1
Modifying the Interpretation and Reception of the
Renaissance Reader: Changing Paratexts in
Josephus’ Antiquitates Iudaicae
MA Thesis Book Studies Supervisor: Dr Paul Dijstelberge
MA Book Studies University of Amsterdam
November 30, 2020 Author: Henriët Graafland Word count: 22,441
2
Illustration 1: First page of the main text of Antiquitates Iudaicae on f.1r of the oldest manuscript in the selection, Cod. Bodmer 98. Cologne, Fondation Martin Bodmer, Cod. Bodmer 98, f1r, https://www.e-codices.ch. This image can be viewed via the following weblink: https://www.e-codices.ch/fr/fmb/cb-0098/1r.
Illustration 2: First page of the main text of Antiquitates Iudaicae on page 13 of the last edition in the selection, the 1599 edition. Bayerische Staatsbibliothek, Flavius
Josephus,
https://reader.digitale-sammlungen.de/de/fs1/object/display/bsb10996348_00017.html, edges cut. This image can be viewed via the following weblink:
3 Table of contents
Summary ... 4
Introduction: Paratext and Antiquitates Iudaicae ... 5
1. Paratext, Reading Habits, and Reception ... 13
2. Paratext by Exception in Manuscript Antiquitates Iudaicae ... 23
3. The Transition ... 41
4. Paratext by Rule of Thumb and Their Printers in the Sixteenth Century ... 59
Student Dropouts’ Editing, Printers’ Mentality, and Secular Readers’ Identities: Comparisons and Conclusions for Changing Paratext in Antiquitates Iudaicae ... 81
Limitations and further research ... 90
Acknowledgements ... 91
Bibliography ... 92
Appendix A: Manuscripts not used (alphabetical) ...120
Appendix B: Error margins for countable paratextual elements ...123
Appendix C: Questionnaire ...125
4 Summary
Although paratext is a popular research subject, no researcher has compared a large amount of paratextual elements in medieval manuscript, incunabula and later printed books yet. Such a comparison is conducted in this thesis, in an analysis of 97 codices of Flavius Josephus’ text Antiquitates Iudaicae. Changes in paratext interact with changes in reading habits. Also, the reception of texts is influenced by paratext. Developments in the production, dissemination and consumption of text between 800 and 1600 point out some remarkable conclusions on paratext in Antiquitates Iudaicae that have influenced the idea of modern readers on what the book is.
5 Introduction: Paratext and Antiquitates Iudaicae
Since Gérard Genette coined the term ‘paratext’ in his book Seuils in 1987, many literary scholars have picked up this new field of research that focuses on the form of texts rather than the content.1 Genette defines paratext as all text that serves as a guide for
the reader to navigate and understand the main text of a book. Paratext can, for instance, be a table of contents, a page number or a title of a chapter. These are called ‘peritext’, which means that they surround the text and are part of the entity that contains the text, such as a printed book or manuscript.2 But paratext can also be a
review of a book or an interview with the author, published in a journal or newspaper – texts that are not part of the book itself, but deal with its production, dissemination or consumption. This is what Genette calls ‘epitext’.3 Importantly, Genette’s main
argument is that paratext, in whatever form, directs the interpretation of the reader. Paratexts are, as he argues, ‘thresholds of interpretation’.4
Of course, this is not new. Scholars have studied the format and lay-out of texts for centuries, although they have done so in different ways and with different purposes.5
Book historians and bibliographers have tried to place manuscripts and printed books – their production, dissemination and consumption – in their historical context. Over the years, a lot of research has been done by medievalists on the structure of medieval manuscripts. For medievalists a focus on paratext is important as it is often the only clue leading to the provenance of medieval manuscripts. Scholars of the early modern
1 Gérard Genette, Paratexts: Thresholds of interpretation (Cambridge: Cambridge University Press,
1997), 1-2. This was the English translation of his French book titled Seuils, which had been published in Paris by Editions du Seuil in 1987. He coined the term ‘paratext’ even earlier, in his book Palimpsestes:
La littérature au second degré (Paris: Éditions du Seuil, 1981). 2 Genette, Paratexts, 4-5.
3 Ibid. 3. 4 Ibid. 1-2.
5 For a clear distinction between the so-called ‘analytical’ and ‘French’ schools in book studies, see Frans
A. Janssen, “Gevechten in de leeuwenkuil: Scholenstrijd in de boekwetenschap”, Boekenwereld 27 (2010-2011), 36-39.
6
European printed book have written extensive and ground-breaking studies on the importance of the book in changing cultural paradigms. Elizabeth Eisenstein’s book The printing press as agent of change, Andrew Pettegree and Arthur der Weduwen's study The Bookshop of the World, or Robert Darnton’s research on eighteenth-century books in France are examples of this kind of study.6
Surprisingly, few researchers have systematically analysed the transition from the late Middle Ages to the Renaissance with regards to paratext on a large scale.7 While
the terms ‘incunabulum’ and ‘post-incunabulum’ – indicating the first printed books produced within the first century of their existence – are well-known, the implication of changes in paratext in incunabula and its consequences for reading habits have hardly been debated.8 It has been suggested that changing paratexts in books had a large
impact on reading habits of readers.9 Paratext and thus reading changed enormously in
6 Robert Darnton, “What is the history of books?”, Daedalus 111 (1982), 65-83; Elizabeth L. Eisenstein, The Printing Press as an Agent of Change (Cambridge: Cambridge University Press, 1980); Andrew
Pettegree and Arthur der Weduwen, The Bookshop of the World: Making and Trading Books in the
Dutch Golden Age (New Haven: Yale University Press, 2019). Darnton’s article was one of the leading
publications that introduced a scientific model into Book Studies, named the ‘communications circuit’. Eisenstein’s book elaborated on the connection between the start of printing in Western Europe and the Reformation at the start of the sixteenth century. She has argued that without the printing press, the Reformation could never have been as successful as it was. Pettegree and Weduwen have written a book that is clearly part of the French school, in which they elaborate on the history of producing, disseminating and consuming books in early modern Europe.
7 Paul Dijstelberge, “Towards an Atlas of Book Design – A Modest Research Proposal”, Quaerendo 42
(2012), 212. This does not mean, however, that no research has been conducted on the changes in paratext between the Late Middle Ages and the Renaissance. Bonnie Mak, for example, wrote a book titled How the page matters, in which she researched the changing page from ancient texts to digital files, but her research is concise and deals only with a few specific paratextual elements. This is the case in other articles and books as well. Another example is the article “Material Approaches to Exploring the Borders of Paratext”, by Sirkku Ruokkeinen and Aino Liira (Textual Cultures 11 (2017), 106-129). What makes my research different, is the comparison of almost 90 paratextual elements instead of only a few.
8 Encyclopaedia Britannica, S.v. “Incunabula”, October 20, 2013, access May 3, 2020,
https://www.britannica.com/topic/incunabula.
9 Mary A. Rouse and Richard H. Rouse, Authentic Witnesses: Approaches to Medieval Texts and Manuscripts (Notre Dame: University of Notre Dame Press, 1991), 191-219; Paul Saenger, ‘Books of
Hours and the Reading Habits of the Later Middle Ages’, in The Culture of Print: Power and the Uses of
7
the period after Gutenberg invented moveable type around 1450.10 Changing reading
habits and paratext, in return, may have influenced reception of texts. Therefore, a study of the changes in paratexts between the late Middle Ages and the end of the Renaissance is essential to understand how the ‘order of the book’ as we know it came about.11
For example, Frans Janssen tentatively suggested that choices in typography – and hence the changes in it over time – were not due to only authors or only printers or only publishers. Changes in typography were sometimes conscious but more often unconscious decisions made in the interaction between the author, printer and publisher.12 Also, changes in typography and paratext may not always have been
chronological and some typographical and paratextual features have actually disappeared over time. An example is described by Lynne Truss, who imagines the horror sixteenth-century authors and readers would probably have felt when confronted with our current use of punctuation.13 Typography could be changed and
adapted along with changing audiences or simply because the book tradition changed – that is, the idea of what the book was. Authors, but more so printers and publishers anticipated on the status quo of the contemporary book and used typography to display the text in the way that it would best be understood and recognised by their readers.14
A comparative approach in this respect is useful, in order to count specific paratexts from the High Middle Ages up to the early modern times in Western Europe. In this period, the number and variety of paratextual elements increased rapidly and the format of the manuscript changed via the incunabulum to the printed book that was
10 Asa Briggs and Peter Burke, A Social History of the Media: From Gutenberg to the Internet, 3rd ed.
(London: Polity Press, 2018 (ebook)), 42 and 112-121.
