The Eusebian Canon Tables Ordering Textual Knowledge in Late Antiquity Oxford Early Christian Studies Matthew R Crawford Full Chapter
The Eusebian Canon Tables Ordering Textual Knowledge in Late Antiquity Oxford Early Christian Studies Matthew R Crawford Full Chapter
The Eusebian Canon Tables Ordering Textual Knowledge in Late Antiquity Oxford Early Christian Studies Matthew R Crawford Full Chapter
General Editors
GILLIAN CLARK ANDREW LOUTH
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
The Eusebian
Canon Tables
Ordering Textual Knowledge in Late Antiquity
M A T T HEW R . C R AW F O R D
1
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
3
Great Clarendon Street, Oxford, OX2 6DP,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
© Matthew R. Crawford 2019
The moral rights of the author have been asserted
First Edition published in 2019
Impression: 1
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, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2018957123
ISBN 978–0–19–880260–0
Printed and bound by
CPI Group (UK) Ltd, Croydon, CR0 4YY
Links to third party websites are provided by Oxford in good faith and
for information only. Oxford disclaims any responsibility for the materials
contained in any third party website referenced in this work.
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
Acknowledgements
Eusebius was able to create his Canon Tables only because of the prior
intellectual labour of Ammonius of Alexandria, and the ingenuity apparent
in his creation owes a great deal to the stimulating environment of the
Caesarean library. The present project is similarly indebted to a number of
scholars who have offered insightful suggestions and criticisms along the way,
and was similarly nurtured by two intellectually fertile institutional settings.
I began writing this book while employed as a postdoctoral researcher in the
Department of Theology and Religion at Durham University and completed it
after making a transition to the Institute for Religion and Critical Inquiry at
Australian Catholic University. The initial research was supported by a grant
from the UK Arts and Humanities Research Council for the project ‘The
Fourfold Gospel and its Rivals’, whose chief investigator was Francis Watson
of Durham. The book owes more to Francis than to any other single person,
since he was the one who introduced me to the Canon Tables and convinced
me that it was a worthy object of scholarly scrutiny. His influence on my
thinking on this matter extends well beyond the citations to his own scholar-
ship in the pages that follow.
The first presentation I gave on the Canon Tables was to the New Testa-
ment research seminar at Durham in early 2013, and shortly thereafter the
Reverend Canon Rosalind Brown kindly invited me to give a public lecture on
the topic to a lunchtime crowd at Durham Cathedral, in connection with the
Lindisfarne Gospels exhibit in summer 2013. One of the surprising outcomes
of both those events (surprising to me at least) was the degree of interest
shown in the topic, an experience that has been replicated many times over in
the subsequent four and a half years. In fact, I never would have imagined
I would write a book on the Eusebian Canon Tables, but after repeatedly
seeing the enthusiasm the material generated among scholarly audiences
I finally realized that to do so was imperative. Papers that eventually became
chapters were presented at the annual meetings of the North American
Patristics Society, the Society of Biblical Literature, the Australasian Associ-
ation for Byzantine Studies, and the Asia-Pacific Early Christian Studies
Society, and I am grateful for the feedback I received on each of those
occasions. An earlier version of chapter two was published as an article in
New Testament Studies in 2015, and an earlier version of chapter six was
published in 2017 in an edited collection entitled Producing Christian Culture:
Medieval Exegesis and Its Interpretative Genres, edited by Giles Gasper, Francis
Watson, and myself. Alessandro Bausi generously invited me to give a talk on
Canon Tables at a conference on multiple-text manuscripts organized by the
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
vi Acknowledgements
Centre for the Study of Manuscript Cultures in November 2016, which
prodded me to delve into the world of manuscript studies, and he invited
me to return in May 2018 for a stimulating conference entirely devoted to
Canon Tables. In October 2017 I was able to return to Durham and, thanks to
the invitation of Jane Heath, present some of the material in more developed
form to the New Testament research seminar, where I was once again re-
minded that there are few places that are as collegial and intellectually
engaging. In addition to these public events, chapters or portions of chapters
in various stages were read by Lewis Ayres, Jeremiah Coogan, Ben Edsall, Jaś
Elsner, Brian Gronewoller, Will Kynes, Margaret Mitchell, Judith McKenzie,
Dawn LaValle Norman, Adam Ployd, Robert Thomson, and Jonathan Zecher,
all of whom offered comments that have improved the final form of the book.
Other scholars provided assistance in a variety of ways. Stephen Carlson
helped me on more than one occasion with conversations about the Synoptic
Problem and tracking down and citing Greek manuscripts. Michael Papazian
answered multiple queries about Armenian grammar. Andrew Riggsby and
Nathan Sidoli shared pre-publication versions of their work which turned out
to be crucial for certain stages of the argument, and Peter Williams introduced
me to the fascinating Codex Climaci Rescriptus, which I discuss in Appendix 3.
For pressing me to consider the grander implications of the topic and for
introducing me to the scholarship of Mary Carruthers (cited in two of the
following chapters), I am particularly grateful to T. J. Lang, whom I had the
pleasure of having as a colleague during my final year at Durham. The isolated
hallway in the top level of the Dun Cow Cottage where we had our offices
witnessed many conversations about paratexts, memory, and interpretation,
which resulted in a co-authored article and provided the context within which
the shape for the present book emerged. Four research assistants have helped
in a variety of ways at differing stages of the project: Clift Ward, Carolyn
Alsen, Ed Jeremiah, and Jon Simons. I should also thank the Research Office of
ACU for providing the funds to cover costs associated with securing copyright
permissions and printing the images included in the book, as well as funds for
research trips to UCLA in April 2017 and to the British Library in October
2017 during which I was able to examine several Greek, Latin, Syriac, and
Armenian gospelbooks related to this project. This financial support came as
part of a five-year grant project titled ‘Modes of Knowing and the Ordering of
Knowledge in Early Christianity’ (2017–21), a fitting theme since the present
volume draws attention to one of the most innovative attempts at ordering
knowledge in late antiquity.
