
prehending a brief, but emphatical and clear explication of all sorts of difficult
words (London, 1721).
KILVERT, R., A reply to a most untrue relation made and set forth in print by certaine
vintners in excuse of their wine proiect (London, 1641). WING K478.
LEFÈVRE, R., [here begynneth the volume intituled and named] the recuyell of the his-
toryes of Troye, trans. W. CAXTON ([Bruges, 1473-1474]). STC 15375.
LINDSAY, R., The History of Scotland, ed. R. URIE (Glasgow, 1749).
LINDSAY, R., The History of Scotland, ed. T. CADELL (London, 1778).
LINDSAY, R., The History of Scotland, ed. J.G. DALYELL (Edinburgh, 1814).
256 Bibliography

LODGE, T., A treatise of the plague (London, 1603). STC 16676.

MARTIN, B., Lingua Britannica reformata: or, a New English Dictionary (London,
1749).
M.R., The meanes of preventing, and preserving from, and curing of the most conta-
gious disease, called the plague (London, 1665). WING R45.
MOULTON, T., This is the Myrrour or Glasse of helth necessary and nedefull for euery
person to loke in, that wil kepe their bodye from the syckenesse of the Pestilence,
and it sheweth how the planettes do raygne in euery houre of the daye and nyghte,
with the natures and exposicio~s of the XII sygnes, deuyded by the .XII. Monethes
of the yeare, and shewed the remedies for many dyuers infirmities and dyseases
that hurteth the bodye of Manne, 5th edn. (London, 1540?). STC 18225.2.
MOXON, J., Regulae Trium Ordinum literarum typographicarum: or the Rules of the
three orders of Print Letters: viz. The Roman, Italick, English, Capitals and Small
(London, 1676). WING M3019.
MOXON, J., Mechanick exercises, or, the doctrine of handy-works ..., 2 vols. (London,
1677-1680). WING M3014.
MULCASTER, R., The first part of the elementarie vvhich entreateth chefelie of the right
writing of our English tung (London, 1582). STC 18250.
PRYNNE, W., A legall vindication of the liberties of England, against illegall taxes and
pretended Acts of Parliament lately enforced on the people (London, 1649). WING
P3996A.
ROYAL COLLEGE OF PHYSICIANS, The Kings medicines for the plague (London, 1630).
STC 16770.
SCAPULA, J., Lexicon Graecolatinum novvm in qvo ex primitivorvm & simplicivm fonti-
bus derivata atque composit (Basileæ, 1580).
SHERWOOD, T., The charitable pestmaster, or, the cure of the plague (London, 1641).
WING S3416. Thomason E.176[10].
SOMNER, W., Dictionarium Saxonico-Latino-Anglicum voces, phrasesque præcipuas
Anglo-Saxonicas, e libris, sive manuscriptis, sive typis excusis, aliisque monumen-
tis tum publicis tum privatis, magna diligentia collectas; cum Latina et Anglica vo-
cum interpretatione complectens (Oxonii, 1659). WING S4663.
SPELMAN, H., Sir, Glossarium archaiologicum Part 1 (Londini, 1626). STC 23065.5.
SPELMAN, H., Sir, Glossarium archaiologicum: continens latino-barbara, peregrina,
obsoleta, & novatæ significationis vocabula (Londini, 1664). WING S4925.
T.D., Food and physick for every householder & his family during the time of the
plague (London, 1665). WING D88.
THAYRE, T., A treatise of the pestilence (London, 1603). STC 23929.
THOMSON, G., Loimologia. A consolatory advice, and some brief observations con-
cerning the present pest (London, 1665) WING T1026.
THOMSON, G., Ëoimotomia: or, the pest anatomized (London, 1666).WING T1027.
Bibliography 257

WHARTON, T., Directions for the prevention and cure of the plague (London, 1665).
WING W1577.
WITHALS, J., A shorte dictionarie for yonge begynners (London, 1553). STC 25874.
W.J., A collection of seven and fifty approved receipts good against the plague (Lon-
don, 1665). WING J47.
WOODALL, J., The cure of the plague by an antidote called aurum vitae (London,
1640). STC 25961.
(W.W.), The Christians refuge: or heavenly antidotes against the plague (1665). Not
in WING.

Editions and Corpora

Aristotle, Art of Rhetoric, trans. J.H. FREESE (Cambridge, MA, 1926: Loeb Classical
Library 193).
Bess of Hardwick’s Letters: The Complete Correspondence, c.1550-1608, ed. A. WIG-
GINS, A. BRYSON, D. STARZA SMITH, A. TIMMERMANN, G. WILLIAMS, and K.
ROGERS, University of Glasgow, web development by K. ROGERS, University of
Sheffield Humanities Research Institute (April 2013) <http://www.bessofhardwick.
org>.
The Cecil Papers <cecilpapers.chadwyck.co.uk>.
The Dream of the Rood, ed. M. SWANTON, 2nd edn. (Exeter, 1987).
Early Modern English Medical Texts (EMEMT), ed. I. TAAVITSAINEN, P. PAHTA, M.
MÄKINEN, T. HILTUNEN, V. MARTTILA, M. RATIA, C. SUHR, and J. TYRKKÖ. Cor-
pus, CD-ROM (Amsterdam and Philadelphia, PA, 2010).
ECCO: Eighteenth-Century Collections Online <http://www.gale.com/primary-sources/
eighteenth-century-collections-online/>.
EEBO: Early English Books Online. < http://eebo.chadwyck.com/home>.
The Exeter Book, ed. G.P. KRAPP and E. VAN KIRK DOBBIE (New York, 1936: Anglo-
Saxon Poetic Records 3).
The Lampeter Corpus of Early Modern English Tracts, comp. J. SCHMIED, C. CLARIDGE
and R. SIEMUND (1999). Available from the Oxford Text Archive <ota.ox.ac.uk>.
Late Modern English Medical Texts (LMEMT), ed. I. TAAVITSAINEN, P. PAHTA, T. HIL-
TUNEN, M. MÄKINEN, V. MARTTILA, M. RATIA, C. SUHR, and J. TYRKKÖ. Corpus
(forthcoming).
The Lisle Letters, ed. M. ST. CLARE-BYRNE, 6 vols. (Chicago, IL, 1981).
The Northern Homily Cycle: The Expanded Version in MSS Harley 4196 and Cotton
Tiberius E VII, ed. S. NEVANLINNA, 3 vols. (Helsinki, 1972-1984: Mémoires de la
Société néophilologique 38, 41, 43).
The Old English Riddles of the Exeter Book, ed. C. WILLIAMSON (Chapel Hill, 1977).
The Vercelli Book, ed. G.P. KRAPP (New York, 1932: Anglo-Saxon Poetic Records 2).
258 Bibliography

Literature

AKINNASO, F.N., “The consequences of literacy in pragmatic and theoretical perspec-

tives”, Anthropology and Education Quarterly 12 (1981), pp. 163-200.
ÁLVAREZ, A.A., “Enriching digitized medieval manuscripts: Linking image, text and
lexical knowledge”, LaTeCH 2015 (2015), pp. 73-77.
ANGERMEYER, P.S., “Bilingualism meets digraphia: Script alternation and hybridity in
Russian-American writing and beyond”, in: Language Mixing and Code-Switching
in Writing: Approaches to Mixed-Language Written Discourse, ed. M. SEBBA, S.
MAHOOTIAN, and C. JONSSON (New York and London, 2012), pp. 255-272.
ANTTILA, R., Historical and Comparative Linguistics (Amsterdam and Philadelphia,
PA, 1989).
ARMSTRONG, E., Robert Estienne, Royal Printer: An Historical Study of the Elder Ste-
phanus (Cambridge, 2011).
ARMSTRONG, G., “Coding continental: Information design in sixteenth-century English
vernacular manuals and translations”, Renaissance Studies 29 (2015), pp. 78-102.
ATKIN, T., and A.S.G. EDWARDS, “Printers, publishers and promoters to 1558”, in: A
Companion to the Early Printed Book in Britain, 1476-1558, ed. V. GILLESPIE and
S. POWELL (Cambridge, 2014), pp. 27-44.
BAHR, A.W.,“Reading codicological form in John Gower’s Trentham manuscript”,
Studies in the Age of Chaucer 33 (2011), pp. 219-262.
BAMMESBERGER, A., “Three Old English runic inscriptions”, in: Old English Runes and
Their Continental Background, ed. A. BAMMESBERGER (Heidelberg, 1991), pp.
125-136.
BAMMESBERGER, A. “The Brandon Antler runic inscription”, Neophilologus 86 (2002),
pp. 129-131.
BARNES, M.P., Runes: A Handbook (Woodbridge, 2012).
BARKER, N., “Caxton’s typography”, Journal of the Printing Historical Society 11
(1976-1977), pp. 114-143.
BARTHES, R., “Rhetoric of the image”, in: ID., Image–Music–Text, selected and trans.
S. HEATH (New York, 1977), pp. 32-51.
BATEMAN, J.A., Multimodality and Genre: A Foundation for the Systematic Analysis
of Multimodal Documents (London, 2008).
BATEMAN, J.A., Text and Image: A Critical Introduction to the Visual / Verbal Divide
(Abingdon, 2014).
BEAL, P., A Dictionary of English Manuscript Terminology (Oxford, 2008).
BECKER, D.P., The Practice of Letters: The Hofer Collection of Writing Manuals 1514-
1800 (Cambridge, MA, 1997).
BERGS, A., Social Networks and Historical Sociolinguistics: Studies in Morphosyn-
tactic Variation in the Paston Letters (1421-1503) (Berlin, 2005).
Bibliography 259

Bi-Graphic Differences: Languages in Con(tact)(flict), ed. R. HODGSON and R. SARKO-