11 Dijstelberge, “Towards an Atlas of Book Design”, 209-211; Adriaan van der Weel, Changing our textual minds: Towards a digital order of knowledge (Manchester: Manchester University Press, 2011), 2. 12 Frans A. Janssen, Auteur en drukker in de geschiedenis van de typografische vormgeving (Amsterdam:
De Buitenkant, 1989), 18.
13 Lynne Truss, Eats, Shoots and Leaves: The Zero Tolerance Approach to Punctuation (London: Profile
Books, 2007), 22-23.
8
familiar to readers for more than two centuries and is still recognised by readers today. The numerous paratextual elements that were included in printed books guided the reader in his or her understanding of the text. Notes in the margins by readers testify of this transition and new text receptions.15
By studying one text that has been written, copied, and read from Antiquity well into the Renaissance, a good comparison can be made between paratext in the Early and High Middle Ages (500-1200), the late Middle Ages (1200-1450), and the Renaissance (1450-1600). The text chosen for this purpose was a bestseller over the ages. The author Flavius Josephus was a Jewish historian that lived in the first age of the common era (c. 38-100). According to his autobiography, he betrayed his fellow Jewish soldiers to serve the Roman empire for the rest of his life.16 As a Roman citizen and follower of the
emperor, he wrote Antiquitates Iudaicae. This work, written in Greek, consists of ten books on the history of the Jewish people from the moment of God's creation described in Genesis up to around 400 BCE, the moment from which the Tanakh is silent on the history of the Jews. In books eleven to twenty, Josephus described Jewish history from that moment up to his own times, when the Jewish War over Jerusalem was fought. He described that war and his part in it in detail in another text, De bello Iudaico.17
His history was copied and read throughout Antiquity and the early Middle Ages because of its high quality, but more so because of Josephus' testimony on Jesus Christ, described in the eighteenth book of Antiquitates Iudaicae. Josephus' text is one of the most important sources, apart from the Bible, that testifies of Jesus' existence. This passage, named the “Testimonium Flavianum” was copied frequently, apart or in the context of its original text. According to medieval copies and translations, Josephus recognised Jesus Christ as the Jewish Messiah of whom the Tanakh prophesied. Later,
15 Dijstelberge, “Towards an Atlas of Book Design”, 213.
16 David Daube, Typologie im Werk des Flavius Josephus (Munich: Bayerische Akademie der
Wissenschaften, 1977), 4-5.
17 Richard Matthew Pollard, “The De excidio of “Hegesippus” and the reception of Josephus in the early
9
already in 1592, the Testimonium Flavianum was doubted for its authenticity, precisely because Josephus named Jesus the Messiah. This part was probably added or adapted by later Christian copyists.18 Nonetheless, it is no surprise that in the Middle Ages, his
text was regarded to be very important for the Christian faith and hence was translated from Greek to Latin and read by monks, priests, scholars and others well into the Renaissance.19
The Renaissance period saw a rising interest in the ancient authors and their non-Christian texts. As such, Josephus was not overlooked and cited frequently by scholars in this era too.20 The first printed edition of his text already appeared in 1470, in
Augsburg.21 After this, many more printed editions appeared in Western Europe,
although translations in vernacular languages were scarce.
The scope and aim of this thesis, then, are ambitious and broad, since the research question is:
How and why have paratexts, reading habits, and reception of Antiquitates Iudaicae by Flavius Josephus changed between 800 and 1600?
Because of the large number of manuscripts and printed editions, the changes in the paratext of Antiquitates Iudaicae are well traceable from the Middle Ages up to the Renaissance.
18 G. Caruso, s.v. “Testimonium Flavianum", Encyclopedia of Ancient Christianity, vol. 3, gen. ed. Angelo
di Berardino (Westmont: Intervarsity Press, 2014).
19 Daniel Stein Kokin, “The Josephan Renaissance: Flavius Josephus and his writings in Italian humanist
discourse”, Viator 47 (2016), 205.
20 Stein Kokin, “The Josephan Renaissance”, 205.
21 USTC 746500. Flavius Josephus, De antiquitate Judaica/De bello Judaico (Augsburg: Johann Schüssler,
10 Methodology
Due to the measures against the coronavirus, this investigation on changing paratexts in Antiquitates Iudaicae could only be conducted digitally. The majority of digitised printed editions from the incunabula and other printed books of Josephus' text were found using the Universal Short Title Catalogue (USTC).22 A great number of European
libraries and archives from all over the world are gathered in this database, creating an enormous amount of metadata on early modern books printed between 1450 and 1650. For the author-entry "Josephus, Flavius", 294 different editions turn up. Narrowing this selection down to digitised editions only, 108 editions appear. From these 108 editions, I discerned 60 books that included Antiquitates Iudaicae, ranging from being published between 1475 and 1600.
For the manuscripts, I used the digital inventory made by Roger Pearse, who lists 30 manuscripts that contain (parts of) the first ten books of Antiquitates Iudaicae and another 38 manuscripts that contain (parts of) the other ten books and the text Vita from Flavius Josephus. Four of these entries contain both texts, which means that Pearse lists 64 manuscripts in total that contain Josephus’ text, written in Greek or Latin.23 The
scholar Franz Blatt has listed 171 surviving Latin copies of Josephus’ Antiquitates Iudaicae, but due to the limited library services from April 2020 onwards, I could not get access to his book.24 Searching the individual databases of the libraries Pearse mentions,
I found 51 Latin manuscripts containing Antiquitates Iudaicae that have been digitised.
22 USTC.ac.uk. “USTC: A digital bibliography of early modern print culture”. Access May 4, 2020.
https://www.ustc.ac.uk/.
23 Roger Pearse, “Josephus: all the Manuscripts”, Tertullian.org, March 27, 2004, access April 28, 2020,
http://www.tertullian.org/rpearse/manuscripts/josephus_all.htm. I did not expect Roger Pearse to be a very trustworthy source, but due to the coronavirus, I could not reach the books that list Josephus’ medieval manuscripts in the university library at the start of the research period. Pearse is very open about his approach on his blog: https://www.roger-pearse.com/weblog/about/ and I do not think he has any interest in pre-selecting or misinforming the readers of his page on Josephus’ manuscripts.
24 Franz Blatt, The Latin Josephus (Kopenhagen: Universitetsforlaget I Aarhus Ejnar Munksgaard, 1958),
11
Comparing these entries with Blatt’s list later, I concluded that I searched most libraries and archives that have several manuscript texts digitised. From these 51 Latin manuscripts, however, I could only use 37 manuscripts as thirteen of them only contain small excerpts from Antiquitates Iudaicae and one digital copy was not available. In the bibliography of primary sources, the manuscripts and printed books from 800 to 1600 are listed. The manuscripts that were unfit for research are presented in appendix A.
Considering that I would have used 163 manuscripts if I had had access to all the manuscripts listed by Blatt, this means that my sample ratio is 37:163.25 Based on
normal statistical practice, this means that my largest error margin is 14%. But for specific paratextual elements, this error margin is lower. The error margins for every countable paratextual element are listed in appendix B. Although an error margin of 14% is not low, it is not too high either and I concluded that the sample of 37 digitised copies suffices to represent all surviving medieval manuscripts of Antiquitates Iudaicae.