Finally, I should mention the libraries and individuals who have assisted me
with finding the images that appear in the book and arranging the permissions
to publish them here, including Judith McKenzie (Faculty of Oriental Studies,
Oxford), Simon Elliott (Charles E. Young Research Library, UCLA), Julia
Rodwell (National Gallery of Victoria, Melbourne), Sandra Powlette (British
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
Acknowledgements vii
Library, London), Alexander Devine and Anne McLaughlin (The Parker
Library, Corpus Christi College, Cambridge), Carl Graves (Egypt Exploration
Society, London), Alessandro Moro (Shylock E-Solutions, Venice, who pro-
vided images of materials in the Biblioteca Nazionale Marciana), Gevorg
Ter-Vardanian (Matenadaran, Erevan), Michael Gervers, Mikheil Tsereteli
(Georgian National Museum, Tbilisi), Simone Verde (Complesso Monumen-
tale della Pilotta, Parma), Florent Palluault (Médiathèque François-Mitterrand,
Poitiers), Kerstin Herzog (Universitätsbibliothek, Augsburg), Mary Haegert
(Houghton Library, Harvard), Sharon Sutton (The Library of Trinity College
Dublin), Benedicta Erny (Universitätsbibliothek, Basel), Stefano Grigolato and
Maddalena Piotti (Biblioteca Queriniana, Brescia), and Anna Rita Fantoni and
Eugenia Antonucci (Biblioteca Medicea Laurenziana).
This book is dedicated to two people who will probably never read it, but
without whose friendship and kindness it would never have been completed:
Hamish and Andy.
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
Contents
Introduction 1
PART 1: ORIGINS
1. Eusebius’ Canon Tables as a Paratext for Ordering Textual
Knowledge 21
Canon Tables as a Paratext 21
Defining Paratextuality 21
The Development of Paratexts in Greco-Roman Book Culture
up to the Fourth Century CE 28
Canon Tables and Tabular Presentation 33
Tables as Information Visualization 33
Tabular Presentation in Babylon and Rome 40
Tabular Presentation in Ptolemy and Eusebius 43
Conclusion 54
2. The Origins of Scholarship on the Fourfold Gospel:
From Alexandria to Caesarea 56
The Diatessaron-Gospel of Ammonius of Alexandria 57
Who was Ammonius? 57
Eusebius’ Description in the Letter to Carpianus 63
Alexandrian Scholarly Traditions and Ammonius’ Work 70
Eusebius’ Continuation of Ammonius’ Scholarly Project 74
Tabular Methods in Eusebius’ Broader Corpus 74
What did Eusebius Take from Ammonius? 80
Eusebius’ Modus Operandi 84
From Formlessness to Polymorphic Diversity 91
3. Reading the Gospels with the Eusebian Canon Tables 96
The Canonizing Effect of the Canon Tables 96
Hypertextual Reading within a Bounded Corpus 100
The Indeterminate Quality of Eusebius’ Parallels 105
Case Studies in Eusebius’ Parallels 112
Case Study 1: Divine and Human Origins 112
Case Study 2: Jesus’ Passover Journeys to Jerusalem 113
Case Study 3: The Cleansing(s) of the Temple 114
Case Study 4: Jesus’ Body as Bread 116
Case Study 5: Peter’s Confession 117
Case Study 6: The Petrine Commission 119
Conclusion 119
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
x Contents
PA RT 2: RE CE PTIO N S
4. ‘The Diversity of Agreement among the Four Evangelists’:
Augustine’s Usage of the Canon Tables 125
Augustine’s Access to Jerome’s Vulgate 128
Methodological Clarifications 130
Augustine’s Dependence upon the Canon Tables in De Consensu
Evangelistarum 133
Memoria Rerum, Memoria Verborum, and the Truth of the Gospel 145
Conclusion 154
5. Canon Tables 2.0: The Peshitta Version of the Eusebian Apparatus 156
The Syriac Version of Eusebius’ Letter to Carpianus 159
Marginal Concordance Tables in Syriac Manuscripts 165
The Peshitta Revision of Eusebius’ Sectioning and Parallels 171
Conclusion 191
6. Scholarly Practices: The Eusebian Canon Tables in the
Hiberno-Latin Tradition 195
Ailerán of Clonard, Canon Evangeliorum 200
Pauca de Libris Catholicorum Scriptorum in Evangelia Excerpta 213
The Irish Reference Bible 216
Sedulius Scottus 221
Conclusion 226
7. Seeing the Salvation of God: Images as Paratext in Armenian
Commentaries on the Eusebian Canon Tables 228
Artistic Adornment of Canon Tables in Late Antique and
Medieval Gospelbooks 231
Two Armenian Interpretations of Canon Table Decoration 248
Introduction to Step‘anos of Siwnik‘ and Nerses Šnorhali 248
The Function of Architecture and Ornamentation according
to Step‘anos and Nerses 252
Interpretations of Specific Motifs in the Decorative Scheme 259
Image and Symbol: Canon Table Decoration as a System of Loci
and Imagines 268
Image and Text: Armenian Canon Tables as Bildeinsätze 278
Conclusion 283
Conclusion 285
Contents xi
Bibliography 317
Index of Ancient Sources 351
Index of Manuscripts and Papyri 358
Index of Modern Authors 360
Subject Index 363
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
List of Illustrations
List of Illustrations xv
23. Lk §94 (Lk §ϘΔ = Lk 9:18–20) in The Gospel Book of Theophanes
(second quarter of the twelfth c.) 118
National Gallery of Victoria, Melbourne, Felton Bequest, 1960 (710-5),
fol. 153r.