NAK (= Visible Language 21 (1987)).
Bilingualism in Ancient Society: Language Contact and the Written Text, ed. J.N. AD-
AMS, M. JANSE, and S. SWAIN (Oxford, 2002), Oxford Scholarship Online <www.
oxfordscholarship.com>.
BITTERLI, D., Say What I Am Called: The Old English Riddles of the Exeter Book and
the Anglo-Saxon Riddle Tradition (Toronto, 2009).
BLACKBURN, P.L., The Code Model of Communication: A Powerful Metaphor in Lin-
guistic Metatheory (Dallas, 2007).
BLADES, W., The Life and Typography of William Caxton, England’s First Printer, 2
vols. (Cambridge, 1861-1863, repr. 2014).
BLAKE, N.F., “Manuscript to print”, in: Book Production and Publishing in Britain
1375-1475, ed. J. GRIFFITHS and D. PEARSALL (Cambridge, 1989), pp. 403-432.
BLAND, M., “The appearance of the text in early modern England”, Text 11 (1998), pp.
91-154.
BLAND, M., A Guide to Early Printed Books and Manuscripts (Hong Kong, 2010).
BRAGG, L., “Runes and readers: In and around ‘The Husband’s Message’”, Studia
Neophilologica 71 (1999), pp. 34-50.
BREDEHOFT, T.A., “Multiliteralism in Anglo-Saxon verse inscriptions”, in: Conceptual-
izing Multilingualism in England, c. 800-c. 1250, ed. E.M. TYLER (Turnhout,
2011: Studies in the Early Middle Ages 27), pp. 15-32.
BROWN, M.P., A Guide to Western Historical Scripts from Antiquity to 1600 (London,
1993).
BROWN, M.P., Understanding Illuminated Manuscripts: A Guide to Technical Terms
(Malibu, CA, and London, 1995).
BUCHSTALLER, I., Quotatives: New Trends and Sociolinguistic Implications (Chich-
ester, 2014).
BÜHLER, C.F., Standards of Bibliographical Description (Philadelphia, PA, 1949).
BURNARD, L., “On the hermeneutic implications of text encoding”, in: New Media and
the Humanities: Research and Applications, ed. D. FIORMONTE and J. USHER (Ox-
ford, 2001).
BURNARD, L., What is the Text Encoding Initiative? How to Add Intelligent Markup to
Digital Resources, new ed. (online) (Marseille, 2014) <http://books.openedition.
org/oep/426>.
CAMPBELL, A., Old English Grammar (Oxford, 1959).
CARLSON, D.R., “Manuscripts after printing: Affinity, dissent and display in the texts of
Wyatt’s Psalms”, in: Prestige, Authority, and Power, ed. F. RIDDY (Woodbridge,
2000), pp. 171-188.
CARLSON, D.R., “A theory of the early English printing firm: Jobbing, book publishing,
and the problem of productive capacity in Caxton’s work”, in: Caxton’s Trace:
260 Bibliography

Studies in the History of English Printing, ed. W. KUSKIN (Notre Dame, IN, 2006),
pp. 35-68.
CARROLL, R., “Middle English recipes”, in: Medical and Scientific Writing in Late
Medieval English, ed. I. TAAVITSAINEN and P. PAHTA (Cambridge, 2004), pp. 174-
191.
CARROLL, R., M. PEIKOLA, H. SALMI, M.-L. VARILA, J. SKAFFARI, and R. HILTUNEN,
“Pragmatics on the page: Visual text in late medieval English books”, European
Journal of English Studies 17 (2013), pp. 54-71.
CARRUTHERS, M.J., The Book of Memory: A Study of Memory in Medieval Culture
(Cambridge, 1992).
CARRUTHERS, M., and J.M. ZIOLKOWSKI, “General introduction”, in: The Medieval
Craft of Memory: An Anthology of Texts and Pictures, ed. M. CARRUTHERS and
J.M. ZIOLKOWSKI (Philadelphia, PA, 2004), pp. 1-31.
CARTER, H., and H.D.L. VERVLIET, Civilité Types (Oxford, 1966).
CHARTIER, R., The Cultural Uses of Print in Early Modern France, trans. L.G. COCH-
RANE (Princeton, NJ, 1987).
CHARTIER, R., “Texts, printing, readings”, in: The New Cultural History, ed. L. HUNT
(Berkeley, CA, 1989), pp. 154-175.
The Chicago Manual of Style, 14th edn. (Chicago and London, 1993).
CLANCHY, M.T., From Memory to Written Record: England 1066-1307 (London,
1979).
CLARIDGE, C., “Life is ruled and governed by opinion”: The Lampeter Corpus of Early
Modern English Tracts Manual of Information (1999; 2003) <www.tu-chemnitz.
de/phil/english/sections/linguist/real/independent/lampeter/manual/pages/content
s.html>.
CLARIDGE, C., “Pamphlets and early newspapers: Political interaction in news report-
ing”, in: English media texts past and present: Language and textual structure, ed.
F. UNGERER (Amsterdam and Philadelphia, PA, 2000: Pragmatics and Beyond:
New Series 80), pp. 25-43.
CLARIDGE, C., “From page to screen: The relevance of encoded visual features in the
Lampeter Corpus”, in: Principles and Practices for the Digital Editing and Anno-
tation of Diachronic Data, ed. A. MEURMAN-SOLIN and J. TYRKKÖ (Helsinki,
2013: Studies in Variation, Contacts and Change in English 12) <http://www.
helsinki.fi/varieng/series/volumes/14/claridge/>.
Code-Switching in Early English, ed. H. SCHENDL and L. WRIGHT (Berlin, 2011).
COLEMAN, J., Public Reading and the Reading Public in Late Medieval England and
France (Cambridge, 1996).
Communicating Early English Manuscripts, ed. A.H. JUCKER and P. PAHTA (Cam-
bridge, 2011).
CONSIDINE, J., Dictionaries in Early Modern Europe: Lexicography and the Making of
Heritage (Cambridge, 2008).
Bibliography 261

CONSIDINE, J., Academy Dictionaries 1600-1800 (Cambridge, 2014).

CRYSTAL, D., Making a Point: The Pernickety Story of English Punctuation (London,
2015).
CRYSTAL, D., Dictionary of Linguistics and Phonetics, 6th edn. (Oxford, 2009).
DAHOOD, R., “Abbreviations, otiose strokes and editorial practice: The case of South-
well Minster MS 7”, in: New Perspectives on Middle English Texts: A Festschrift
for R.A. Waldron, ed. S. POWELL and J.J. SMITH (Cambridge, 2000), pp. 141-149.
DANE, J.A., The Myth of Print Culture: Essays on Evidence, Textuality and Biblio-
graphical Method (Toronto, 2003).
DANE, J.A., Out of Sorts: On Typography and Print Culture (Philadelphia, PA, 2011).
DANYLENKO, A., “The Uniate ‘Leksikon’ of 1722: A case of regional relapse from
script- to language-switching”, Slavia Orientalis 56 (2007), pp. 543-557.
DARNTON, R., “Towards a history of reading”, The Wilson Quarterly 13.4 (1989), pp.
86-102.
DAWSON, G.E., and L. KENNEDY-SKIPTON, Elizabethan Handwriting 1500-1650: A
Guide to the Reading of Documents and Manuscripts, 2nd edn. (Chichester, 1981).
DAYBELL, J., and P. HINDS, “Introduction: Material matters”, in: Material Readings of
Early Modern Culture: Texts and Social Practices, 1580-1730, ed. J. DAYBELL and
P. HINDS (Basingstoke, 2010), pp. 1-20.
DEN HEIJER, J., and A.B. SCHMIDT, “Scripts beyond borders: Allographic traditions and
their social, cultural and philological aspects. An analytical introduction”, in:
Scripts Beyond Borders: A Survey of Allographic Traditions in the Euro-Mediter-
ranean World, ed. J. DEN HEIJER, T. PATARIDZE, and A.B. SCHMIDT (Louvain-la-
Neuve, 2014: Publications de l’Institut Orientaliste de Louvain 62), pp. 1-64.
DEROLEZ, A., The Palaeography of Gothic Manuscript Books: From the Twelfth to the
Early Sixteenth Century (Cambridge, 2003).
DEROLEZ, R., Runica Manuscripta: The English Tradition (Bruges, 1954).
DEROLEZ, R., “Runic literacy among the Anglo-Saxons”, in: Britain 400-600: Lan-
guage and History, ed. A. BAMMESBERGER and A. WOLLMANN (Heidelberg, 1990),
pp. 397-436.
DE YOUNG, G., “Mathematical diagrams from manuscript to print: Examples from the
Arabic Euclidean transmission”, Synthese 186 (2012), pp. 21-54.
DOBRES, J., B. REIMER, L. PARIKHAL, E. WEAN, and N. CHAHINE, “The incredible
shrinking letter: How font size affects the legibility of text viewed in brief glances”,
paper presented at the 8th International Driving Symposium on Human Factors in
Driver Assessment, Training and Vehicle Design, Salt Lake City, UT, 22-25 June
2015, <agelab.mit.edu>.
DOLLINGER, S., “‘Philological computing’ vs. ‘philological outsourcing’ and the compi-
lation of historical corpora: A Late Modern English test case”, Vienna English
Working Papers (VIEWS) 13 (2004), pp. 3-23 <anglistik.univie.ac.at/views>.
262 Bibliography