I constructed a questionnaire to analyse the 37 manuscripts and 60 printed books for paratextual elements.26 This questionnaire can be found in appendix C. Before I started
the research, I tried the questionnaire on three texts, one manuscript, one incunabulum and one printed book from after 1550 and improved the questionnaire accordingly. Even though the starting point for the questionnaire was one specified to printed books, I took care to include paratextual elements that are common in manuscripts as well. For this purpose, I also used Rosalind Brown-Grant et al.’s book, titled Inscribing Knowledge in the Medieval Book: The Power of Paratexts.27 In my opinion and to the best of my
25 Blatt lists the number of folia of the manuscripts. Based on my criteria for the selection, I would not
have used eight manuscripts because they contain merely excerpts of Antiquitates Iudaicae.
26 The idea to use a questionnaire and to construct a database came from my supervisor, Dr Paul
Dijstelberge, and he got part of this idea from Frans A. Janssen. Dijstelberge’s ideas are explicated in the article “Towards an Atlas of Book Design”. Janssen’s construction of a typographical matrix can be found in the essay ‘Layout as means of identification?’ (part of Technique and Design in the history of printing) in which he argues that “[a] model of description for layout would have to contain at least some twenty elements” (p. 105 and 107). The questionnaire I constructed meets this requirement easily, as it counts 89 questions about various typographical elements.
27 Rosalind Brown-Grant et al. (eds.), Inscribing Knowledge in the Medieval Book: The Power of Paratexts
12
knowledge, this is one of the most elaborate volumes on paratext in medieval manuscripts. During the analysis, I altered some questions, so multiple versions are listed in the appendix.
To clearly write down the findings of the comparison and the dynamics between paratext, reading habits, and reception are discussed more elaborately in the first chapter. In the second chapter, the paratexts of the medieval manuscripts are analysed, followed by the incunabula in chapter three and the sixteenth-century printed editions in chapter four. In a concluding chapter, the comparison is elaborated and conclusions are drawn in order to answer the main question of the thesis.
13 1. Paratext, Reading Habits, and Reception
‘(…) paratext, like texts themselves, has a history’.28 Paratext developed and changed
over time. When one scans the manuscripts, incunabula and other printed editions of Josephus’ Antiquitates Iudaicae, such a conclusion is not surprising. Manuscripts of the ninth century look very different from the printed editions of the sixteenth. And indeed, one of the major differences between the two is the lack of paratext in the former and the abundance of it in the latter. When paratext changes, it has repercussions for the whole text. As defined by Genette and Thomas Bredehoft and in my own definition, paratext guides the reader through the pages and the text’s interpretation and it puts the text into a historical context.29 According to Dorothee Birke and Birte Christ, paratext
can be identified based on three functions: 1. Interpretative paratext, 2. commercial paratext, and 3. navigational paratext. Interpretative paratext guides the reader in his or her understanding of the text, while commercial paratext aims to sell the text to the reader, and navigational paratext helps the reader to locate certain passages in the text. In this way, changes in paratext can affect reading habits, certainly when paratextual elements function in an interpretative or commercial manner. These kinds of paratexts are easily moved around within the order of the book, as they are about the main text, but not directly connected to the main text, like navigational paratext is.30 This feature
of ‘optionality’ of changing paratexts is one of the main distinctions between text and paratext, Sirkku Ruokkeinen and Aino Liira have argued.31
An example of this is given by Paul Saenger, who suggests that spacing between words promoted silent reading in the High Middle Ages. Because of spacing, there was
28 Rosalind Brown-Grant et al. ‘Introduction’, in Inscribing Knowledge in the Medieval Book: The Power of Paratexts (Berlin: De Gruyter, 2020), 1.
29 Thomas A. Bredehoft, The Visible Text: Textual Production and Reproduction from Beowulf to Maus
(Oxford: Oxford University Press, 2014), 103; Genette, Paratexts, 1-2.
30 Dorothee Birke and Birte Christ, “Paratext and Digitized Narrative: Mapping the Field”, Narrative 21
(2013), 76-78.
14
no longer a need to pronounce every letter separately in order to understand the text, since words were presented as separate entities. Silent reading promoted the individual reading of texts and provided readers with a means to read what they wanted to read rather than what they were expected to read.32 Although spacing in words had existed
already in Antiquity, reading aloud became much more common from the twelfth century onwards, for example among university students.33
The dynamic between paratext and reading habits could be reversed as well. Although many divisions in text like chapters, paragraphs and books were not new, their systematic introduction in late medieval texts reflects a new attitude towards information. Divisions in text longer than one page were created and increasingly systematically applied from the High and Late Middle Ages onwards. This was the result of a growing body of literature that became available to readers and the shift to a different manner of preaching.34 The structure of texts could have been remembered
by readers in the early and High Middle Ages, but memorisation was insufficient for reading and interpreting all available texts in the late Middle Ages. The gloss solved this problem, for example in Bibles. It was a short note on the explanation or translation of a word or a sentence. Preachers could refer to and easily search for specific parts of text in the gloss. Glosses in the Bible, however, grew to be long texts, mostly in the margins, explaining the meaning and interpretation of certain passages.35 In the thirteenth
century, therefore, even the gloss was no longer sufficient to hold all the information that preachers and readers were looking for. This was why the alphabetical concordance
32 Paul Saenger, ‘Reading in the Later Middle Ages’, in A History of Reading in the West, eds. Guglielmo
Cavallo and Roger Chartier (Hoboken: Wiley, 1999), 130, 133 and 139-140.
33 M.B. Parkes, ‘Reading, Copying and Interpreting a Text in the Early Middle Ages’, in A History of Reading in the West, eds. Guglielmo Cavallo and Roger Chartier (Hoboken: Wiley, 1999), 92; Saenger,
‘Reading in the Later Middle Ages’, 120, 130, 133 and 139-140.
34 Rouse and Rouse, Authentic Witnesses, 192-193, 197, 201 and 213-214; Saenger, ‘Reading in the Later
Middle Ages’, 133-134.
35 Peter Beal, s.v. “Gloss”, in A Dictionary of English Manuscript Terminology, 1450 to 2000 (New York:
15
was introduced in texts on a large scale.36 More generally, tables of contents and
capitalisation of section titles represented new hierarchies within texts.37 The
systematic introduction of these kind of reading aids show that paratext could also follow changing reading habits and new functions and uses of the book.
To speak of “paratext” in the case of medieval manuscripts is sometimes contested, as paratext in medieval manuscripts was not always clearly separated from the text itself and was often fully part of it. Since scribes copied all text from their exemplar, it is odd to speak of paratext in medieval manuscripts in this way.38 In line with this argument, it
seems paratext in the Middle Ages might mainly have had a navigational function, as it was mostly closely connected to the main text and was changed only when the main text changed.39 Therefore, Bredehoft suggests that when paratext was copied with the
text, it may have been regarded not as paratext that guided the reader through the text at all, but rather as part of the text itself.40 If this is indeed the case, “mise-en-page”
might actually be a term better suited in the analysis of medieval manuscripts, as mise-en-page reflects the scribe’s or the patron’s view of the structure and hierarchy of the text on the parchment leaf.41
Still and certainly for the late Middle Ages, paratext is a useful term, as scribes became aware of the interpretative functions of paratext long before the printed book became common, nor are the different functions of paratextual elements always very clear-cut.42 Whereas mise-en-page influences the visual structure of the book, paratext
is the intention behind the structure of the book and changes the text consciously. An
36 Rouse and Rouse, Authentic Witnesses, 192-193, 197, 201 and 213-214. 37 Saenger, ‘Reading in the Later Middle Ages’, 133-134.
38 Bredehoft, The Visible Text, 55-56.
39 Erik Kwakkel, “‘Dit boek heeft niet de vereiste breedte’: Afwijkende bladdimensies in de elfde en
twaalfde eeuw”, Jaarboek voor Nederlandse boekgeschiedenis 19 (2012), 34.