24. The Johannine version of Peter’s commission in
The Lindisfarne Gospels (eighth c.) 120
London, British Library, Cotton MS Nero D IV, fol. 258v
(© The British Library Board).
25. First half of the Letter to Carpianus in the Rabbula Gospels
(sixth c.) 160
Florence, Biblioteca Medicea Laurenziana, MS. Plut. 1.56, fol. 2v.
26. Mt §343–7 (= Mt 26:35ff.) in the Rabbula Gospels (sixth c.) 166
Florence, Biblioteca Medicea Laurenziana, MS. Plut. 1.56, fol. 84r.
27. Mt §354–5 in Codex Sangallensis 1395 (first half of fifth c.) 169
St Gall, Stiftsbibliothek 1395, p. 132.
28. Jn §55–8 (Greek numerals ΝΕ, ΝϚ, ΝΖ, ΝΗ) in Codex
Basilensis A.N.III.12 (ninth c.) 170
Basel, Universitätsbibliothek, A.N.III.12, fol. 269r.
29. The poem Canon Evangeliorum by Ailerán of Clonard in
the Augsburg Gospels (eighth c.) 201
Augsburg, Universitätsbibliothek, Cod. I.2.4°.2, fol. 1v.
30. Canon II in The Book of Kells (c.800) 202
Dublin, Trinity College Library, MS 58, fol. 2v.
(© The Board of Trinity College Dublin).
31. A word square facing Ailerán’s poem in the Augsburg
Gospels (eighth c.) 206
Augsburg, Universitätsbibliothek, Cod. I.2.4°.2, fol. 2r.
32. Christ in Majesty in Livre d’Evangiles de l’abbaye Sainte-Croix
(eighth c.) 208
Poitiers, Médiathèque François-Mitterrand, MS 17 (65), fol. 31r.
33. Canons V, VI, VII, and VIII in The Book of Durrow (seventh c.) 232
Dublin, Trinity College Library, MS 57, fol. 9v
(© The Board of Trinity College Dublin).
34. Canons XLk and XJn in Abba Garima III (fifth–seventh c.) 234
Ethiopia, Abba Garima Monastery, AG II, fol. 260v
(© Michael Gervers, 2004).
35. Canon V in the Lindisfarne Gospels (eighth c.) 235
London, British Library, Cotton MS Nero D IV, fol. 15r
(© The British Library Board).
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
Abbreviations
BO Biblica et Orientalia
CCSL Corpus Christianorum Series Latina
CSCO Corpus Scriptorum Christianorum Orientalium
CSEL Corpus Scriptorum Ecclesiasticorum Latinorum
GCS Die griechischen christlichen Schriftsteller
LCL Loeb Classical Library
LSJ Liddell, Scott, Jones, A Greek-English Lexicon (Oxford, 1843; 9th edn 1940)
LXX Septuagint
OECS Oxford Early Christian Studies
OECT Oxford Early Christian Texts
PG Patrologia Cursus Completus: Series Graeca
PL Patrologia Cursus Completus: Series Latina
PTS Patristische Texte und Studien
SC Sources Chrétiennes
ST Studi e testi
STAC Studien und Texte zu Antike und Christentum
TU Texte und Untersuchungen zur Geschichte der altchristlichen Literatur
VCSup Supplements to Vigiliae Christianae
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
Introduction
When modern readers open up nearly any edition of the Bible available for
purchase today, they encounter much more than the bare text of the ancient
documents that comprise the canon of Christian scripture. Any modern
reader of the Bible will instead be confronted with a penumbra of additional
material intended by editors as aids for the reading or interpretation of the
main text. In some cases, such as with study Bibles, this secondary material
perhaps outweighs the primary text, with the inclusion of tables of contents,
prefaces, explanatory notes, alternative translations or textual traditions, sec-
tion headings, cross-references, and maps—not to mention chapters and
verses. Most Bibles contain less than this, but almost all have at a minimum
cross-references to guide the reader to other pertinent passages, or perhaps
indexes to allow a reader to find with ease passages that speak to a given topic.
So common are these reading aids that they hardly seem remarkable, just
another mundane feature of modern life that one can count on always being
present, just as one can assume that the local coffee shop will have Wi-Fi
available for its customers. Yet it was not always so. Just as the technology that
allows a laptop to connect to the wider world through the Internet has a
distinct history of its development and use, so also the technology of reading
aids—for that is what they are, a technology—did not always exist in the ‘taken
for granted’ category that we now perceive them to be. Rather, they too have a
certain background out of which they emerged, and, again, just like the
Internet, the personal computer, or moveable type, they were a technological
advance that created hitherto unforeseen possibilities.