DOYLE, A.I., and M.B. PARKES, “A palaeographical introduction”, in: A Variorum

Edition of the Works of Geoffrey Chaucer, 1, ed. P.G. RUGGIERS and D.C. BAKER
(Folkestone, 1979), pp. XIX-XLIV.
DOTY, K., “Telling tales: The role of scribes in constructing the discourse of the Salem
witchcraft trials”, Journal of Historical Pragmatics 8 (2007), pp. 25-41.
DRISCOLL, M.J., “Levels of transcription”, in: Electronic Textual Editing, ed. L.
BURNARD, K. O’KEEFFE, and J. UNSWORTH (Modern Language Association of
America, 2007) <http://www.tei-c.org/About/Archive_new/ETE/Preview/driscoll.
xml>.
Early Modern English Medical Texts: Corpus Description and Studies, ed. I. TAAVI-
TSAINEN and P. PAHTA (Amsterdam and Philadelphia, PA, 2010).
EISENSTEIN, E.L., The Printing Press as an Agent of Change: Communications and
Cultural Transformations in Early Modern Europe, 2 vols. (Cambridge, 1979).
EISENSTEIN, E.L., The Printing Revolution in Early Modern Europe, 2nd edn. (Cam-
bridge, 2005).
EVANS, M., “A sociolinguistics of early modern spelling: An account of Queen Eliza-
beth I’s correspondence”, in: Outposts of Historical Corpus Linguistics: From the
Helsinki Corpus to a Proliferation of Resources. ed. J. TYRKKÖ, M. KILPIÖ, T. NE-
VALAINEN, and M. RISSANEN (Helsinki, 2012: Studies in Variation, Contacts and
Change in English 10) <http://www.helsinki.fi/varieng/series/volumes/10/evans/>.
FARMER, A.B., and Z. LESSER, “What is print popularity? A map of the Elizabethan
book trade”, in: The Elizabethan Top Ten: Defining Print Popularity in Early
Modern England, ed. A. KESSON and E. SMITH (London and New York, 2013), pp.
19-54.
FELL, C.E., “Anglo-Saxon England: A three-script community?”, in: Proceedings of the
Third International Symposium on Runes and Runic Inscriptions: Grindaheim,
Norway, 8-12 August 1990, ed. J.E. KNIRK (Uppsala, 1994), pp.119-137.
FETZER, A., “Challenges in contrast: A function-to-form approach”, in: Contrastive
Pragmatics, ed. K. AIJMER (Amsterdam and Philadelphia, PA, 2011).
FISCHER, S.R., A History of Reading (London, 2003).
FOX, A., Oral and Literate Culture in England 1500-1700 (Oxford, 2002).
FRITZ, G.,“Topics in the history of dialogue forms”, in: Historical Pragmatics, ed. A.H.
JUCKER (Amsterdam and Philadelphia, PA, 1995), pp. 469-498.
GALLAGHER, J., “The Italian London of John North: Cultural contact and linguistic en-
counter in early modern England”, Renaissance Quarterly (forthcoming).
GARDNER-CHLOROS, P., Code-Switching (Cambridge, 2009).
GENETTE, G., Paratexts: Thresholds of Interpretation, trans. J.E. LEWIN (Cambridge,
1997).
GIBBONS, A., and S. WHITELEY, Style and Cognition: An Introduction (Edinburgh,
forthcoming).
Bibliography 263

GIBSON, J., “Letters”, in: A Companion to English Renaissance Literature and Culture,
ed. M. HATTAWAY (New York, 2000), pp. 615-621.
GIBSON, J., “The queen’s two hands”, in: Representations of Elizabeth I in Early Mod-
ern Culture, ed. A. PETRINA and L. TOSI (New York, 2011), pp. 47-65.
GILLIAM, T., R.C. WILSON, and J.A. CLARK, “Scribe identification in medieval English
manuscripts”, in: Pattern Recognition (ICPR), 20th International Conference
(2010), pp. 1880-1883 <http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=
5597227>.
GOLDBERG, J., Writing Matter: From the Hands of the English Renaissance (Stanford,
CA, 1990).
GRANT-RUSSELL, P., “The influence of French on Quebec English: Motivation for lex-
ical borrowing and integration of loanwords”, LACUS Forum 25 (1998), pp. 473-
486.
GRIFFIN, C., “Instruction and information from manuscript to print: Some English litera-
ture, 1400-1650”, Literature Compass 10 (2013), pp. 667-676.
GUMBERT, J. P., “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 and New Castle, DE, 1999), pp. 7-13.
HASSNER, T., M. REHBEIN, P. STOKES, and L. WOLF, “Computation and palaeography:
Potentials and limits”, Dagstuhl Manifestos 2 (2012), pp. 14-35.
HAUGEN, E., “Dialect, language, nation”, in: The Ecology of Language: Essays by
Einar Haugen, ed. A.S. DIL (Stanford, CA, 1966 [1972]), pp. 237-254.
HAUGEN, E., “National and international languages”, in: The Ecology of Language, pp.
255-264.
HEALY, M., “Discourses of the plague in early modern London”, in: Epidemic Disease
in London, ed. J.A.I. CHAMPION (London, 1993: Centre for Metropolitan History
Working Papers Series 1), pp. 19-34. <http://www.history.ac.uk/cmh/epiheal.
html>.
HELLINGA, L., Caxton in Focus: The Beginning of Printing in England (London, 1982).
HELLINGA, L., “Manuscripts in the hands of printers”, in: Manuscripts in the Fifty Years
after the Invention of Printing, ed. J.B. TRAPP (London, 1983), pp. 3-11.
HELLINGA, L., “Notes on the order of setting a fifteenth-century book”, Quaerendo 4
(1974), pp.64-69.
HELLINGA, L., “Printing”, in: The Cambridge History of the Book in Britain, 3, 1400-
1557, ed. L. HELLINGA and J.B. TRAPP (Cambridge, 1999), pp. 65-108.
HELLINGA, L., “Printing types and other typographical materials”, in: Catalogue of
Books Printed in the XVth Century Now in the British Library, Part XI, England,
ed. L. HELLINGA (’t Goy-Houten, 2007).
HELLINGA, L., William Caxton and Early Printing in England (London, 2010).
HELLINGA, L., “William Caxton, Colard Mansion, and the Printer in Type 1”, Bulletin
du bibliophile (2011), pp. 86-114.
264 Bibliography

HELLINGA, L., Texts in Transit: Manuscript to Proof and Print in the Fifteenth Century
(Leiden, 2014).
HELLINGA, W., and L. HELLINGA, The Fifteenth-Century Printing Types of the Low
Countries, trans. D.A.S. REID (Amsterdam, 1966).
HIIPPALA, T., Modelling the Structure of a Multimodal Artefact (doctoral thesis, Uni-
versity of Helsinki, 2013) <http://urn.fi/URN:ISBN:978-952-10-9427-9>.
HIIPPALA, T., The Structure of Multimodal Documents: An Empirical Approach (New
York, 2015).
HILLS, C., “The archaeological context of runic finds”, in: Old English Runes and Their
Continental Background, ed. A. BAMMESBERGER (Heidelberg, 1991), pp. 41-59.
HILTUNEN, R., and M. PEIKOLA, “Trial discourse and manuscript context: Scribal pro-
files in the Salem witchcraft records”, Journal of Historical Pragmatics 8.1 (2007),
pp. 43-68.
HINES, J., “The runic inscriptions of early Anglo-Saxon England”, in: Britain 400-600:
Language and History, ed. A. BAMMESBERGER and A. WOLLMANN (Heidelberg,
1990), pp. 437-455.
HINES, J., “Some observations on the runic inscriptions of early Anglo-Saxon England”,
in: Old English Runes and Their Continental Background, ed. A. BAMMESBERGER
(Heidelberg, 1991), pp. 61-83.
HINES, J., “New light on literacy in eighth-century East Anglia: A runic inscription from
Baconsthorpe, Norfolk”, Anglia 129 (2011), pp. 281-296.
HOLDER, N., “Inscriptions, writing and literacy in Saxon London”, Transactions of the
London and Middlesex Archaeological Society 49 (1998), pp. 81-97.
HOROBIN, S., “The criteria for scribal attribution: Dublin, Trinity College MS 244 recon-
sidered”, Review of English Studies 60 (2009), pp. 371-381.
HOROBIN, S., “Mapping the words”, in: The Production of Books in England, 1350-
1500, ed. A. GILLESPIE and D. WAKELIN (Cambridge, 2011), pp. 59-78.
HOTCHKISS, V., and F.C. ROBINSON, English in Print from Caxton to Shakespeare to
Milton (Urbana, IL, 2008).
HOWLETT, D.R., “A corrected form of the reconstructed Ruthwell Crucifixion Poem”,
Studia Neophilologica 80 (2008), pp. 255-257.
IOPPOLO, G., “Early modern handwriting”, in: A New Companion to English Renais-
sance Literature and Culture, ed. M. HATTAWAY, 2 vols. (Chichester, 2010), 1, pp.
177-189.
JAJDELSKA, E., Silent Reading and the Birth of the Narrator (Toronto, 2007).
JENKINSON, H., “English current writing and early printing”, The Library: Transactions
of the Bibliographical Society 13 (1913), pp. 273-295.
JENKINSON, H., “Elizabethan handwritings: A preliminary sketch”, The Library, 4th
ser., 3 (1922), pp. 1-50.
Bibliography 265