40 Bredehoft, The Visible Text, 29, 56, 66-67 and 124. 41 Brown-Grant et al., ‘Introduction’, 5.
42 Bredehoft, The Visible Text, 63-64; Brown-Grant et al., ‘Introduction’, 5; Ruokkeinen and Liira,
“Material Approaches to Exploring The Borders of Paratext”, 116; Saenger, ‘Reading in the Later Middle Ages’, 120.
16
example of the former is the division of the text in two columns, while paratext is the choice to start a new book at the top of a new column in order to make it better recognisable. In this thesis, I define mise-en-page as a practical view to the page and space division, while paratext is the choice of the producer of a text to change its structure and meaning and guide the interpretation of the reader. Both medieval scribes and rubricators as well as early modern printers and editors made these choices, consciously or unconsciously. Scribes and rubricators made those choices themselves or were paid by their patrons to make changes to paratext, while printers and editors were often guided by the demands of their customers.43
With the growing awareness that paratext was apart from text, scribes started to adapt old texts to the new context of their own time. In this way, scribes, rubricators, and illuminators influenced the reading and reception of texts by changing and adding paratext.44 For example, an illuminator could enlarge an existing miniature in the new
copy or even add a whole new miniature to guide the reader to a chapter or a part of the codex that interested him or her most.45 Also, scribes could emphasise certain parts
of the text by rubricating them more heavily, or diminish their importance and conspicuousness by not adding any rubrication at all.46 In many occasions new paratexts
were added or adapted based on the reading skills, interests and preferences of the patrons of the manuscript, or to make the inexperienced lay reader better understand the original context of a text.47 The latter may seem a contradiction, but it is not. The
43 Kwakkel, “‘Dit boek heeft niet de vereiste breedte’”, 37; Margareth M. Smith, ‘Red as a Textual
Element during the Transition from Manuscript to Print’, in Textual Cultures: Cultural Texts, eds. Orietta da Rold and Elaine Treharne (Cambridge: D. S. Brewer, 2010), 27
44 Géraldine Veysseyre, ‘Structuring, Stressing, or Recasting Knowledge on the Page? Rubrication in the
Manuscript Copies of the Pèlerinage de l’âme by Guillaume de Deguileville’, in Inscribing Knowledge in
the Medieval Book: The Power of Paratexts, eds. Rosalind Brown-Grant et al. (Berlin: De Gruyter, 2020),
162.
45 Anne D. Hedeman, ‘Translating Prologues and Prologue Illustration in French Historical Texts’, in Inscribing Knowledge in the Medieval Book: The Power of Paratexts, eds. Rosalind Brown-Grant et al.
(Berlin: De Gruyter, 2020), 210.
46 Veysseyre, ‘Structuring, Stressing, or Recasting Knowledge on the Page?’, 164.
47 Hedeman, ‘Translating Prologues and Prologue Illustration in French Historical Texts’, 205; Veysseyre,
17
scribe changed the paratext for a new audience to clarify and translate it to the reader’s own era.48
This may also be the reason why sixteenth-century printed editions had so many more paratextual elements that were used in a more consistent way than in the earlier codices. Publishers – or their editors – wanted to show the original context of a text to the contemporary reader.49 It was central to the Renaissance period that antique texts
became relevant and were thus read again in the original language, translated and printed. New paratextual elements were introduced to make these texts understandable to early modern readers.50 Since books were no longer read only by the
clergy, rich patrons or scholars, the need for more extensive paratext grew as well.51 In
the manuscript culture of monasteries, it had even been possible to explain or convey the content of paratext orally, as scribe and reader were likely to know each other.52 In
such an environment there was stability and control over the reading of texts.53 With
the advent of universities in the twelfth and thirteenth centuries, the access to texts opened up to students and the late Middle Ages saw a growing readership consisting of nobles and wealthy urban families.54 Using notes and citation, scholars could still have
control over the text and explain it to readers, by explaining the author’s meaning in the margins.55
Selling books in a bookshop was even less controlled by religious or governmental authorities, as lay readers could interpret the text on their own once they had bought a book.56 Literacy rose from the fifteenth century onwards, with new audiences that
48 Veysseyre, ‘Structuring, Stressing, or Recasting Knowledge on the Page?’, 176. 49 Bonnie Mak, How the Page Matters (Toronto: University of Toronto Press, 2011), 40. 50 Brown-Grant et al., ‘Introduction’, 1.
51 Mak, How the Page Matters, 26 and 43.
52 Patrizia Carmassi, ‘Book Material, Production, and Use from the Point of View of the Paratext’, in Inscribing Knowledge in the Medieval Book: The Power of Paratexts, eds. Rosalind Brown-Grant et al.
(Berlin: De Gruyter, 2020), 313; Mak, How the Page Matters, 26.
53 Saenger, ‘Reading in the Later Middle Ages’, 137.
54 Mak, How the Page Matters, 43; Saenger, ‘Reading in the Later Middle Ages’, 130 and 148. 55 Saenger, ‘Reading in the Later Middle Ages’, 130.
18
bought and read books.57 To these new readers the context of a text was no longer
explained orally. It is one of the reasons why the need for paratext grew in the Renaissance, as the context of the text had to be presented to the reader visually.58
Even the absence of text can be part of paratext. This chapter opened with a direct quote from Brown-Grant et al.’s book. Instead of inserting this quotation in the text, marked by direct quotation marks, I could also have put the quote in italics in a separate sentence, leaving a line open between the quote and the text. Such a manner of citing would have presented the quote with more authority to the reader. It illustrates the power paratext holds over the text, and also how it both influences and is influenced by reading habits.59
The dynamic interaction should be explained more in-depth. Paratext points to the main passages of the text, to the most important parts according to the editor. In this way, paratextual elements can pave the way for what the reader will read with more attention than other passages. Paratext in the form of marginalia, in particular, can have a great impact on the reader’s reception of the text, as the marginal notes often guide the reader’s interpretation of the text.60 Of course, readers are not obliged to follow the
paratext’s instructions and emphases, but a reader who is used to see and read a lot of paratextual elements in codices is likely to follow its structure and hierarchy unconsciously.61
57 Ian Frederick Moulton, ‘Introduction’, in Reading and Literacy in the Middle Ages and Renaissance
(Turnhout: Brepols Publishers, 2004), xi.
58 Mak, How the Page Matters, 26. 59 Brown-Grant et al., ‘Introduction’, 11.
60 Dirk Schultze, “Wisdom in the margins: text and paratext in The Seven Points of True Love and Everlasting Wisdom”, Études Anglaises 66 (2013), 348 and 353-354.
61 Roger Chartier, ‘General Introduction: Print Culture’, in The Culture of Print: Power and the Uses of Print in Early Modern Europe (Princeton: Princeton University Press, 1989), 8; Schultze, “Wisdom in the
19
In the incunabula period, the paratextual elements of well-known texts did not change a lot at first, even if the manner of production had changed radically.62 After this
period, the paratext of books started changing more rapidly. For example, the production of cheaper books on the printing press allowed for printers to create bigger “collected works” of well-known authors.63 For Flavius Josephus, this meant that rather
than just the Antiquitates Iudaicae, and sometimes De bello Iudaico – texts that were sometimes copied in one edition – the printed Opera also included Josephus’ other texts: Vita, Contra Apionem, and a philosophical treatise on the martyrdom of the Maccabees attributed to Josephus.64 The printer no longer depended on available
exemplars of Josephus’ texts, as the exchange of texts was much faster due to the growing trade in printed books and to scholarly networks.65 In these “collected works”,
it was important to distinguish between the different texts of Josephus.66 Paratext was
used to create this distinction.
Translations from Latin into vernacular languages provided new readers without a classical education with literature and history works like Antiquitates Iudaicae. Translations, however, always have a need for paratext, as they imply the existence of an “original” text.67 In a translation, not only the text and the paratext changes, but even
the interpretation of the reader is influenced. Languages can never be fully translated
62 Martin K. Foys, ‘Medieval manuscripts: media archaeology and the digital incunable', in The Medieval Manuscript Book: Cultural Approaches, eds. Michael Johnston and Michael van Dussen (Cambridge:
Cambridge University Press, 2015), 119-120.