This book is about one specific form of reading technology that arose in
the early fourth century CE and has influenced the reading of the Christian
scriptures ever since. The most prolific Christian author of this period, who
experimented with an astonishingly diverse range of genres, topics, and
technologies, was Eusebius Pamphili, who lived through the great persecution
under Diocletian, then became bishop of Caesarea Maritima, was present at
the Council of Nicaea in 325, and delivered an oration for the thirtieth
anniversary of Constantine’s reign in 335. Famous as the author of the
earliest surviving ecclesiastical history, Eusebius also left to posterity a world
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
Introduction 3
whether or not they had parallels in other gospels, and, if so, how many other
gospels, with the result that some sections are longer than a single chapter in
today’s reckoning, while others are merely half a verse. Finally, Eusebius
collated the numbers for each of the gospels in ten tables, called ‘Canons’,
which he placed at the beginning of the fourfold gospel, following his Letter to
Carpianus.⁴ Each of these Canons represents a different set of passages that
could be classified together on the basis of which gospels they appeared in.⁵ So,
for example, Canon I presented the numbers for those passages that occurred
in all four gospels, with the numbers for each set of parallel passages grouped
together, placed alongside one another in a given row of the table. He
continued on in this fashion as follows for the remaining tables:⁶
Canon II Matthew–Mark–Luke
Canon III Matthew–Luke–John
Canon IV Matthew–Mark–John
Canon V Matthew–Luke
Canon VI Matthew–Mark
Canon VII Matthew–John
Canon VIII Luke–Mark
Canon IX Luke–John
Canon XMt Matthew alone
Canon XMk Mark alone
Canon XLk Luke alone
Canon XJn John alone
The rationale of the system is that by reading horizontally across the rows
within the first nine tables one can find passages that are similar in another
mark the latter. A useful conversion table between the two reference systems can be found at
https://danielbwallace.files.wordpress.com/2014/04/eusebian-canons-conversion-table.pdf [accessed
on 21 Dec 2017].
⁴ Although strictly speaking only these prefatory tables are the ‘Canon Tables’, in keeping
with prior scholarly convention I will on occasion use the phrase to refer to the entire paratextual
system, comprising the ten tables, the marginal notation throughout the gospels, and the Letter
to Carpianus.
⁵ In keeping with the convention of the NA²⁸, I will always give the section number in Arabic
numerals and Canon numbers in Roman numerals. Eusebius made a similar typographic
distinction, writing the section number in black and the Canon number below it in red (both
of course as Greek numerals), a convention that he describes in his Letter to Carpianus. pace
Thomas O’Loughlin, who asserted that the colour distinction and placement of one numeral
above the other in the margin were innovations introduced by Jerome in his Latin translation:
Thomas O’Loughlin, ‘The Eusebian Apparatus in Some Vulgate Gospel Books’, Peritia 13
(1999): p. 4; Thomas O’Loughlin, ‘Harmonizing the Truth: Eusebius and the Problem of the
Four Gospels’, Traditio 65 (2010): p. 14; Thomas O’Loughlin, ‘The Eusebian Apparatus in the
Lindisfarne Gospels: Ailerán’s Kanon Euangeliorum as a Lens for Its Appreciation’, in The
Lindisfarne Gospels: New Perspectives, ed. Richard Gameson (Leiden: Brill, 2017), p. 99.
⁶ The reader will observe that the tenth and final Canon actually comprises four separate
tables. On this aspect of the system, see the discussion in chapter two.
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
Fig. 1. Mk §64 (in Greek ξΔ = Mk 6:35ff.) in The Gospel Book of Theophanes (second
quarter of the twelfth c.).
National Gallery of Victoria, Melbourne, Felton Bequest, 1960 (710-5), fol. 94r
Beneath or alongside the black number is another number in red that tells
you in which of the tables you should look to find this passage. Here the red
digit is α (= 1), so you turn to the first Canon, which reports parallels among
Matthew, Mark, Luke, and John. Immediately, then, you realize that this is a
story that occurs in all four gospels. Now, within Canon I, if you scan down the
column that lists numbers for Mark, you should eventually come to the
number 64 (see fig. 2). For the sake of illustration, here is the row that includes
this number, with the preceding and subsequent rows included for comparison:
Mt Mk Lk Jn
142 51 21 35
147 64 93 49
166 82 94 17
Once you have located Mark section 64, you can read horizontally, to the
left and to the right, to find that that the feeding of the 5,000 is §147 in
Matthew, §93 in Luke, and §49 in John. Then, if you were to turn to those
numbered passages in the other gospels, you would find the parallel accounts
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
Introduction 5
Fig. 2. Canon I in The Gospel Book of Theophanes. Each column contains numbers
for passages within a given gospel, as indicated by the two-letter abbreviations at
the top of each column. The second column is for Mark and the first number listed on
the right side of the page is ξδ (= Mk §64). (Note that the scribe has here written a
lowercase delta, whereas in the page of Mark shown previously he has used an
uppercase delta.)
National Gallery of Victoria, Melbourne, Felton Bequest, 1960 (710–5), fol. 3r
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
Fig. 3. Mt §147 (in Greek ρμζ = Mt 14:14ff.) in The Gospel Book of Theophanes.
National Gallery of Victoria, Melbourne, Felton Bequest, 1960 (710-5), fol. 40r
Eusebius’ system, therefore, had the same effect as the cross-references that
often occur today in the margins of printed Bibles, though he had to devise a
method that worked in the absence of chapter and verse divisions, and, indeed,
⁷ The preceding paragraph is largely a paraphrase of the latter half of the Letter to Carpianus.