JEWITT, C., “Introduction: Handbook rationale, scope and structure”, in: The Routledge
Handbook of Multimodal Analysis, ed. C. JEWITT (Abingdon, [2009] 2011), pp. 1-7.
JOHN, J.J., “Latin paleography”, in: Medieval Studies: An Introduction, ed. J.M.
POWELL (Syracuse, NY, 1992), pp. 1-68.
JONES, P.M., “Medical literacies and medical culture in early modern England”, in:
Medical Writing in Early Modern English, ed. I. TAAVITSAINEN and P. PAHTA
(Cambridge, 2011), pp. 30-43.
JUCKER, A.H., and P. PAHTA, “Communicating manuscripts: Authors, scribes, readers,
listeners and communicating characters”, in: Communicating Early English Manu-
scripts, ed. A.H. JUCKER and P. PAHTA (Cambridge, 2011), pp. 2-10.
JUCKER, A.H., and I. TAAVITSAINEN, English Historical Pragmatics (Edinburgh, 2013).
KAISLANIEMI, S., “The early English East India Company as a community of practice:
Evidence of multilingualism”, in: Merchants of Innovation: The Language of Trad-
ers, ed. E.-M. WAGNER, B. BEINHOFF, and B. OUTHWAITE (Berlin, 2017: Studies in
Language Change 15), pp. 132-157.
KAISLANIEMI, S., “Encountering and appropriating the other: East India Company mer-
chants and foreign terminology”, in: The Language of Daily Life in England 1450-
1800, ed. A. NURMI, M. NEVALA, and M. PALANDER-COLLIN (Amsterdam, 2009:
Pragmatics and Beyond: New Series 183), pp. 219-251.
KAISLANIEMI, S., Material Communities of Practice: The Letters of Richard Cocks,
English Merchant and Intelligencer, 1603-1609 (doctoral thesis, University of
Helsinki, forthcoming).
KARKOV, C.E., The Art of Anglo-Saxon England (Woodbridge, 2011: Boydell Studies
in Medieval Art and Architecture).
KARKOV, C.E., “Art and writing: Voice, image, object”, in: The Cambridge History of
Early Medieval English Literature, ed. C.A. LEES (Cambridge, 2012: New Cam-
bridge History of English Literature), pp. 145-153.
KATAOKA, K., “Affect and letter-writing: Unconventional conventions in casual writing
by young Japanese women”, Language in Society 26 (1997), pp. 103-136.
KENDALL, C.B., “From sign to vision: The Ruthwell Cross and ‘The Dream of the
Rood’”, in: The Place of the Cross in Anglo-Saxon England, ed. C.E. KARKOV, S.
LARRATT KEEFER and K.L. JOLLY (Woodbridge, 2006), pp. 129-144.
KERLING, J., Chaucer in Early English Dictionaries: The Old-Word Tradition in Eng-
lish Lexicography down to 1721 and Speght’s Chaucer Glossaries (Leiden, 1979).
KOPACZYK, J., “In search of a framework for written code-switching: Administrative
multilingualism on the page in early modern Poland”, in preparation.
KRESS, G., and T. VAN LEEUWEN, “Colour as a semiotic mode: Notes for a grammar of
colour”, Visual Communication 1 (2002), pp. 343-368.
KUSKIN, W., Symbolic Caxton: Literary Culture and Print Capitalism (Notre Dame,
IN, 2008).
266 Bibliography

KYTÖ, M., Manual to the Diachronic Part of The Helsinki Corpus of English Texts:
Coding Conventions and Lists of Source Texts, 3rd edn. (Helsinki, 1996) <clu.uni.
no/icame/manuals/HC/INDEX.HTM>.
LABOV, W., and J. WALETZKY, “Narrative analysis: Oral versions of personal experi-
ence”, in: Essays on the Verbal and Visual Arts, ed. J. HELM (Seattle, WA, 1967).
LABOV, W., “Some further steps in narrative analysis”, The Journal of Narrative and
Life History 7 (1997), pp. 395-415.
LAKOFF, G., and M. JOHNSON, Metaphors We Live By (Chicago, IL, 1980).
Language Mixing and Code-Switching in Writing: Approaches to Mixed-Language
Written Discourse, ed. M. SEBBA, S. MAHOOTIAN, and C. JONSSON (New York and
London, 2012).
LASS, R. “Ut custodiant litteras: Editions, corpora and witnesshood”, in: Methods and
Data in English Historical Dialectology, ed. M. DOSSENA and R. LASS (Bern,
2004: Linguistic Insights 16), pp. 21-48.
LAUERBACH, G.E., s.v. “Emphasis” in: Handbook of Pragmatics Online, ed. J.-O. ÖST-
MAN and J. VERSCHUEREN (Amsterdam, 2002-2015), DOI: 10.1075/hop.8.emp1.
LEECH, G., and M. SHORT, Style in Fiction: A Linguistic Introduction to English Fic-
tional Prose (Harlow, 2007).
LEIWO, M. “From contact to mixture: Bilingual inscriptions from Italy”, in Bilingualism
in Ancient Society: Language Contact and the Written Text (Oxford, 2002), pub-
lished by Oxford Scholarship Online: January 2010, DOI: 10.1093/acprof:oso/
9780199245062.003.0007.
LERER, S., Literacy and Power in Anglo-Saxon Literature (Lincoln, NE, 1991: Regents
Studies in Medieval Culture).
LEUNG, G.Y., “Hong Kong university students’ language use in blogs: Tensions be-
tween creativity and prescriptivism (‘omg my grammar is rubbish’)”, in: Sino-
graphic Languages: The Past, Present, and Future of Script Reform, ed. V.H.
MAIR (= Sino-Platonic Papers 189 (2009)), pp. 30-62, <sino-platonic.org>.
LOOIJENGA, T., Texts and Contexts of the Oldest Runic Inscriptions (Leiden, 2003).
LUBORSKY, R.S., “Connections and disconnections between images and texts: The case
of secular Tudor book illustrations”, Word & Image 3 (1987), pp. 74-85.
LUNA, P., “Not just a pretty face: The contribution of typography to lexicography”, in:
Proceedings of the Eleventh EURALEX International Congress, EURALEX 2004, ed.
G. WILLIAMS and S. VESSIER, 3 vols. (Lorient, 2004), pp. 847-857.
LYONS, M., Books: A Living History (London, 2013).
MCARTHUR, T., The English Languages (Cambridge, 1998).
MCATEER, E., “Typeface emphasis and information focus in written language”, Applied
Cognitive Psychology 6 (1992), pp. 345-359.
MCCONCHIE, R.W., “The real Richard Howlet”, in: Words and Dictionaries from the
British Isles in Historical Perspective, ed. J. CONSIDINE and G. IAMARTINO (New-
castle upon Tyne, 2007), pp. 39-49.
Bibliography 267

MCCONCHIE, R.W., “Some reflections on early modern printed title-pages”, in: Princi-
ples and Practices for the Digital Editing of Diachronic Data, ed. A. MEURMAN-
SOLIN and J. TYRKKÖ (Helsinki, 2013: Studies in Variation, Contacts and Change
in English 14), available at <http://www.helsinki.fi/varieng/series/volumes/14/
mcconchie/>.
MACHAN, T.W., “The visual pragmatics of code-switching in Late Middle English litera-
ture”, in: Code-Switching in Early English, ed. H. SCHENDL and L. WRIGHT (Ber-
lin, 2011), pp. 303-333.
MCINTOSH, A., “Scribal profiles from Middle English texts”, in: Middle English Dia-
lectology: Essays on Some Principles and Problems, ed. A. MCINTOSH, M.L. SAM-
UELS, and M. LAING (Aberdeen, 1989), pp. 32-54, earlier published in Neuphi-
lologische Mitteilungen 76 (1975), pp. 218-235.
MCINTOSH, A., M.L. SAMUELS, and M. BENSKIN, A Linguistic Atlas of Late Mediaeval
English, 4 vols. (Aberdeen, 1986).
MACKAY, Æ.J.G., The Historie and Cronicles of Scotland: Robert Lindesay of Pitscot-
tie (Edinburgh and London, 1899).
MCKERROW, R.B., An Introduction to Bibliography for Literary Students (Oxford,
1928).
MCKERROW, R.B., and F.S. FERGUSON, Title-Page Borders Used in England & Scot-
land 1485-1640 (London, 1932).
MCSHANE, A., “Typography matters: Branding ballads and gelding curates in Stuart
England”, in: Book Trade Connections from the Seventeenth to the Twentieth Cen-
turies, ed. J. HINKS and C. ARMSTRONG (New Castle, DE, and London, 2008), pp.
19-44.
MAK, B., How the Page Matters (Toronto, 2011).
MÄKINEN, M. “Herbal recipes and recipes in herbals – Intertextuality in early English
medical writing”, in: Medical and Scientific Writing in Late Medieval English, ed.
I. TAAVITSAINEN and P. PAHTA (Cambridge, 2004), pp. 144-173.
MANGUEL, A., A History of Reading (London, 1997).
MANOVICH, L., “How to compare one million images?”, in: Understanding Digital
Humanities, ed. D. BERRY (New York, 2012), pp. 249-278.
MANOVICH, L., “Media visualization: Visual techniques for exploring large media col-
lections”, in: The International Encyclopaedia of Media Studies, 6, Media Studies
Futures, ed. K. GATES (London, 2013), DOI: 10.1002/9781444361506.wbiems
144.
Manuscript Studies and Codicology: Theory and Practice, ed. M. PEIKOLA and M.
KYTÖ (= Studia Neophilologica 86, Supplement 1 (2014)).
MARCUS, I., The Linguistics of Spoken Communication in Early Modern English Writ-
ing: Exploring Bess of Hardwick’s Manuscript Letters (London, forthcoming).
268 Bibliography