63 Vincent Gillespie, ‘Syon and the English Market for Continental Printed Books: The Incunable Phase',
in Syon Abbey and its Books: Reading, Writing and Religion, c. 1400-1700, ed. E.A. Jones. (Suffolk: Boydell & Brewer, 2010), 110.
64 The first edition in the USTC that includes all these works in the title, is the book printed by Nicolas
des Prez in Paris, 1511: USTC 180621, Flavius Josephus, Que hoc volumine contineantur Josephi, vita,
tabula sive index alphabeticus continens precipuas materias. Antiquitatum libri viginti. Contra Appionem grammaticum libri duo. De bello Iudaico libri septem (Paris: Nicholas des Prez, 1511), access May 8,
2020, https://www.ustc.ac.uk/editions/180621.
65 In the incunabula period, the reader was probably still very dependent on the supply of the printer
closest to him or her. See Gillespie, ‘Syon and the English Market for Continental Printed Books’, 116.
66 Gillespie, ‘Syon and the English Market for Continental Printed Books’, 110.
67 Bredehoft, The Visible Text, 70; Freyja Cox Jensen, “The Popularity of Ancient Historians, 1450-1600", Historical Journal 61 (2018), 569.
20
because they cross borders between cultural systems of meaning.68 In some
translations, then, paratext was easily changed and prologues or frontispieces were added to the translation that the exemplar never had. Using paratext, the translation was adapted to new audiences.69 Not all marginalia included in Latin editions were
copied in vernacular ones.70 Therefore, translations had different paratexts from the
Latin codices.
In the case of Flavius Josephus, the reception of Antiquitates Iudaicae may not have changed much between the Middle Ages and the Renaissance. In the Middle Ages, Josephus was read because of his value in securing the dogmas of Christianity. During the Renaissance, this was still the case. In both eras, however, it seems remarkable that Josephus, a Jew, was a popular author.71 Jews, after all, were far from accepted by
Christian society in Western Europe. In the same period that universities grew into important institutions and focus shifted from reading aloud to silent reading, the states of Western Europe grew stronger and centralised. With centralisation and the growing awareness of a common jurisdictional identity, Jews were increasingly perceived as outcasts.72 Already in Late Antiquity, Christians had reproached Jews for not accepting
Jesus as their Messiah, and indeed condemning him to crucifixion. In this line of thought, the destruction of Jerusalem and the Jewish diaspora was the fault of the Jews themselves, since it was God’s punishment for their unwillingness to believe in Christ.73
68 Cox Jensen, “The Popularity of Ancient Historians”, 569.
69 Brown-Grant et al., ‘Introduction’, 11; Hedeman, ‘Translating Prologues and Prologue Illustration in
French Historical Texts’, 208.
70 Saenger, ‘Reading in the Later Middle Ages’, 145.
71 Cox Jensen, “The Popularity of Ancient Historians”, 580-581 and 586; Nicholas Vincent, ‘William of
Newburgh, Josephus and the New Titus', in Christians and Jews in Angevin England: The York Massacre
of 1190, Narratives and Contexts, eds. Sarah Rees Jones and Sethina Watson (Suffolk: Boydell & Brewer,
2013), 58.
72 Deeana Copeland Klepper, ‘Disentangling Heretics, Jews, and Muslims: Imagining Infidels in Late
Medieval Pastoral Manuals’, in Late Medieval Heresy: New Perspectives, eds. Michael D. Bailey and Sean L. Field (Suffolk: Boydell & Brewer, 2018), 137-138.
21
These arguments fitted the beliefs of medieval Christians in Europe, as they forged a distinctive and uniform Christian identity.74
Flavius Josephus was actually an important source for establishing this train of thought. The Testimonium Flavianum was seen as the proof that Jews knew that Jesus Christ was the Messiah, but still had him killed.75 Even though this excerpt is highly
contested nowadays, the Testimonium Flavianum was a convenient document in Christian justification for exclusion and persecution of Jews.76 Humanist authors and
readers held this view too, since Josephus was still a popular author who was being read and printed in the fifteenth and sixteenth century.77 His popularity may seem to be a
paradox, but because of the Testimonium Flavianum, Josephus himself was excused from the anti-Semite feelings among medieval and Renaissance readers.78
Conclusion
Although the connection between paratext and reading habits has been proven plenty in literature, the direct connection between paratext, reading habits and reception is not always clear. This is certainly the case with Antiquitates Iudaicae. Since I will focus mostly on peritext in my research due to the high number of primary sources, one could argue that I do not analyse the dynamic between paratext and reception. Throughout this thesis I will try to connect developments and changes in paratext to the contemporary reception of Antiquitates Iudaicae. Marginalia in particular can shed light on the reception of intended readers, as they interact directly with the assertions in the
74 Copeland Klepper, ‘Disentangling Heretics, Jews, and Muslims’, 139-140.
75 Pollard, “The De excidio of “Hegesippus” and the reception of Josephus in the early Middle Ages”, 78. 76 Stein Kokin, “The Josephan Renaissance”, 230.
77 Cox Jensen, “The Popularity of Ancient Historians”, 580-581; Stein Kokin, “The Josephan Renaissance”,
247.
22
text. Whether paratext influences reading habits or the other way around, I hope to show the dynamics of changing reception as well, however subtle these changes might be.
23
2. Paratext by Exception in Manuscript Antiquitates Iudaicae
Unlike printed editions, manuscripts are unique copies of a text. Even though workshops were founded in the late Middle Ages where manuscripts were copied and illuminated in the same style by different hands, each manuscript varied from other copies.79
Whenever scribes copied text from other manuscripts, they were likely to imitate the structure and paratext of the exemplar. Some scribes, however, changed the mise-en-page a little, especially when new audiences were to be addressed.80 Over time, the
copying habits of scribes resulted in “book families” of manuscripts that are similar to each other as they originated from the same exemplar, but differ because they were copied by different scribes. This logic – that goes against the ‘logic of the copy’, in which the scribe is aware that he is making a copy rather than a new book – may indeed contradict the concept of paratext in manuscripts, as some scribes simply copied all the text from his predecessor, regardless whether it was a chapter, a header, or an incipit.81
In the analysis of the 37 manuscript Antiquitates Iudaicae, however, some ideas on paratext can still be distilled, as, according to Bredehoft, the ‘logic of the copy’ already emerged in the late Middle Ages.82 Some scribes did make deliberate choices – based
on their own insight or on the wishes of their patrons – to change the structure and hierarchy of texts.83
Out of the 37 manuscripts in the selection, eighteen manuscripts include all twenty books of Josephus’ Antiquitates Iudaicae. These “complete” manuscripts transmitted
79 Christine Mary Geisler Andrews, The Boucicaut Workshop and the Commercial Production of Books of Hours in Early Fifteenth-Century Paris (Evanston: Northwestern University, 2006), 48 and 209.
80 Orietta da Rold, ‘Textual Copying and Transmission’, in The Oxford Handbook of Medieval Literature in English, eds. Greg Walker and Elaine Treharne (Oxford: Oxford University Press, 2010), 48-49; Veysseyre,
‘Structuring, Stressing, or Recasting Knowledge on the Page?’, 176.
81 Bredehoft, The Visible Text, 6. 82 Ibid. 6, 23 and 66.
83 For examples, see Brown-Grant et al. (eds.), Inscribing Knowledge in the Medieval Book. Although the
authors in this edited volume give many examples of paratext in medieval manuscripts, the examples all concern individual manuscripts and a larger argument on the scale of the structural implementation of the paratextual elements discussed is not provided.