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
Introduction 7
in the absence of any comparable system of numerical cross-references for a
corpus of literature that he might have used as a model. Though obvious
enough to modern readers, this was an innovation of remarkable proportions
for its day. Accordingly, the aim of the present monograph is to highlight the
ingenuity that went into its creation and the way in which readers interacted
with it in the eight centuries thereafter. Its influence may be seen simply by
taking stock of the surviving gospel manuscripts in the various languages of late
antiquity. Three of the four early Greek uncials from the fourth and fifth
centuries contain traces of the Eusebian apparatus in the margins, though
none of them preserves the prefatory Canons: Codex Sinaiticus, Codex Alexan-
drinus, and Codex Ephraemi Rescriptus.⁸ The oldest of these, Codex Sinaiticus,
was probably written only a few decades at most after Eusebius’ death (see fig. 6).
Moving from the world of Greek into Latin, we find the Canons in the
majority of Vulgate gospelbooks, including most of the well-known ones, such
as Codex Fuldensis (6th c.), the St. Augustine Gospels (6th c.), and the
Lindisfarne Gospels (early 8th c.),⁹ not surprisingly, since Jerome incorporated
a Latin version of the apparatus in the new Vulgate edition he created for Pope
Damasus in the early 380s (see fig. 7).
At some point in the century after Jerome finished his Vulgate translation,
an unknown scribe writing in Syriac produced a revised version of Eusebius’
apparatus to accompany his new translation of the gospels, known today as
the Peshitta, and as a result the majority of Peshitta manuscripts likewise
contain the marginal system, including well-known examples such as the
Rabbula Gospels, copied in the year 586 (see figs 19, 25, 26, 42),¹⁰ and a
sixth-century copy of the gospels in Paris (see fig. 8).
Fig. 6. The beginning of the gospel of John in Codex Sinaiticus (mid fourth c.). The
text of the Gospel of John begins at the top of the left column, and, in the margin next
to the text, one can see an Α, indicating Jn §1 (= Jn 1:1ff.) and beneath it a Γ, showing
that this is a Canon III passage. This specific parallel, which combines the Johannine
prologue with the synoptic genealogies, is discussed in chapter three.
London, British Library, Add MS 43725, fol. 247r (© The British Library Board)
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
Introduction 9
The earliest copy of the gospels in Gothic, the manuscript known as Codex
Argenteus, which was probably made in Ostrogothic Italy in the sixth century
for King Theodoric the Great, also contains the Eusebian system, though the
ten prefatory tables are now lost (see fig. 9).¹¹
¹¹ On Codex Argenteus, see further Carla Falluomini, The Gothic Version of the Gospels and Pauline
Epistles: Cultural Background, Transmission and Character (Berlin: De Gruyter, 2015), pp. 32–4, and
on Canon Tables in Gothic, see Carl Nordenfalk, Die spätantiken Kanontafeln, 2 vols (Göteborg:
O. Isacsons boktryckeri a.-b., 1938), pp. 261–9; Falluomini, The Gothic Version, pp. 53–4.
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
Fig. 8. Canon IV in Peshitta Gospels (sixth c.), showing parallels in Matthew, Mark,
and John (moving from right to left).
Paris, Bibliothèque nationale de France, syr. 33, fol. 5v
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
Introduction 11
Fig. 9. Lk §215–18 (= Lk 18:14ff.) in Codex Argenteus (sixth c.). The parchment page
has been dyed purple and the text has been written in silver, with gold being used for
the incipit of each Eusebian section. The architectural frames in the bottom margin list
the Lukan passages on the page and the parallels they have in the other gospels, in the
sequence Luke, Mark, Matthew, John. For a discussion of this feature, see the section
‘Marginal Concordance Tables in Syriac Manuscripts’ in chapter five.
Uppsala, Uppsala University Library, MS DG 1, fol. 170r
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
Introduction 13
Similarly, the earliest surviving Georgian gospelbook, the Adishi Gospels,
dated to 897, contains the Eusebian apparatus (see fig. 40 in chapter 7), as
does a tenth-century copy of the gospels at Harvard (see fig. 11).
¹² Cf. Judith McKenzie and Francis Watson, The Garima Gospels: Early Illuminated Gospel
Books From Ethiopia (Oxford: Manar al-Athar, University of Oxford, 2016).
¹³ Cf. Jost Gippert et al., eds, The Caucasian Albanian Palimpsests of Mt. Sinai, 2 vols,
Monumenta Palaeographica Medii Aevi. Series Ibero-Caucasica (Turnhout: Brepols, 2008).
The remains of the two palimpsests discussed in this volume, one a lectionary and one a copy
of the Gospel of John, are the only surviving manuscripts containing text written in Caucasian
Albanian. I am grateful to Philip Forness for drawing my attention to them.