MARTTILA, V., Creating Digital Editions for Corpus Linguistics: The Case of Potage
Dyvers, a Family of Six Middle English Recipe Collections (doctoral thesis, Uni-
versity of Helsinki, 2014) <http://urn.fi/URN:ISBN:978-951-51-0060-3>
MEURMAN-SOLIN, A., “Taxonomisation of features of visual prosody”, in: Principles
and Practices for the Digital Editing and Annotation of Diachronic Data, ed. A.
MEURMAN-SOLIN and J. TYRKKÖ (Helsinki, 2013: Studies in Variation, Contacts
and Change in English 14) <http://www.helsinki.fi/varieng/series/volumes/14/
meurman-solin_c/>.
MEURMAN-SOLIN, A., “Visual prosody in manuscripts of letters in the study of syntax
and discourse”, in: Principles and Practices for the Digital Editing and Annotation
of Diachronic Data, ed. A. MEURMAN-SOLIN and J. TYRKKÖ (Helsinki, 2013:
Studies in Variation, Contacts and Change in English 14) <http://www.helsinki.fi/
varieng/journal/volumes/14/meurman-solin_a/>.
MICHEL, J.-B., et al., “Quantitative analysis of culture using millions of digitized
books”, Science 331 (2011), pp. 176-182.
MIKKELI, H., and V. MARTTILA,“Change and continuity in early modern medicine
(1500-1700)”, in: Early Modern English Medical Texts: Corpus Description and
Studies, ed. I. TAAVITSAINEN and P. PAHTA (Amsterdam and Philadelphia, PA,
2010), pp. 13-28.
MILLETT, B., “What is mouvance?”, Wessex Parallel Web Texts (24 Aug. 2011),
<http://www.southampton.ac.uk/~wpwt/mouvance/mouvance>.
MISH, C.C., “Black letter as a social discriminant in the seventeenth century”, Proceed-
ings of the Modern Language Association 68 (1953), pp. 627-630.
MOORE, C., Quoting Speech in Early English (Cambridge, 2011).
MOORE, C., “Visual pragmatics: Speech presentation and Middle English manuscripts”,
in: The Cambridge Handbook of English Historical Linguistics, ed. M. KYTÖ and
P. PAHTA (Cambridge, 2016), pp. 481-496.
MORETTI, F., Graphs, Maps, Trees: Abstract Models for a Literary History (London,
2005).
MORISON, S., “The development of hand-writing: An outline”, in: ID., Selected Essays
on the History of Letter-Forms in Manuscript and Print, ed. D. MCKITTERICK, 2
vols. (Cambridge, 1981), 1, pp. 161-176, first publ. as introduction to A. HEAL, The
English Writing-masters and Their Copy-books, 1570-1800 (1931), pp. XXIII-XL.
MORISON, S., “On script types”, in ID., Selected Essays, 1, pp. 47-80, first publ. in: The
Fleuron 4 (1925), pp. 1-39.
Multilingual Practices in Language History: New Perspectives, ed. P. PAHTA, J. SKAF-
FARI, and L. WRIGHT (in preparation).
MUZERELLE, D., Vocabulaire codicologique: Répertoire méthodique des termes fran-
çais relatifs aux manuscrits (Paris, 1985).
Bibliography 269

NEEDHAM, P., The Printer & the Pardoner: An Unrecorded Indulgence Printed by
William Caxton for the Hospital of St. Mary Rounceval, Charing Cross (Washing-
ton, DC, 1986).
NELSON, V., “Problems of transcription in the ‘Speculum Vitae’ MSS”, Scriptorium 31
(1977), pp. 254-259.
NEVALA, M., Address in Early English Correspondence: Its Forms and Socio-Prag-
matic Functions (Helsinki, 2004: Mémoires de la Société Néophilologique de
Helsinki 64).
NEVALA, M., “Accessing politeness axes: Forms of address and terms of reference in
early English correspondence”, Journal of Pragmatics 36 (2004), pp. 2125-2160.
NEVALAINEN, T., and H. RAUMOLIN-BRUNBERG, Historical Sociolinguistics: Language
Change in Tudor and Stuart England (London and New York, 2003).
NICHOLS, S.G., “Introduction: Philology in a manuscript culture”, Speculum 65 (1990),
pp. 1-10.
NIEDZWì IEDZì, J., “Cyrillic and Latin script in late medieval Vilnius”, in: Uses of the
Written Word in Medieval Towns: Medieval Urban Literacy II, ed. M. MOSTERT
and A. ADAMSKA (Turnhout, 2014: Utrecht Studies in Medieval Literacy 28), pp.
99-116.
NILES, J.D. Old English Enigmatic Poems and the Play of the Texts (Turnhout, 2006).
NOORDZIJ, G., “Broken scripts and the classification of typefaces”, Journal of Typo-
graphic Research [Visible Language] 4 (1970), pp. 213-240.
NORDBY, K.J., “Ráð þat, if you can!”, Futhark 3 (2012), pp. 81-88.
NORRICK, N.R., “Conversational storytelling”, in: The Cambridge Companion to Nar-
rative, ed. D. HERMAN (Cambridge, 2007), pp. 127-141.
NURMI, A., Manual for the Corpus of Early English Correspondence Sampler CEECS
(Helsinki, 1998) <clu.uni.no/icame/manuals/CEECS/INDEX.HTM>.
NURMI, A., and P. PAHTA, “Multilingual practices in the language of the law: Evidence
from the Lampeter corpus”, in: Ex Philologia Lux: Essays in Honour of Leena
Kahlas-Tarkka, ed. J. TYRKKÖ, O. TIMOFEEVA, and M. SALENIUS (Helsinki, 2013:
Mémoires de la Société Néophilologique de Helsinki 90), pp. 187-205.
NURMI, A., and P. PAHTA, “Social stratification and patterns of code-switching in early
English letters”, Multilingua 23 (2004), pp. 417-456.
Ó CARRAGÁIN, É., Ritual and the Rood: Liturgical Images and the Old English Poems
of the Dream of the Rood Tradition (London, 2005).
OED = The Oxford English Dictionary Online <www.oed.com>.
OKASHA, E., “Anglo-Saxon inscribed rings”, Leeds Studies in English 34 (2003), pp.
29-45.
OKASHA, E., “Script-mixing in Anglo-Saxon inscriptions”, in: Writing and Texts in
Anglo-Saxon England, ed. A.R. RUMBLE (Woodbridge, 2006: Publications of the
Manchester Centre for Anglo-Saxon Studies 5), pp. 62-70.
ONG, W., Orality and Literacy, 2nd edn. (Abingdon, 2002).
270 Bibliography

ORGEL, S., “Textual icons: Reading early modern illustrations”, in: The Renaissance
Computer: Knowledge Technology in the First Age of Print, ed. N. RHODES and J.
SAWDAY (London and New York, 2000), pp. 59-94.
PAGE, R.I., An Introduction to English Runes, 2nd edn. (Woodbridge, 1999).
PAGE, R.I., “Quondam et futurus”, in: ID., Runes and Runic Inscriptions: Collected
Essays on Anglo-Saxon and Viking Runes, ed. D. PARSONS (Woodbridge, 1995),
pp. 1-16.
PAGE, R.I., “Roman and runic on St Cuthbert’s coffin”, in: St Cuthbert, His Cult and
His Community to AD 1200, ed. G. BONNER, D. ROLLASON, and C. STANCLIFFE
(Woodbridge, 1989), pp. 257-265.
PAHTA, P., and A. NURMI, “Code-switching in the Helsinki Corpus: A thousand years
of multilingual practices”, in: Medieval English and its Heritage: Structure, Mean-
ing and Mechanisms of Change, ed. N. RITT, H. SCHENDL, C. DALTON-PUFFER, and
D. KASTOVSKY (Frankfurt, 2006), pp. 203-220.
PAHTA, P., and A. NURMI, “Multilingual discourse in the domain of religion in medieval
and early modern England: A corpus approach to research on historical code-
switching”, in: Code-Switching in Early English, ed. H. SCHENDL and L. WRIGHT
(Berlin and Boston, MA, 2011: Topics in English Linguistics 76), pp. 219-251.
PAHTA, P., and M. RATIA, “Category 2: Treatises on specific topics”, in: Early Modern
English Medical Texts: Corpus Description and Studies, ed. I. TAAVITSAINEN and
P. PAHTA (Amsterdam and Philadelphia, PA, 2010), pp. 73-100.
PAHTA, P., and I. TAAVITSAINEN, “Introducing early modern English medical texts”, in:
Early Modern English Medical Texts: Corpus Description and Studies, ed. I. TAA-
VITSAINEN and P. PAHTA (Amsterdam and Philadelphia, PA, 2010), pp. 1-8.
PARKES, M.B., English Cursive Book Hands, 1250-1500 (Oxford, 1969; Berkeley,
1980).
PARKES, M.B., “The influence of the concepts of ordinatio and compilatio on the devel-
opment of the book”, in: Medieval Learning and Literature: Essays Presented to
Richard William Hunt, ed. J.J.G. ALEXANDER and M.T. GIBSON (Oxford, 1976),
pp. 115-141 and Plates IX-XVI.
PARKES, M.B., Pause and Effect: An Introduction to the History of Punctuation in the
West (Aldershot, 1992; Berkeley and Los Angeles, 1993).
PARKES, M.B., Their Hands Before Our Eyes: A Closer Look at Scribes: The Lyell
Lectures Delivered in the University of Oxford 1999 (Aldershot, 2008).
PARSONS, D.N., Recasting the Runes: The Reform of the Anglo-Saxon “Futhorc”
(Uppsala, 1999).
PARTRIDGE, S., “Designing the page”, in: The Production of Books in England 1350-
1500, ed. A. GILLESPIE and D. WAKELIN (Cambridge, 2011), pp. 79-103.
PARTRIDGE, W.J., “The use of William Caxton’s Type 3 by John Lettou and William
De Machlinia in the printing of their ‘Yearbook 35 Henry VI, c.1481-1482’”, The
British Library Journal (1983), pp. 56-65.
Bibliography 271