24
the prologue as well. Thirteen manuscripts not only included Antiquitates Iudaicae but also De bello Iudaico.84 Two texts included Contra Apionem.85 Remarkably, the undated
manuscript Città del Vaticano, Biblioteca Apostolica Vaticana, Ottobonianus latinus 84 has added Contra Apionem, but left out the more popular and better-known text De bello Iudaico.86 Città del Vaticano, Biblioteca Apostolica Vaticana, Vaticanus latinus 1994
included De bello Iudaico, but between books 1-12 and 18-20 of the Antiquitates Iudaicae. It was not uncommon to put De bello Iudaico within the Antiquitates Iudaicae, but all the other manuscripts of this selection count 27 books, twenty for Antiquitates Iudaicae and seven for De bello Iudaico, whereas Vaticanus latinus 1994 counts 22 books. Books 13-17 were never copied or may have gone missing over time.
Another case stands out. Paris, Bibliothèque nationale de France, Latin 5763 includes books 13-18 of Antiquitates Iudaicae, that was sometimes divided in two codices, but these books are bound together with De bello gallico of Gaius Iulius Caesar. Whether
84 Berlin, Staatsbibliothek zu Berlin, Lat. Fol. ms 625, 1300-1400, 167 fols., access May 26, 2020,
http://resolver.staatsbibliothek-berlin.de/SBB0000530400000000; Città del Vaticano, Biblioteca
Apostolica Vaticana, Archivio del Capitoli di San Pietro ms A.39, 1075-1125, 212 fols., access May 26,
2020, https://opac.vatlib.it/mss/detail/215323; Città del Vaticano, Biblioteca Apostolica Vaticana, Palatini latini ms 815, 1439, 243 fols., access May 26, 2020,
https://digi.ub.uni-heidelberg.de/diglit/bav_pal_lat_815/0007; Città del Vaticano, Biblioteca Apostolica Vaticana, Vaticani latini ms 1994, 1300-1400, 119 fols., access May 26, 2020, https://opac.vatlib.it/mss/detail/3162; Città del Vaticano, Biblioteca Apostolica Vaticana, Vaticani latini ms 1997, 1125-1175, 238 fols., access May 26, 2020, https://opac.vatlib.it/mss/detail/3170; London, British Library, Arundel ms 94, 1125-1175, 83 fols., access May 26, 2020, http://www.bl.uk/manuscripts/FullDisplay.aspx?ref=Arundel_MS_94; London, British Library, Royal ms 13 D VII, 1120-1140; Paris, Bibliothèque nationale de France, Français ms 6446, 1400-1415, 413 fols., access May 26, 2020, https://gallica.bnf.fr/ark:/12148/btv1b8451605w; Paris, Bibliothèque nationale de France, Latin ms 5045 (2), 1100-1125, 102 fols., access May 26, 2020,
https://gallica.bnf.fr/ark:/12148/btv1b52000584z; Paris, Bibliothèque nationale de France, Latin ms 8835, 1461, 178 fols., access May 26, 2020, https://gallica.bnf.fr/ark:/12148/btv1b52511828f; Paris,
Bibliothèque nationale de France, Latin ms 8959, 1100-1200, 251 fols., access May 26, 2020, https://gallica.bnf.fr/ark:/12148/btv1b8510034c; Paris, Bibliothèque nationale de France, Latin ms 14361, 1100-1200, 264 fols., access May 26, 2020, https://gallica.bnf.fr/ark:/12148/btv1b10036672q; Paris, Bibliothèque nationale de France, NAF ms 21013, 1400-1420, 298 fols., access May 26, 2020,
https://gallica.bnf.fr/ark:/12148/btv1b84559070.
85 Berlin, Staatsbibliothek zu Berlin, Lat. Fol. ms 625, 1300-1400; Città del Vaticano, Biblioteca Apostolica Vaticana, Ottoboniani latini ms 84, n.d., 197 fols., access May 26, 2020,
https://opac.vatlib.it/mss/detail/3176.
86 Although undated, I suspect from the paratext and the script that it was written in the late Middle
25
this was a mistake of a later binder (the cover is not original and the two hands of the respective texts differ) or a conscious decision is impossible to establish, but the similarity in titles might well have caused a mix-up.
Front matters
Antiquitates Iudaicae starts with a prologue and for that reason is a good case study in establishing how medieval scribes perceived paratext. For today’s readers, prologues belong to the front matter of a book, separated from the main text. Modern front matters may include a half-title page, a frontispiece, a title page, a colophon, dedications, acknowledgments, prefaces, a table of contents, list of illustrations etc. All of these different paratextual elements have a history and were added over time. Some were used during specific periods only.87 For example, laudatory poems were quite
common in the first two centuries of the printed book, but are hardly found in modern books.
The manuscripts of the selection barely have any paratextual elements before the main text apart from the prologues. However, there are two codices that include a “proto-title page”, in-between a half-title page that only mentions title and author and the full title page, which enumerates all details of a book. Firenze, Biblioteca Medicea Laurenziana, San Marco 385 is dated 1100-1200 and lists the title and author of the text on folio recto 1 (illus. 3). Although it fills a complete page, this proto-title page can best be seen as an extraordinary incipit, as its text announces the start of the prologue of the book rather than the title of the book itself. Still, it was included originally by the scribe or rubricator.
26
The other manuscript with a proto-title page is Città del Vaticano, Biblioteca Apostolica Vaticana, Urbinatus latinus 400, dated 1300-1400 (illus. 4). The title is set on the verso before the prologue, but its ornamentation corresponds in style and colour with the rest of the manuscript. The proto-title page announces the title and the author of the text and proclaims Antiquitates Iudaicae as a most marvellous read. Another detail that is included is the translator’s name, Rufinus, who allegedly translated Antiquitates Iudaicae from Greek to Latin.88 As such it is a more recognisable title page
for modern readers, which contains the title, name of the author and the production details.89
Noteworthy is that both manuscripts are written in scripta continua with capital letters, which presents the text as Roman inscriptions and the following tomes as objects to observe and admire. Also, in both cases, the title consists of one sentence and includes the author’s name, instead of breaking the various pieces of information (author, title, translator) apart. Apart from that, the two manuscripts cannot be linked in time or place of origin, although the proto-title page was probably a feature of manuscripts for wealthy owners in particular, as the proto-title page presented the text as a remnant of great Roman and Greek culture, monuments to be admired by rich patrons. Both manuscripts are elaborately decorated, which suggests their original owners were rich indeed.
Another element of front matters is the table of contents that lists parts, chapters and paragraphs with references to page numbers. No medieval manuscript of this selection includes a table of contents at the front of the Antiquitates Iudaicae, although the majority of the manuscripts has a table of contents before each book (illus. 5). The exception on this rule is Paris, Bibliothèque Mazarine, Bibliothèque Mazarine 1581, which includes a table of contents for all twenty books at the front (illus. 6a and b). This
88 Pollard, “The De excidio of “Hegesippus” and the reception of Josephus in the early Middle Ages”, 69. 89 Craig Kallendorf, The Virgilian Tradition: Book History and the History of Reading in Early Modern Europe (Aldershot: Ashgate Publishing, 2007), 15.
27
is, however, no medieval codex, as it is dated between 1525 and 1550. At this time both the printed book and tables of content were well-known. Ms 1581 still has medieval features, as it does not refer to folium or page numbers, but merely describes what the chapters in the books are about.
With regards to tables of contents, the other major exception is Paris, Bibliothèque nationale de France, Latin 5051, dated 1400-1450. This manuscript is divided in books, not in chapters or paragraphs. Two books contain smaller initials indicating new parts of the text, but these do not follow the normal order or number of chapters. Only one book is preceded by a table of contents (book 17 on f.189v-190r). Despite the lack of chapters, tables of contents or even marginalia, the codex is still easily navigable because of the clean, light parchment, and the large initials at the beginning of each book. However, no other manuscript in the selection lacks a structure in textual hierarchy and contents like Latin 5051. The table of contents was actually widely used by medieval readers and therefore, it points to navigational paratext guiding the reader through the text.