¹⁴ Some readers will no doubt notice that Coptic is missing from this list. The earliest
surviving evidence for the transmission of the Eusebian apparatus in Coptic comes from the
twelfth and thirteenth centuries, and it is unclear when this Coptic translation was made. See
G. Horner, The Coptic Version of the New Testament in the Northern Dialect (Oxford: Clarendon
Press, 1898), pp. xli–xliv, lviii–lxi, lxxii–lxxiv; Harold H. Oliver, ‘The Epistle of Eusebius to
Carpianus: Textual Tradition and Translation’, Novum Testamentum 3 (1959): p. 142; Carl
Nordenfalk, ‘Canon Tables on Papyrus’, Dumbarton Oaks Papers 36 (1982): p. 30. The Coptic
version of the Letter to Carpianus, based on one of these later manuscripts (Vat.Copt. 9), was
published in Ad. Hebbelynck, ‘Les κεϕάλαια et les τίτλοι des évangiles’, Le Muséon 41 (1928):
pp. 114–18. Fragments from a Sahidic gospelbook containing traces of the Eusebian apparatus
were mentioned by J. B. Lightfoot in Frederick Henry Ambrose Scrivener, A Plain Introduction
to the Criticism of the New Testament for the Use of Biblical Students, 3rd edn (Cambridge:
Deighton, Bell and Co., 1883), pp. 397–9, who described them as ‘ancient’, but did not give any
more precise date. On the Old Church Slavonic version of the Canon Tables found in Codex
Zographensis, see Leszek Moszynski, ‘Kanony Euzebiusza w głagolskim rękopisie kodeksu
Zografskiego’, Slovo 25–6 (1976): pp. 77–119. I have been unable to find any information
about Canon Tables in Arabic manuscripts. e.g., no mention is made of their existence in
Hikmat Kashouh, The Arabic Versions of the Gospels: The Manuscripts and Their Families
(Berlin: De Gruyter, 2012).
¹⁵ Cf. Martin Wallraff, Kodex und Kanon: Das Buch im frühen Christentum (Berlin: De
Gruyter, 2013), p. 34, who pointed out that, after the Bible itself, the apparatus is ‘den
bestüberlieferten Text der Antike’.
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
Introduction 15
¹⁶ For a discussion of its inclusion in the seventh edition, see E. Nestle, ‘Die Eusebianische
Evangeliensynopse’, Neue kirchliche Zeitschrift 19 (1908): pp. 40–51, 93–114, 219–32. The
version of Canon Tables printed in subsequent editions, right up to NA²⁸, is unchanged from
that of Nestle’s seventh edition, though Prof. Martin Wallraff of the University of Munich is
currently preparing a new critical edition to be published by Mohr Siebeck.
¹⁷ Nordenfalk, Die spätantiken Kanontafeln.
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
Introduction 17
I am calling Canon Tables 2.0—in other words, the new and improved version
of the original, in which the paratext is more finely tuned to accord with the
text to which it relates. In chapter 6 I highlight little-studied literature pro-
duced by Irish or Irish-influenced scholars between the seventh and the ninth
centuries, in which Canon Tables figure prominently as a symbolic but also
utilitarian means of engaging with the text of the gospels and analysing the
relationship between parallel passages. Finally, carrying forward the theme of
the symbolism of the Canon Tables, chapter 7 focuses on other little-studied
texts, in this instance, Armenian commentaries from the eighth century and
later, which explore how the rich decorative motifs that grew up around the
Canon Tables (e.g., architectural frames, birds, plants) can be used as the basis
for a programme of contemplative meditation understood as a preparatory
(and so paratextual) task required before the reading of the fourfold gospel.
This is by no means a comprehensive survey of the reception history of the
Eusebian apparatus from late antiquity into the Middle Ages. Given how little
scholarly attention has been devoted to this topic, too many questions remain
unanswered and too many sources unstudied (and even unpublished) to
attempt such a project. Instead, the four traditions considered in the latter
portion of this book have been chosen because they represent four different
modes of using the Canon Tables as a paratext and so illustrate the potential
inherent in the Eusebian apparatus for engaging with the fourfold gospel in a
variety of ways, from the philological to the theological to the visual.
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
Part 1
Origins
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
Defining Paratextuality
The term ‘paratext’ was coined three decades ago by the literary theorist
Gérard Genette and it came eventually to refer to one of the five types of
textual transcendence he outlined in his poetics.¹ By ‘textual transcendence’,
¹ Note that Genette’s usage of the term underwent development. As he points out, what
he referred to as ‘paratextuality’ in the first volume of his trilogy, he would later term
21
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
‘hypertextuality’ (Gérard Genette, The Architext: An Introduction, trans. Jane E. Levin (Berkeley,
CA: University of California Press, 1992), p. 82 (originally published as Gérard Genette,
Introduction à l’architexte (Paris: Editions du Seuil, 1979)); Gérard Genette, Palimpsests: Litera-
ture in the Second Degree, trans. Channa Newman and Claude Doubinsky (Lincoln, NE:
University of Nebraska Press, 1997), 1 (originally published as Gérard Genette, Palimpsestes:
la littérature au second degré (Paris: Editions du Seuil, 1982))).
² Genette, The Architext, p. 81.
³ Genette, Palimpsests, pp. 1–7. See also the brief summary of the five categories in Richard
Macksey’s foreword to Gérard Genette, Paratexts: Thresholds of Interpretation, trans. Jane
E. Lewin, Literature, Culture, Theory 20 (Cambridge: Cambridge University Press, 1997),
pp. xviii–xix (originally published as Gérard Genette, Seuils (Paris: Editions du Seuil, 1987)).
⁴ Genette, The Architext. ⁵ Genette, Palimpsests. ⁶ Genette, Paratexts.
⁷ Note that both Genette and his later followers point out that the prefix para- denotes not
simply proximity but also liminality. Cf. Genette, Paratexts, p. 1; Laura Jansen, ‘Introduction:
Approaches to Roman Paratextuality’, in The Roman Paratext: Frame, Texts, Readers, ed. Laura
Jansen (Cambridge: Cambridge University Press, 2014), p. 5.