PEARSON, D., “What can we learn by tracking multiple copies of books”, in: Books on
the Move: Tracking Copies through Collections and the Book Trade, ed. R. MYERS
et al. (New Castle, DE, and London, 2007), pp. 17-38.
PEIKOLA, M., “Guidelines for consumption: Scribal ruling patterns and designing the
mise-en-page in later medieval England”, in: Manuscripts and Printed Books in
Europe 1350-1500: Packaging, Presentation and Consumption, ed. E. CAYLEY
and S. POWELL (Liverpool, 2013), pp. 14-31.
PETTEGREE, A., The Book in the Renaissance (New Haven, CT, and London, 2010).
PETTI, A.G., English Literary Hands from Chaucer to Dryden (London, 1977).
PILZ, T., A. ERNST-GERLACH, S. KEMPKEN, P. RAYSON, and D. ARCHER, “The identifi-
cation of spelling variants in English and German historical texts: Manual or auto-
matic?”, Literary and Linguistic Computing 23 (2008), pp. 65-72.
PORTER STORK, N., Through a Gloss Darkly: Aldhelm’s Riddles in the British Library
MS Royal 12.C.XXIII (Toronto, 1990).
A Prague School Reader on Esthetics, Literary Structure, and Style, ed. and trans. P.
GARVIN (Washington, DC, 1964).
Principles and Practices for the Digital Editing and Annotation of Diachronic Data,
ed. A. MEURMAN-SOLIN and J. TYRKKÖ (Helsinki, 2013: Studies in Variation, Con-
tacts and Change in English 14), <www.helsinki.fi/varieng/series/volumes/14>.
PROOT, G., “Converging design paradigms: Long-term evolutions in the layout of title
pages of Latin and vernacular editions published in the Southern Netherlands,
1541-1660”, The Papers of the Bibliographical Society of America 108 (2014),
269-305.
PULSIANO, P., “The prefatory matter of London, British Library, Cotton Vitellius
E.XVIII”, in: Anglo-Saxon Manuscripts and Their Heritage, ed. E.M. TREHARNE
and P. PULSIANO (Brookfield, VT, 1998), pp. 85-116.
RASBAND, W.S., “ImageJ”, U.S. National Institutes of Health, Bethesda, Maryland,
USA (1997-2014) <http://imagej.nih.gov/ij/>.
RATIA, M., Texts “Con and Pro”: The Early Modern Medical Controversy over To-
bacco (Helsinki, 2011: Mémoires de la Société Néophilologique 82).
RATIA, M., “Investigating genre through title-pages: Plague treatises of the Stuart period
in focus”, in: Principles and Practices for the Digital Editing of Diachronic Data,
ed. A. MEURMAN-SOLIN and J. TYRKKÖ (Helsinki, 2013: Studies in Variation, Con-
tacts and Change in English 14) < http://www.helsinki.fi/varieng/series/volumes/
14/ratia/>.
RATIA, M., and C. SUHR, “Medical pamphlets: Controversy and advertising”, in: Early
Modern English Medical Texts: Corpus Description and Studies, ed. I. TAAVITSAI-
NEN and P. PAHTA (Amsterdam and Philadelphia, PA, 2010), pp. 180-201.
RAUMOLIN-BRUNBERG, H., and T. NEVALAINEN, “Historical sociolinguistics: The Cor-
pus of Early English Correspondence”, in: Creating and Digitizing Language Cor-
272 Bibliography

pora, 2: Diachronic Databases, ed. J.C. BEAL, K.P. CORRIGAN, and H.L. MOISL
(Houndsmills, 2007), pp. 148-171.
RAYNER, K., and A. POLLATSEK, The Psychology of Reading (Hillsdale, NJ, 1989).
REDDY, M.J., “The conduit metaphor: A case of frame conflict in our language about
language,” in: Metaphor and Thought, ed. A. ORTONY (Cambridge, 1979), pp.
284-310.
REIMER, B., B. MEHLER, and J.F. COUGHLIN, “An evaluation of typeface design in a
text-rich automotive user interface”, Massachusetts Institute of Technology AgeLab
white paper 2012-12 (2012) <agelab.mit.edu>.
REIMER, S.R., “Unbinding Lydgate’s Lives of Ss. Edmund and Fremund”, in: The Book
Unbound: Editing and Reading Medieval Manuscripts and Texts, ed. S. ECHARD
and S. PARTRIDGE (Toronto, 2004), pp. 169-189.
RHODES, N., “Status anxiety and English Renaissance translation”, in: Renaissance
Paratexts, ed. H. SMITH and L. WILSON (Cambridge, 2011), pp. 107-120.
ROBINSON, P., “Materials: Paper and type”, in: A Companion to the Early Printed Book
in Britain 1476-1558, ed. V. GILLESPIE and S. POWELL (Cambridge, 2014), pp. 61-
74.
SAENGER, P., Space Between Words: The Origins of Silent Reading (Stanford, CA,
1997).
SALMI, H., “Visual presentation of English Body and Soul Debates”, in: Manuscript
Studies and Codicology: Theory and Practice, ed. M. PEIKOLA and M. KYTÖ (=
Studia Neophilologica 86, Supplement 1 (2014)), pp. 144-156.
SALMON, V., “Orthography and punctuation”, in: The Cambridge History of the English
Language, ed. R. LASS, 5 vols. (Cambridge, 1999), 3, pp. 13-55.
SCHMIDT, C.K.. “Loanwords in Japanese”, in: Loanwords in the World’s Languages: A
Comparative Handbook, ed. U. TADMOR and M. HASPELMATH (Berlin, 2009), pp.
545-574.
SCHNEIDER, C.A., W.S. RASBAND, and K.W. ELICEIRI, “NIH Image to ImageJ: 25 years
of image analysis”, Nature Methods 9 (2012), pp. 671-675.
SCHULZ, H.C., “The teaching of handwriting in Tudor and Stuart times”, Huntington
Library Quarterly 6 (1943), pp. 381-425.
SCRAGG, D.G., A History of English Spelling (Manchester, 1974: Mont Fellick Series
3).
Scripts Beyond Borders: A Survey of Allographic Traditions in the Euro-Mediterra-
nean World, ed. J. DEN HEIJER, T. PATARIDZE and A.B. SCHMIDT (Louvain-la-
Neuve, 2014: Publications de l’Institut Orientaliste de Louvain 62).
SEBBA, M., “Multilingualism in written discourse: An approach to the analysis of multi-
lingual texts”, International Journal of Bilingualism 17 (2013), pp. 97-118.
Semantics versus Pragmatics, ed. Z. GENDLER SZABÓ (Oxford, 2005).
Bibliography 273

SHERMAN, W., “On the threshold: Architecture, paratext and early print culture”, in
Agent of Change: Print Culture Studies after Elizabeth L. Eisenstein, ed. S. AL-
CORN, E.N. LINDQUIST and E.F. SHELVING (Amherst, MA, 2007), pp. 67-81.
SHEVLIN, E.F., “‘To reconcile book and title, and make’em kin to one another’: The
evolution of the title’s contractual functions”, Book History 2 (1999), pp. 42-77.
SHORT, M., Exploring the Language of Poems, Plays and Prose (Harlow, 1996).
SHORT, M., “Graphological deviation, style variation and point of view in Marabou
Stork Nightmares by Irvine Welsh”, Journal of Literary Studies 15 (1999), pp.
305-323.
SLACK, P., The Impact of Plague in Tudor and Stuart England (London, Boston, BA,
Melbourne, and Henley, 1985).
SLIGHTS, W.W.E., Managing Readers: Printed Marginalia in English Renaissance
Books (Ann Arbor, MI, 2001).
SLOWIACZEK, M.L. and C. CLIFTON, “Subvocalization and reading for meaning”, Jour-
nal of Verbal Learning and Verbal Behaviour 19 (1980), pp. 573-582.
SMITH, J., “Narrative: Sociolinguistic research”, in: Encyclopedia of Language and
Linguistics, ed. K. BROWN, R.E. ASHER, and J.M.Y. SIMPSON (Amsterdam, 2006),
pp. 473-476.
SMITH, J.J., A Historical Study of English: Function, Form and Change (London,
1996).
SMITH, J.J., “Punctuating Mirk’s Festial: A Scottish text and its implications”, in:
Preaching the Word in Manuscript and Print in Late Medieval England, ed. M.W.
DRIVER and V. O’MARA (Turnhout, 2013), pp. 161-192.
SMITH, M.M., The Title-Page: Its Early Development 1460-1510 (London, 2000).
SNIJDERS, T., Manuscript Communication: Visual and Textual Mechanics of Communi-
cation in Hagiographical Texts from the Southern Low Countries, 900-1200
(Turnhout, 2015: Utrecht Studies in Medieval Literacy 32).
SÖNMEZ, M., “Perceived and real differences between men’s and women’s spellings of
the early to mid-seventeenth century”, in: The History of English in a Social Con-
text: A Contribution to Historical Sociolinguistics, ed. D. KASTOVSKY and A.
METTINGER (Berlin, 2000: Trends in Linguistics, Studies and Monographs 129),
pp. 405-440.
STEIN, G., Sir Thomas Elyot as Lexicographer (Oxford, 2014).
STEINBERG, S.H., Five Hundred Years of Printing, 2nd edn. (Harmondsworth, 1961).
STROHM, P., “‘Passioun, lyf, miracle, legende’: Some generic terms in Middle English
hagiographical narrative”, The Chaucer Review 10 (1975), pp. 62-75, 154-171.
SUHR, C., Publishing for the Masses: Early Modern English Witchcraft Pamphlets
(Helsinki, 2011: Mémoires de la Société Néophilologique 83).
SUHR, C., “News and relations: Textual labels in the titles of early modern news pam-
phlets”, in: Sociocultural Dimensions of Lexis and Text in the History of English,
274 Bibliography