The prologue could provide stronger evidence for medieval paratextual sensitivity as the majority of the selection has included the prologue by Josephus to readers of his own time. Thirteen manuscripts do not include the prologue, but six of these start with books thirteen, fifteen or seventeen and two manuscripts have lost pages at the front and in the back of the codex.90 Of the other five, it can only be confirmed of Berlin,
Staatsbibliothek zu Berlin, Lat. Fol. ms 263 that the prologue was never copied. Book one of this manuscript starts on folium verso and the preceding folio recto is empty, filled with possession marks and notes of later readers. In the other four cases, the first book
90 Respectively London, British Library, Arundel ms 94, 1125-1175; London, British Library, Royal ms 13 D
VII, 1120-1140; Milano, Biblioteca Ambrosiana, ms A 220 inf, 800-850, 56 fols., access May 26, 2020,
https://ambrosiana.comperio.it/opac/detail/view/ambro:catalog:26885; Paris, Bibliothèque nationale
de France, Latin ms 5045 (2), 1100-1125; Paris, Bibliothèque nationale de France, Latin ms 5763,
1000-1200, 68 fols., access May 26, 2020, https://gallica.bnf.fr/ark:/12148/btv1b8426038x; Paris,
Bibliothèque nationale de France, NAF ms 21013, 1400-1420 and Città del Vaticano, Biblioteca
Apostolica Vaticana, Archivio del Capitoli di San Pietro ms A.39, 1075-1125; Città del Vaticano, Biblioteca Apostolica Vaticana, Vaticani latini ms 1996, 1300-1400, 193 fols., access May 26, 2020,
28
starts on the folio recto, which could point to a lost prologue on a preceding leaf.91 Still,
this is not very likely, as in many manuscripts new books do not start on a new leaf or in a new column.
Moreover, all these four copies have their incipit before book one on the top left of the page. In most cases, when the prologue is separated from book one, the incipit immediately follows the explicit of the prologue, meaning that the incipit for book one would not be written on the first page of book one, but on the preceding page of the prologue. For Città del Vaticano, Biblioteca Apostolica Vaticana, Archivio del Capitoli di San Pietro ms A.37, this is different, as it starts not immediately with the incipit and text of book one, but with a table of contents for book one. Still, as the other books and their respective tables of contents immediately follow on the text of the prologue, it is unlikely that this book was originally preceded by a prologue. Later in time, a prologue was added on the folio verso before book one, as was also done for Wolfenbüttel, Herzog August Bibliothek, Guelf. ms 14 Blank. For all five copies without prologues at the start of book one, it seems likely their scribes simply did not have access to it.
The unavailability of texts may have resulted in another remarkable feature of the manuscript Antiquitates Iudaicae prologues as well. The manuscripts London, British Library, Royal 13 D VI (1120-1140) and Paris, Bibliothèque Mazarine, Bibliothèque Mazarine 1581 (1525-1550) have added, before Josephus’ own prologue, an excerpt of St Jerome on Flavius Josephus. Both codices were written for rich owners, as they were both decorated in multiple colours by professional rubricators. Because only these two manuscripts of the selection added St Jerome’s prologue, it seems that this text was not often combined with Antiquitates Iudaicae. Another explanation of their inclusion might
91 Città del Vaticano, Biblioteca Apostolica Vaticana, Archivio del Capitoli di San Pietro ms A.37,
1300-1400, 181 fols., access May 26, 2020, https://opac.vatlib.it/mss/detail/215204; Città del Vaticano,
Biblioteca Apostolica Vaticana, Palatini latini ms 814, 975-1025, 145 fols., access May 26, 2020,
https://opac.vatlib.it/mss/detail/3161; Cologny, Fondation Martin Bodmer, Bodmer ms 98, 800-850, 169 fols., access May 26, 2020, https://www.e-codices.ch/fr/list/one/fmb/cb-0098; Wolfenbüttel, Herzog
August Bibliothek, Guelf. ms 14 Blank., 1100-1150, 223 fols., access May 26, 2020, http://diglib.hab.de/mss/14-blank/start.htm?image=00003.
29
be that these books were written for a scholarly audience, readers who would know what St Jerome had written on Josephus and wanted it to be included. It could also be that inclusion of St Jerome’s excerpt justified reading Flavius Josephus. Jewish authors were generally not widely read. St Jerome, however, approved of Josephus, including him in his De viris illustribus. This allowed Christian readers to read Josephus. In this way, the inclusion of St Jerome’s text on Josephus as a prologue could have functioned as a sort of approbation and contextualisation of the text.92
Defining the prologues in Antiquitates Iudaicae as front matter paratextual elements is, however, a bit problematic. As it was custom in medieval manuscripts to include the table of contents before books, one would expect that the prologue always preceded the table of contents of book one. This is the case for seven manuscripts.93 In thirteen
cases, however, the table of contents for book one was not added at all.94 In four
92 Jerome, On Illustrious Men, ed. Thomas P. Halton (Washington: Catholic University of America Press,
1999), 28-29.
93 Città del Vaticano, Biblioteca Apostolica Vaticana, Urbinati latini ms 400, 1300-1400, 358 fols., access
May 26, 2020, https://opac.vatlib.it/mss/detail/3173; Città del Vaticano, Biblioteca Apostolica Vaticana, Vaticani latini ms 1995, 1400-1500, 152 fols., access May 26, 2020,
https://opac.vatlib.it/mss/detail/3169; Città del Vaticano, Biblioteca Apostolica Vaticana, Vaticani latini ms 7015, 1200-1300, 209 fols., access May 26, 2020, https://opac.vatlib.it/mss/detail/3172; Firenze,
Biblioteca Medicea Laurenziana, San Marco ms 385, 1100-1200, 247 fols., access May 26, 2020, http://mss.bmlonline.it/s.aspx?Id=AWOS4yilI1A4r7GxMdjs&c=S.%20IOSEPHI%20ANTIQUITATES%20IUD AICE#/book; London, British Library, Royal ms 13 D VI, 1120-1140, 218 fols., access May 26, 2020,
http://www.bl.uk/manuscripts/FullDisplay.aspx?ref=Royal_MS_13_D_VI; Paris, Bibliothèque Mazarine, Bibliothèque Mazarine ms 1581, 1525-1550, 399 fols., access May 26, 2020,
https://mazarinum.bibliotheque-mazarine.fr/ark:/61562/mz1824; Wolfenbüttel, Herzog August
Bibliothek, Guelf. ms 22 Weiss., 825-875, 249 fols., access May 26, 2020, http://diglib.hab.de/mss/22-weiss/start.htm?image=00010.
94 Città del Vaticano, Biblioteca Apostolica Vaticana, Ottoboniani latini ms 84, n.d.; Città del Vaticano, Biblioteca Apostolica Vaticana, Palatini latini ms 815, 1439; Città del Vaticano, Biblioteca Apostolica Vaticana, Reginensi latini ms 1935, 1400-1500, 391 fols., access May 26, 2020,
https://opac.vatlib.it/mss/detail/3177; Città del Vaticano, Biblioteca Apostolica Vaticana, Vaticani latini ms 1994, 1300-1400; Cologny, Fondation Martin Bodmer, Bodmer ms 181, 1475-1500, 308 fols., access May 26, 2020, https://www.e-codices.ch/fr/list/one/fmb/cb-0181; London, British Library, Harley ms 3883, 1175-1200, 306 fols., access May 26, 2020,
http://www.bl.uk/manuscripts/FullDisplay.aspx?ref=Harley_MS_3883; Oxford, Exeter College, Exeter College ms 25, 1175-1200, 130 fols., access May 26, 2020,
https://digital.bodleian.ox.ac.uk/inquire/p/1707829f-fdc9-4bfe-aa56-d390e71bc68a; Paris, Bibliothèque
nationale de France, Français ms 247, 1410-1420, 312 fols., access May 26, 2020,
30
manuscripts, the table of contents for book one preceded the prologue.95 These four
cases refute that Josephus’ prologue was perceived as (para)text separate from the main text, as is suggested by the proto-title page of San Marco 385 as well. Rather, for these scribes the prologue belonged fully to the main text of book one.