OUP CORRECTED PROOF – FINAL, 10/4/2019, SPi
¹³ See, e.g., Jean-Claude Fredouille et al., eds, Titres et articulations du texte dans les oeuvres
antiques (Paris: Institut d’études Augustiniennes, 1997); Bianca-Jeanette Schröder, Titel und
Text: Zur Entwicklung lateinischer Gedichtüberschriften. Mit Untersuchungen zu lateinischen
Buchtiteln, Inhaltsverzeichnissen und anderen Gliederungsmitteln, Untersuchungen zur antiken
Literatur und Geschichte 54 (Berlin: Walter de Gruyter, 1999); Menico Caroli, Il titolo iniziale nel
rotolo librario greco-egizio: con un catalogo delle testimonianze iconografiche greche e di area
vesuviana (Bari: Levante, 2007); Francesca Schironi, To Mega Biblion: Book-Ends, End-Titles,
and Coronides in Papyri with Hexametric Poetry, American Studies in Papyrology 48 (Durham,
NC: American Society of Papyrologists, 2010); Gianluca del Mastro, Titoli e annotazioni
bibliologiche nei papiri greci di Ercolano (Naples: Centro internazionale per lo studio dei papiri
ercolanensi, 2014).
¹⁴ Laura Jansen, ed., The Roman Paratext: Frame, Texts, Readers (Cambridge: Cambridge
University Press, 2014). See also Andrew M. Riggsby, ‘Guides to the Wor(l)d’, in Ordering
Knowledge in the Roman Empire, ed. Jason König and Tim Whitmarsh (Cambridge: Cambridge
University Press, 2007), pp. 88–107; C. T. Mallan, ‘The Book Indices in the Manuscripts of
Cassius Dio’, The Classical Quarterly 66 (2017), pp. 705–23; Aaron Pelttari, ‘Speaking From the
Margins: Paratexts in Greek and Latin Poetry’, in Walking the Wire: Latin and Greek Poetry in
Dialogue, ed. Tine Scheijnen and Berenice Verhelst (Cambridge: Cambridge University Press,
forthcoming).
¹⁵ Jansen, ‘Introduction’, p. 2.
¹⁶ So Thomas O’Loughlin in ‘De Bruyne’s Sommaires on its Centenary: Has its Value for
Biblical Scholars Increased?’, in Donatien De Bruyne, Pierre-Maurice Bogaert, and Thomas
O’Loughlin, Summaries, Divisions and Rubrics of the Latin Bible, Studia Traditionis Theologiae
18 (Turnhout: Brepols, 2015), pp. xix–xxvi. According to O’Loughlin, the situation was different
for Roman Catholic scholars committed to the Vulgate, for whom ‘the sacred text, for all its
problems, did not come alone but was embedded within a web of other material: various ways
of gathering books together, lists of chapter headings, a variety of division systems, cross-
referencing systems, along with aids to readers which, starting with the work of Eusebius of
Caesarea, seemed to have been added to by every generation until the time of printing’ (p. xx).
This focus on such later material was thought to be appropriate because ‘the tradition was part of
the work of the Spirit speaking in the Church and it was to be respected as such . . . By culture,
training, and temperament the Vulgate editors were inclined to value everything they found in a
codex: it was, in its totality, part of the tradition’ (p. xxi). This impulse, however, is not restricted
to modern scholarly treatments of biblical texts. Shane Butler has argued that ‘the most
Another random document with
no related content on Scribd:
milloin minnekin, niin, että ne loivat varjoja, hämyvaloja ja
kummallisia valovaikutuksia. Keskelle näitä pani hän tavallisesti
ristiinnaulitun-kuvan. Ja hän maalasi milloin täplillä ja pirskeillä, joita
hän näytti huolimattomasti pergamentille pärskyttelevän, milloin
levein pensselivedoin. Arvi arveli toisinaan ettei tuo ollut juuri muuta
kuin tuhrailua; välimmiten häntä ihmetytti tuhrailun tekemä
tenhovaikutus. Hänen käytöksessään mestaria kohtaan oli viime
aikoina kunnioitus alkanut vivahtaa hellyydelle. Hän tiesi, mikä
mestaria, sekä sen ohessa Margitia, rasitti, ja hän tuumaili turhaan
keinoa auttaakseen heitä ahdingosta.
— Niin on. Enkä minä osaa kiivauttani hillitä, vaikka minun olisi
pitänyt se oppia autuaan isävainajani nuhteista, autuaan äitivainajani
esimerkistä ja ystävä vainajani, esimunkki Mathiaksen seurustelusta
sekä lukemalla uutta testamenttia ja Tuomas Kempiläistä.
Kaksikymmentä vuotta ja enemmänkin aikaa on kulunut minun
tietämättäni, että tämä intohimo entisellään rehenteli rinnassani. Sen
uinutti uneen täällä talossa ennen vallitseva rauha; mutta viime
aikoina on se näyttänyt ilkeitä elon oireita. Ja kun en ole laskenut
sitä irroilleen, on se ruvennut työntämään juuriaan syvemmälle ja
tuntuu usein ikäänkuin kirveltävänä polttona sydämen ympärillä.
Näytäpäs nyt minulle, Arvi, piirustustasi. Se on asuinkartano…
— Niin se, jonka aijon rakentaa isäni asumattomalle talontilalle
Kortebohon. Siitä tulee minun taloni, on isä sanonut. Tuohon
kylkirakennukseen laitetaan maalarimaja, jos suinkin ihan tämän
kaltainen, mutta näköala Vetterille päin.
— Ah, mestari, sanoi Fabbe, ette näy tietävän että olen erot
saanut.