ed. P. PETRÉ, H. CUYCKENS and F. D’HOEDT (Amsterdam and Philadelphia, PA,

forthcoming).
SUHR, C., and J. TYRKKÖ, “Investigating the cultures and practices of print: Corpus-
based taxonomies of medical title-pages 1500-1800”, a conference presentation at
ICAME37, Hong Kong, 26 May 2016.
SULLIVAN, C., “Disposable elements? Indications of genre in early modern titles”, Mod-
ern Language Review 102 (2007), pp. 641-653.
SZABO, V.E., Monstrous Fishes and the Mead-Dark Sea: Whaling in the Medieval
North Atlantic (Leiden, 2008).
TAAVITSAINEN, I., “Changing conventions of writing: The dynamics of genres, text
types, and text traditions”, European Journal of English Studies 5 (2001), pp. 139-
150.
TAAVITSAINEN, I., “Discourse and genre dynamics in early modern English medical
writing”, in: Early Modern English Medical Texts: Corpus Description and Stud-
ies, ed. I. TAAVITSAINEN and P. PAHTA (Amsterdam and Philadelphia, PA, 2010),
pp. 29-53.
TAAVITSAINEN, I., Middle English Lunaries: A Study of the Genre (Helsinki, 1998:
Mémoires de la Société Néophilologique 47).
TANNENBAUM, S.A., The Handwriting of the Renaissance (New York, 1930).
TANSELLE, G.T., Bibliographical Analysis: A Historical Introduction (Cambridge,
2009).
THOMAS, K., “The meaning of literacy in early modern England”, in: The Written
Word: Literacy in Transition: Wolfson College Lectures 1985, ed. G. BAUMANN
(Oxford, 1986), pp. 97-131.
TILGHMAN, B.C., “Writing in tongues: Mixed scripts and style in insular art”, in: Insu-
lar & Anglo-Saxon Art and Thought in the Early Medieval Period, ed. C. HOURI-
HANE (University Park, PA, 2011: Index of Christian Art Occasional Papers 13),
pp. 92-108.
TOKUNAGA, S., “A digital approach to the history of the book: The case of Caxton”,
Poetica (Tokyo) 60 (2003), pp. 65-76.
TOKUNAGA, S., “Early English printing and the hands of compositors”, International
Journal of English Studies 5 (2005) pp. 149-160.
TOTARO, R., Suffering in Paradise: The Bubonic Plague in English Literature from
More to Milton (Pittsburgh, PA, 2005).
TRUELOVE, A., “Commanding communications: The fifteenth-century letters of the
Stonor women”, in: Early Modern Women’s Letter Writing, 1450-1700, ed. J.
DAYBELL (Basingstoke, 2001), pp. 42-58.
TUOMINEN, J., “‘Trifling Shews of Learning’? Patterns of code-switching in English
sermons 1640-1740”, in: Multilingual Practices in Language History, ed. P. PAH-
TA, J. SKAFFARI and L. WRIGHT (in preparation).
Bibliography 275

TYRKKÖ, J., “Printing houses as communities of practice: Orthography in early modern

medical books”, in: Communities of Practice in the History of English, ed. J.
KOPACZYK and A.H. JUCKER (Amsterdam and Philadelphia, PA, 2013), pp. 151-
176.
TYRKKÖ, J., V. MARTTILA, and C. SUHR, “The Culpeper Project: Digital editing of title-
pages”, in: Principles and Practices for the Digital Editing of Diachronic Data, ed.
A. MEURMAN-SOLIN and J. TYRKKÖ (Helsinki, 2013: Studies in Variation, Con-
tacts and Change in English 14), < http://www.helsinki.fi/varieng/series/volumes/
14/tyrkko_marttila_suhr/>.
UPDIKE, D.B., Printing Types, Their History, Forms, and Use: A Study in Survivals, 1
(Cambridge, MA, 1922).
VAN LEEUWEN, T., The Language of Colour: An Introduction (London, 2011).
VAN LEEUWEN, T., “Towards a semiotics of typography”, Information Design Journal
14 (2006), pp. 139-155.
VAN LEEUWEN, T., “Typographic meaning”, Visual Communication 4 (2005), pp. 137-
143.
VARILA, M.-L., In Search of Textual Boundaries: A Case Study on the Transmission of
Scientific Writing in 16th-Century England (Turku, 2016).
Visual Text, ed. J. KENDALL, M. PORTELA, and G. WHITE (= European Journal of Eng-
lish Studies 17.1 (2013)).
WALKER, S., Typography & Language in Everyday Life: Prescriptions and Practices
(Harlow, 2001).
WALKER, T. and KYTÖ, M., “Features of layout and other visual effects in the source
manuscripts of An Electronic Text Edition of Depositions 1560-1760 (ETED)”, in:
Principles and Practices for the Digital Editing and Annotation of Diachronic
Data, ed. A. MEURMAN-SOLIN and J. TYRKKÖ (Helsinki, 2013: Studies in Varia-
tion, Contacts and Change in English 12) <http://www.helsinki.fi/varieng/series/
volumes/14/walker_kyto/>
WALLER, R., The Typographic Contribution to Language: Towards a Model of Typo-
graphic Genres and Their Underlying Structures (unpublished doctoral thesis,
University of Reading, 1987), <www.robwaller.org>.
WEAR, A., Knowledge and Practice in English Medicine, 1550-1680 (Cambridge,
2000), pp. 275, 277-278.
WERLICH, E., A Text Grammar of English (Heidelberg, 1976).
WERNER, S., “Where material book culture meets digital humanities”, blogpost in Wyn-
ken De Worde, posted 29 April 2012 at <http://sarahwerner.net/blog/2012/04/
where-material-book-culture-meets-digital-humanities/>.
WERTHEIM, A.H., “Visual conspicuity: A new simple standard, its reliability, validity
and applicability”, Ergonomics 53 (2010), pp. 421-442.
WHATLEY, E.G., A.B. THOMPSON, and R.K. UPCHURCH, “St. Andrew and the Three
Questions in the Scottish Legendary”, in: Saints’ Lives in Middle English Collec-
276 Bibliography

tions, ed. E.G. WHATLEY, A.B. THOMPSON, and R.K. UPCHURCH (Kalamazoo,
2004) <http://d.lib.rochester.edu/teams/text/whatley-saints-lives-in-middle-english-
collections-st-andrew-and-the-three-questions-introduction>.
WIECK, R.S., Painted Prayers: The Book of Hours in Medieval and Renaissance Art
(New York, 2004).
WILLIAMS, G., “‘yr Scribe Can proove no nessecarye Consiquence for you?’: The social
and linguistic implications of Joan Thynne’s using a scribe in letters to her son,
1607-1611”, in: Women and Writing, c. 1340-c. 1650: The Domestication of Print
Culture, ed. P. HARDMAN and A. LAWRENCE-MATHERS (Woodbridge, 2010), pp.
131-145.
WILLIAMS, G., “The language of early modern letters: A reader’s guide”, in: Bess of
Hardwick’s Letters: The Complete Correspondence, c.1550-1608, ed. A. WIGGINS,
A. BRYSON, D. STARZA SMITH, A. TIMMERMANN and G. WILLIAMS, University of
Glasgow, web development by K. ROGERS, University of Sheffield Humanities
Research Institute (April 2013) <https://www.bessofhardwick.org/background.
jsp?id=174>.
WILLIAMS, G., Women’s Epistolary Utterance: A Study of the Letters of Joan and
Maria Thynne, 1575-1611 (Amsterdam and Philadelphia, PA, 2013).
WILSON, A., “Speech, writing and discourse type”, in: English Language: Description,
Variation and Context, ed. J. CULPEPER, F. KATAMBA, P. KERSWILL, R. WODAK,
and T. MCENERY (Basingstoke, 2009), pp. 425-438.
WOLF, L., L. POTIKHA, N. DERSHOWITZ, R. SHWEKA, and Y. CHOUEKA, “Computerized
paleography: Tools for historical manuscripts”, in: 18th IEEE International Confer-
ence on Image Processing (2011), pp. 3545-3548, <http://ieeexplore.ieee.org/
stamp/stamp.jsp?arnumber=6116481>.
Women Letter-Writers in Tudor England, ed. J. DAYBELL (Oxford, 2006).
WOUDHUYSEN, H.R., “The Queen’s own hand: A preliminary account”, in: Elizabeth
I and the Culture of Writing, ed. P. BEAL and G. IOPPOLO (London, 2007), pp. 1-
27.
Writing … in Stereo: Bilingualism in the Text, ed. R. HODGSON and R. SARKONAK (=
Visible Language 27.1-2 (1993)).
ZUMTHOR, P., “The impossible closure of the oral text”, trans. J. MCGARRY, Yale
French Studies 67 (1984), pp. 25-42.
 Index

Abbreviations, 142, 152, 156-157, 159, 233- Cocks, Richard, 20, 178-179, 181-184, 189-
234, 243 193, 195, 198-200
Abcedarium anglico latinum, 209-210 Code-switching, 165-200; definition of -,
Aide-memoire, 45, 51-52, 56, 58 166-167; marking of -, 11, 20, 173, 188,
Allograph, 170 (note), 181, 199 (note), 233, 190
238-240, 243-244, 246 Colour, functions of, 7, 8, 38-39, 99; blue, 37,
Alphabet, 170-171, 233, 236; Latin -, 167; 39; gold, 39; red, 5, 8, 25, 26, 35, 37, 39;
roman -, 126; runic -, 125 white, 8
Alvearie, 210-213 Compositor, 46-47, 69-70, 159-160, 204; ÷
Andrew, St., 18, 25, 35-39 Typesetter
Anglo-Saxon England, 126, 132, 135-136, Computational analysis, 101, 245; ÷ Image
139 analysis
Anglo-Saxon runes, 125, 127-129, 132-134, Corpora, 96, 178, 200; construction of -, 16,
136-139 246, 180; corpus linguistics, 17, 95; corpus
Annotation, 17, 96, 101 studies, 166, 221
Aristotle, 14-15 Cryptography, 133, 138
Artefacts, significance of, 10, 101, 141 Cultural analytics, 100-101
Aubert, David, 145-146 De Saussure, Ferdinand, 48
Auctoritas, 26 De Vigo, Joannes, 114
Baret, John, 210-213, 216 Definitions, 203, 207, 209, 213-216
Bateman, John, 9, 204 Dictionaries, 20, 27, 201-218; bilingual -,
Bess of Hardwick ÷ Talbot, Elizabeth 203, 205, 213, 214, 218; monolingual -,
Biscriptalism, 168-169 (note) 203, 204, 214, 215, 216, 218; polylingual
Black letter ÷ Typeface -, 203, 204, 205
Blank space, 7, 10; ÷ Negative space Digital humanities, 98
Brevigraphs, 234 Digital media, 27, 141
Caistor-by-Norwich astralagus, 127-128 Digitised graphemic identification, 245; ÷
Calepinus (Calepine), Ambrosius, 205 Computational analysis
Capital (letter), 75, 79, 91, 177 (note), 200, Digraphia, 168-169, 199, 205
203 Discourse linguistics, ix, 2, 18, 41, 44
Capitalisation, 13, 173, 235-236, 243 Discourse organisation ÷ Textual organisa-
Cawdrey, Robert, 214-215 tion
Caxton, William, 19-20, 141-164 Distant reading, 99
278 Index