The thirteen manuscripts without a table of contents to introduce the first book can be interpreted in both ways. On the one hand, the absence of table of contents might mean that the scribes understood the prologue to be a special text. The prologue started the codices and therefore, no table of contents for book one was included. On the other hand, in this view Josephus’ prologue and book one were seen as equal parts of the text, not in a hierarchical order. Therefore, there is no clear conclusion on what scribes perceived as main text and paratext, or if they distinguished between the two. If scribes did have a sense of paratext, few visible elements point to a distinction between text and paratext.
Body matters
In thirteen manuscripts ranging in date from 825 to 1500 the text is divided into paragraphs.96 There are two manuscripts in which the start of every sentence was
6446, 1400-1415; Paris, Bibliothèque nationale de France, Latin ms 5045 (1), 1100-1125, 173 fols., access May 26, 2020, https://gallica.bnf.fr/ark:/12148/btv1b8454696n; Paris, Bibliothèque nationale de France, Latin ms 5051, 1400-1450, 230 fols., access May 26, 2020,
https://gallica.bnf.fr/ark:/12148/btv1b8446965g; Paris, Bibliothèque nationale de France, Latin ms 8959, 1100-1200; Paris, Bibliothèque nationale de France, Latin ms 14361, 1100-1200.
95 Berlin, Staatsbibliothek zu Berlin, Lat. Fol. ms 625, 1300-1400; Città del Vaticano, Biblioteca Apostolica Vaticana, Città del Vaticano, Biblioteca Apostolica Vaticana, Vaticani latini ms 8698, 1375-1425, 340
fols., access May 26, 2020, https://opac.vatlib.it/mss/detail/3175; Vaticani latini ms 1997, 1125-1175; Paris, Bibliothèque nationale de France, Latin ms 8835, 1461.
96 Berlin, Staatsbibliothek zu Berlin, Lat. Fol. ms 263, 1000-1100, 209 fols., access May 26, 2020,
http://resolver.staatsbibliothek-berlin.de/SBB0001949C00000000; Città del Vaticano, Biblioteca
Apostolica Vaticana, Archivio del Capitoli di San Pietro ms A.39, 1075-1125; Città del Vaticano, Biblioteca Apostolica Vaticana, Palatini latini ms 814, 975-1025; Città del Vaticano, Biblioteca Apostolica Vaticana,
31
rubricated.97 No chronological pattern can explain the use of paragraphs or lines as
paratextual elements. All the manuscripts make use of the twenty books and their chapters to divide the text. Thus, all copyists followed the original division of Josephus’ text despite its transformation from rolls to codex, for twenty books (i.e. rolls) fitted easily into one codex. They preserved the original structure of the text and adapted new textual elements to this hierarchy.98
Studying the use of capital letters in titles and incipits is more interesting. In twenty manuscripts, the scribe used capital letters to indicate the start of a new part, for example the first three words of book one ‘IN PRINCIPIO CREAUIT’.99 For the other
seventeen cases, the scribe wrote the title in lower-case letters, although most of the time rubricated incipits were added before the start of new text to clearly mark the end and beginning of a new chapter or book. The first occurrence of the use of lower-case letters in titles is dated at the end of the eleventh or the start of the twelfth century.
7015, 1200-1300; Cologny, Fondation Martin Bodmer, Bodmer ms 181, 1475-1500; London, British
Library, Royal ms 13 D VI and VII, 1120-1140; Paris, Bibliothèque nationale de France, Français ms 247,
1410-1420; Paris, Bibliothèque nationale de France, Français ms 6446, 1400-1415; Paris, Bibliothèque
nationale de France, Latin ms 8835, 1461; Paris, Bibliothèque nationale de France, NAF ms 21013,
1400-1420; Wolfenbüttel, Herzog August Bibliothek, Guelf. ms 22 Weiss., 825-875.
97 Berlin, Staatsbibliothek zu Berlin, Lat. Fol. ms 625, 1300-1400; Paris, Bibliothèque nationale de France,
Français ms 247, 1410-1420.
98 Anonymous, “De Oude geschiedenis van de Joden als overlever”, Beyond the Text: text and paratext from manuscript to print, access May 26, 2020,
https://oldbooksnewdirections.wordpress.com/2020/05/01/de-oude-geschiedenis-van-de-joden-als-overlever/.
99 Berlin, Staatsbibliothek zu Berlin, Lat. Fol. ms 263, 1000-1100; Città del Vaticano, Biblioteca Apostolica Vaticana, Ottoboniani latini ms 84, n.d.; Città del Vaticano, Biblioteca Apostolica Vaticana, Palatini latini
ms 814, 975-1025; Città del Vaticano, Biblioteca Apostolica Vaticana, Palatini latini ms 815, 1439; Città del Vaticano, Biblioteca Apostolica Vaticana, Reginensi latini ms 1935, 1400-1500; Città del Vaticano,
Biblioteca Apostolica Vaticana, Urbinati latini ms 400, 1300-1400; Città del Vaticano, Biblioteca
Apostolica Vaticana, Vaticani latini ms 7015, 1200-1300; Cologny, Fondation Martin Bodmer, Bodmer ms
98, 800-850; Firenze, Biblioteca Medicea Laurenziana, San Marco ms 385, 1100-1200; London, British
Library, Arundel ms 94, 1125-1175; London, British Library, Royal ms 13 D VI and VII, 1120-1140; Milano, Biblioteca Ambrosiana, ms A 220 inf, 800-850; Oxford, Exeter College, Exeter College ms 25, 1175-1200;
Paris, Bibliothèque nationale de France, Latin ms 5045 (1) and (2), 1100-1125; Paris, Bibliothèque
nationale de France, Latin ms 5763, 1000-1200; Paris, Bibliothèque nationale de France, Latin ms 8959,
1100-1200; Wolfenbüttel, Herzog August Bibliothek, Guelf. ms 14 Blank., 1100-1150; Wolfenbüttel,
32
Three other cases in which no capital letters for titles are used date from the twelfth century and all the other occurrences are dated in the fourteenth to sixteenth centuries. This change from capital letters in titles to lower-case letters could cohere with the shift of scripts in the late Middle Ages in which the capital letters were simply used less, or it may have been that the use of capital letters, or litterae notabiliores, for titles simply declined (illus. 7 and 8).100
In the selection of Antiquitates Iudaicae manuscripts there are three French manuscripts of which one is split into two codices, dated respectively 1400-1415, 1410-1420 and 1475-1500.101 Remarkably, all of these manuscripts and only the
sixteenth-century Paris, Bibliothèque Mazarine, Bibliothèque Mazarine 1581 manuscript start new books on new pages or columns (illus. 9). Although some of the Latin manuscripts have the first book written on a new page or column after the prologue, none of them create the extra space for other books, instead fitting the table of contents for the new book and its text directly after the explicit of the preceding book, whether this was at the top of the page, in the middle, or at the bottom. Translations became more common in the late Middle Ages, and especially the fifteenth century saw the rise of the vernacular in Western Europe.102 Therefore, it is no surprise that all French translations of Josephus’
Antiquitates Iudaicae in the selection are dated in this century or later. In the selection, however, there may be a bias as well, considering that the four texts are richly decorated and illuminated. All these copies include miniatures before new books, sometimes laid with gold. It can therefore be concluded that these were written for rich patrons. Parchment was expensive and was not to be wasted. Only rich readers would have had the money to spare to start each book on a new page. Therefore, the relation between
100 Carmassi, ‘Book Material, Production, and Use from the Point of View of the Paratext’, 314. 101 Cologny, Fondation Martin Bodmer, Bodmer ms 181, 1475-1500; Paris, Bibliothèque nationale de France, Français ms 247, 1410-1420; Paris, Bibliothèque nationale de France, Français ms 6446,
1400-1415; Paris, Bibliothèque nationale de France, NAF ms 21013, 1400-1420.
102 Denis Renevey and Christiania Whitehead, ‘Introduction’, in Lost in Translation? (Turnhout: Brepols