*****
— Hän ei olisi sitten niin joutunut Laurin yksinvallan alle kuin nyt.
Sinä olet jaloluontoinen mies. Lauri ei ole. Äiti olisi helpommin tullut
sinun vaikutukseksi alaiseksi kuin hänen, sillä äiti on jalo hänkin,
mutta hän ei tule toimeen ilman miehen johtavaa arvostelua.
— Isä, luuletkos meidän siltä olevan kiirettä, että kello soi? Lauri
sitä salinakkunasta kalistelee, eikä minulle soittokellon äänessä, ole
enempää isännyyttä kuin hänenkään äänessään. Isä, minulla on
jotain muutakin sinulle sanottavaa, kun näin saamme tuttavallisesti
jutella eikä minun tarvitse sinun edessäsi hävetä. Etpä tiedä, että
Lauri on antanut minulle vitsaa…
— Mitä sanot?
— Takaisin mies!
Hän astui nyrkit ojossa Klasea kohden. Mutta tämä juoksi pajaan
ja tuli ulos heiluttaen vasaraa, valmiina lyömään. Silloin kajahti
Gudmund mestarin ääni, kaikuvammin kuin milloinkaan ennen
hänen eläessään:
VIIMEINEN ILTAVAHTIKÄVELY.
*****
— Mitä sanot?
*****
Illallista syötäessä ei Margarccta rouva ollut saapuvilla.
Naispalvelija ilmotti hänen aikaisin menneen levolle. Tämä ei ollut
mitään uutta; hän teki nyttemmin aina niin, ja hän tarvitsi sen. Hän,
niin tottunut onneen, oli jo kauvan elellyt aikojaan, saamatta
sielulleen päivänpaistetta. Kesken talonaskareita, jotka muuten
tuottivat hänelle huvia saman verran kuin työtäkin, istahti hän usein
itkemään ja käsiänsä vääntelemään, kun ei kukaan nähnyt. Mikä
vaivasikaan Gudmundia? Sitä hän kyseli. Hän oli toivonut tälle
elonsyksyä ihanaa kuin kaunis takasuvi. Hän oli toivonut hänelle
uuden kevään elpymistä siinä lämpimässä ja valossa, joka säteili
pojan pontevasta toimesta ja Jumalalle altistuneesta innosta. Eikö
tämä poika ollut kuin Libanonin seetri määrätty pylväänä Siionin
templiä kannattamaan? Eikö kaikki mitä hän tahtoi ollut harkittu
kaukaa ennalta näkevällä älyllä ja eikö se tähdännyt ainoastaan
hyvään päämäärään? Ja kuitenkin — pojan palattua kotiin vanheni
Gudmund silminnähtävästi päivä päivältä, hän ei viihtynyt enää
missään, paitsi maalarimajassaan, karttoi mikäli mahdollista pojan
läheisyyttä, ei osottanut sitä isällistä leppeyttä, joka on omalle pojalle
annettava, ja, vaikka olikin ainoastaan koulua käymätön
käsityöläinen, väitteli hän oppinutta vastaan, tämän selittäessä
Jumalan sanan oikeaa käsittämistä, olipa äskettäin kohdellut häntä
varsin sopimattomastikin. Ja kuitenkin oli Lauri menetellyt niin
maltillisesti, jättäen niin kauvan kuin suinkin toimeenpanematta
milloin minkin tarkoin mietityn parannuspuuhan, eikä suinkaan
väkirynnäköllä ajanut käännytystyötä, vaan koetti pikemmin
viittauksilla ohjata isää perikadon tieltä pelastuksen tielle. Mikä
vaivasi hänen Gudmundiaan? Sitä rouva ei saanut selville.
Margit oli eilisestä asti itsekseen hautonut uutista, joka hänen piti
ilmottaa isälleen, jotta tämä aikanaan ehtisi estää aijotun
toimenpiteen. Mutta sopivaa tilaisuutta ilmottamiselleen hän ei vielä
ollut saanut, ja nykyhetki ei ensinkään siihen sopinut. Lauri oli näet
hänen kuullensa lohduttanut äitiään sillä että tuo pakanallinen
tulitupa, josta he molemmat välisti olivat kuulleet kummallisia ääniä,
ennen pitkää tulisi revittäväksi. Hän ei muka enää aikonut isälle sen
enempää asiasta puhua, sillä Jumalan edessä oli hänellä oikeus,
mutta jos häneltä puuttui maallista oikeutta, niin voisihan isä
haastattaa hänet tuomioistuimen eteen. Lisäksi oli hän päättänyt,
isältä lupaa pyytämättä, siirtää kirjansa maalarinmajaan ja sisustaa
sen itselleen lukukammioksi. Jos isää sitten haluttaisi istua siellä
tuhrimassa, olkoon se hänen vallassaan, kunhan vaan luopuu
paavilaisista kuvailuksista. Lauri oli tuonut Saksasta kymmenittäin
Martti Lutherin kuvia, tehtyjä karkean, törkeän ja kaikittain
onnistumattoman puupiirroksen mukaan, sekä sitä paitse uuden,
melkoisesti parannetun puupiirroksen tuosta merkillisestä, Tiberistä
saadusta paavikalasta, siitäkin useita kappaleita. Nämä aikoi Lauri
antaa isänsä väritettäväksi. Siten tuhrailu ei ainakaan vahingoittaisi
hänen sieluansa.
*****
*****
Lauri ei sanonut mitään, mutta ajatus että hän tuhrisi väreillä tuntui
hänestä liian hävyttömältä.