Distinctio (punctuation), 54 History of reading, 46, 47

The Dream of the Rood, 130-133 Holograph ÷ Letter
Editions, 15; digital -, 15, 17, 96, 101, 114; Howlet, Richard, 209
printed -, 69 Illustrations, 7, 67, 71
Elyot, Thomas, 209 Image analysis, 106, 109; methods of -, 97,
Emphasis, visual, 11-14, 53, 55, 56, 77, 85, 103, 108; potential of -, 19, 97, 116, 121;
92, 196; ÷ Flagging, Foregrounding, High- ÷ Computational analysis
lighting ImageJ, 19, 97, 103-104
Estienne, Robert, 207-208, 217, 218 Incunabula, 141
Ethos, 15 Information design, 7; ÷ Ordinatio, Textual
Exeter Book riddles, 132; ÷ Riddles organisation
Extensive reading, 19, 45-46, 54-64 Initials, 35, 37, 39, 56-57, 61
Facsimile images, 19, 98, 106, 118; problems Intensive reading, 45, 51-52, 56-57, 59, 62-63
of -, 96, 102 Italic (script, typeface) ÷ Script, Typeface
Falstone inscription, 126, 130, 135 Johnson, Samuel, 217
Flagging, typographical, 174, 179, 190, 198- Kersey, John, 216
200; visual -, 173; ÷ Highlighting Latin, 38, 75, 133, 170
Foregrounding, 3, 12, 13, 15; ÷ Highlighting Latin alphabet ÷ Alphabet
Format of book, 7, 11, 109, 112, 210; folio, Layout, 5-7, 9-11, 15, 24, 52, 55, 70-71, 204,
98, 109, 147, 210; octavo, 109; quarto, 109, 223
210 Letter, 20, 178, 181, 219-220, 222-223, 229,
Fount, 143, 146, 160, 162, 199, 200 203; ÷ 245-246; holograph -, 221-224, 225, 245,
Typeface 246; scribal -, 221-224, 225-229, 246
Franks (Auzon) casket, 129, 132-134 Lexicographer, 20, 202, 204, 216, 218
Freebairn, Robert, 42-43, 49-50, 59-64 Lindesay, Robert, of Pitscottie, 42-43, 49
Futhorc, 125, 137; ÷ Runes. Literacy, 3, 18, 45, 60, 71, 128-139, 171, 202;
GeM (Genre and Multimodality) Model, 9 Latin -, 126, 133-134, 135; practices, 18,
Genette, Gérard, 69 42, 46-47, 49, 57, 59, 61-64; runic -, 125-
Glossary, 203-204, 214 126, 134, 135, 136, 138
Grapheme, 19, 231, 233, 236, 238-240, 243- Litterae notabiliores, 51, 54, 58, 59
246 Logos, 15
Gutenberg, Project, 100 Macron, 142, 147, 150, 156-158, 233, 243
Hand, 144, 169, 222-225, 236, 245; ÷ Hand- Majuscule, 174, 235; ÷ capital (letter)
writing, Script Malton pin, 137
Handwriting, 222, 236-237 Mansion, Colard, 145-146
Heading, 8, 35 Material realisation, of page, 45, 98; of book,
Headline, 19, 69, 73, 76, 81-82, 87-88, 90-93 document, or text, 4, 69, 98
Henisch, Georg, 213-214 Media, digital, 27, 141
Highlighting, 3, 12, 75-78, 80-93, 165, 192, Medical writing, 68, 73-74, 93
195, 196; methods of -, 5, 70-71,76, 78, Metadata, 102; annotating of -, 101
92, 173; reasons for -, 14, 68, 75, 79, 198; The Mirror of the Blessed Life of Jesus Christ,
textual -, 91, 166, 193; verbal -, 71-72, 91; 25-26, 39
visual -, 12, 13, 19, 68, 75, 91, 167, 171, Mise-en-page ÷ Layout
173-176, 179-180, 199, 200; ÷ Flagging Most Excellent Workes of Chirurgerye, 114,
Historical pragmatics, 18, 24, 41, 44, 75 117
Historie and Cronicles of Scotland, 41-65 Mouvance, 41, 44
Index 279

Moxon, Joseph, 12, 70, 204 Roman alphabet ÷ Alphabet

Multiliteralism, 126 Rubrication, 25, 26, 40
Multimodality, 9; multimodal analysis, 7-8 Runes, 19, 126, 130, 133, 135-139
Negative space, 106, 108, 113; ÷ Blank space Runic inscription, 19, 125-127, 130, 132-135,
New philology, 44 138-139
News pamphlets, 68, 71, 78, 91, 93 Runic alphabet ÷ Alphabet
Northern Homily Cycle, 28-35 Ruthwell Cross, 130-134
Optical character recognition (OCR), 115, 116 Scottish Reformation, 42
Oral storytelling, 34 Scribal profiling, 219, 221, 225-231
Ordinatio, 5, 25, 26, 38, 40 Scribe, 20, 46-47, 58, 59, 142, 147, 156, 219-
Orthography, 135, 144, 149, 150; ÷ Spelling 222, 228, 240, 245; ÷ Letter
Otiose mark, 142, 148, 163 Script, 167, 170, 222; cursive -, 167, 169;
Otiose sort, 162, 164 italic-, 165, 167, 169-172, 222; national -,
Otiose stroke, 142, 164 166, 169-170, 172, 181; roman -, 134, 135;
Palaeography, 2, 6, 142, 167, 219, 220, 230- secretary -, 167, 169, 171, 176, 222; ÷
231, 236, 239-240, 243-246 Hand, Handwriting
Paraph marks, 18, 35, 37 Script-switching, 20, 165-200
Paratext, 18, 35, 48-49, 61, 98, 101; functions Setting ÷ Typesetting
of -, 36, 55, 62 A shorte dictionarie for yonge begynners, 210
Parkes, Malcolm, 5, 6, 222, 230, 237 Sort (typesetting), 20, 142, 144, 146-147,
Pathos, 15 149-164
Penmanship, 231, 236, 237-238 South English Legendary, 35-39
Personal correspondence, 20, 219, 245; ÷ Spelling, 100, 142, 146, 151; variation, 3, 13,
Letter 149; ÷ Orthography
Persuasion, 14-15 Spelman, Henry, 214-215, 218
Plague, 68, 72, 74-75, 77, 81, 82-93 Standardisation, 27, 40; of English spelling,
Pragmaphilology, 95 100, 163, 241; of printing, 26, 164
Pragmatic polysemy, 26 Style-switch, 194-196
Pragmatics, 26; historical -, 18, 24, 41, 44; Surgical books, 19, 97, 109
visual -, 11 Table Alphabetical, 214-215
Printing, 43, 70, 143, 146-147; printers, 63, Talbot, Elizabeth, 20, 219-220, 222, 224-229,
67-68, 93, 110-111, 142-143, 160, 172, 246
202; printing houses, 69-70, 204, 213; Textual element, 2, 20, 68, 92, 93, 203-204;
printing industry, 203, 218; printing prac- non-textual element, 69, 117
tices, 75, 213 Textual label, 68, 71-73, 74, 76, 80-82, 85-92,
Puncti, 54, 59 93
Punctuation, 51, 54, 58-59; in scribal identifi- Textual organisation 23-26, 35; devices of -,
cation, 234-235; practices, 49, 162 89, 104, 176, 195
Reading practices, 11, 18, 19, 44-49, 52-59, Thaïs, St., 18, 25, 28-35
61-65; ÷ Extensive reading, Intensive read- Thames scramasax (seax of Beagnoth), 137
ing Title-page, 19, 67-93, 103, 179, 195, 210-211,
Recuyell of the Histories of Troye, 148-150, 216; early modern -, 104, 118-119, 121
152-153, 155, 157-158 Topical label, 72, 76, 81, 82, 87-93
Rhetoric, 14; bibliographic -, 4; visual -, 14 Translations, 7, 203
Riddles, 129, 132, 133 Typeface, 8, 71, 78, 80, 91, 113, 147, 166,
Roman (script, typeface) ÷ Script, Typeface 171-172, 197, 202; black letter, 11, 76-78,
280 Index

165, 171-172, 176, 203, 218; italic, 77, 87, Typography, 5, 7-8, 11-14, 20, 75, 99, 165,
176, 200, 202; roman, 11, 77, 165, 171- 203-204, 209; changes in, 78, 203, 209,
172, 176, 202; ÷ Fount 217
Typeface-switching, 165-200 Wheatley Hill ring, 127-128
Typesetter, 120, 142, 146, 149, 152, 158, 163, Withals, John, 209-210
187, 197, 198; ÷ Compositor Worde, Wynkyn de, 161, 207
Typesetting, 69, 154; ÷ Sort