BMCR 2007.06.41

Christianity and the Transformation of the Book: Origen, Eusebius, and the Library of Caesarea

, , Christianity and the Transformation of the Book: Origen, Eusebius, and the Library of Caesarea. Cambridge, MA: Harvard University Press, 2006. xvi, 367. $29.95.

Table of Contents

This book is jointly authored by Renaissance and early modern historian Anthony Grafton (Princeton) and Megan Williams, scholar of late antique history and religion (San Francisco State). It is the result of a collaboration that arose at Princeton during Williams’s time there as a graduate student in the Department of Religion (xi-xii). The stated goal of the book, announced first in the preface and reiterated throughout the volume, is to ascertain the connections between the Christian scholars Origen (c. AD 184-c. 254) and Eusebius (c. 260-339) as regards their investment in literary technologies, especially in their shared enthusiasm for multi-columned codices (xi-xii). That may seem a rather narrow focus, but its outworking is expansive. With forays into library history, papyrology, and reception studies, this book is multi-faceted, erudite, and inspiring in its scope. While summarizing much of the best scholarship, Grafton and Williams (G-W) offer some original conclusions, notably on the connection between biblical scholarship and historiography among Christian writers (esp. chapter 3). G-W limit their main arguments to the scholastic tendencies of Origen and Eusebius at the library of Caesarea Maritima, thereby emphasizing their chosen subtitle. As a result, the main title, “Christianity and the Transformation of the Book,” appears somewhat overstated. They only rarely seek to define “Christianity” by its interaction with “the Book”. Nevertheless, if you come to their study to learn about the appropriation of emergent book technologies by one specific late antique school, led by veritable “impressarios of the codex”, then you will be richly rewarded.

The book is organized into four chapters with introduction and “Coda”. Aware of the roadblocks for uninitiated readers, G-W include at the very beginning of the book a “Cast of Characters” (xiv-xvi), which lists the most important dramatis personae from their narrative with short biographical summaries. The first two chapters deal with Origen and the second two with Eusebius. According to the preface, Williams was primarily responsible for the first two chapters and Grafton the second two, with joint revision leading to the final product (xii-xiii).

The introduction begins with a brief meditation on the fame of the school of Caesarea in the Renaissance, then moves quickly into a survey of the bibliography and shape of the subject. G-W announce a theme in the introduction which is carried throughout the book: Origen’s Hexapla was a unique and audacious work of scholarship which had a profound impact on his immediate successors, especially Eusebius. In particular, the tabular form of Eusebius’ famous Chronicle owed its inspiration directly to the Hexapla. Precursors are also named in the introduction, some of whom appear later and others who do not. Julius Africanus, himself the author of a world chronicle, is the subject of further scrutiny in chapters 2 and 3. The Library of Alexandria is noted as a touchstone for ancient scholarship, via Africanus (20-21), but is hardly mentioned again in the course of the book (cf. 52, 67). While not an egregious omission considering the publication of Bernhard Neuschäfer’s, Origenes als Philologe (Basel, 1987), Origen’s debt to Alexandrian scholarship is largely not a concern of this book.

The first chapter, “Origen at Caesarea: A Christian Philosopher among his Books,” situates Origen among ancient philosophers as regards his bibliographic mode of life, not his philosophy (25), neatly avoiding the perennial debate over Origen’s Platonism.1 G-W compare the scholarly culture of Caesarea with that of the Plotinian school in Rome, which was roughly contemporary with Origen’s career (28-41).2 Shared traits include 1) the accumulation of authoritative works in a personal library (58), 2) the collation and correction of texts in a scriptorium environment (ibid.), and 3) the dependence on patrons for the ongoing purchase and copying of books (55-56). G-W also compare Origen’s interest in books to the library of the Villa dei Papiri in Herculaneum, destroyed by the Vesuvius eruption of AD 79 (46-53). This comparison is based on a synthesis of secondary literature and does not make original arguments about the Villa itself.3 However, the comparison is salutary especially in the phenomenon of intratextual variation between different manuscripts of literary and philosophical works. The import for biblical studies is immediately apparent. Equally effective would have been a comparison with the Dead Sea Scrolls at Qumran, which also exhibit intratextual variation, but the use of the Villa dei Papiri (and the school of Plotinus) firmly links Origen to practices of book collection and collation closer to his own day and imperial context.

The chapter closes with a meditation on Origen’s role as master of the Caesarean library: in his teaching, in his book-collecting and collating, and in the church hierarchy (69-85). The transition from Alexandria to Caesarea included Origen selling his library (of ancient classics) in a fit of worldly renunciation (67). Of course, at Caesarea his patron (the layman Ambrose) helped him construct a new one, not least through the gift of numerous scribes.4 Ultimately, G-W conclude, Origen was both connected to the intellectual currents of his day and also unique. One of his most characteristic features was an interest in Hebrew (82). Such an interest cut against the grain of eastern Christianity, which had adopted the Greek Septuagint as the inspired Word of God. Origen also believed it was inspired, but ultimately his linguistic curiosity produced the most audacious book of his career and the most famous product of the Caesarean tradition.

Chapter 2 is entitled “Origen’s Hexapla: Scholarship, Culture, and Power”. G-W begin this chapter with a description of the Hexapla itself, fully aware that any reconstruction is open for debate (88-89). They then proceed to the evidence, weighing the testimonia and the fragments in succession (89-102). The Hexapla (lit., “six-fold”) was a six-columned book across the opening of a codex page which contained the Old Testament in both Hebrew and Greek. The first column was a Hebrew text, a rabbinic version (different from that underlying the Septuagint) which had likely been acquired from the local Jewish community of either Alexandria or Caesarea. The second column was a Greek transliteration from the Hebrew. The third and fourth columns presented Aquila’s and Symmachus’ more literal Greek translations, based on later Hebrew texts than that of the Septuagint, but similar to the Hebrew of the first column. Then came the Septuagint itself, possibly adorned with marginal scholarly apparatus, the asterisk and obelus, borrowed from Alexandrian scholarship.5 This text also included block Hebrew letters for the name of God. Finally, the sixth text was Theodotion’s Greek translation, which was also based on a similar Vorlage to the Hebrew in the first column but which was less Semiticizing in character. (Aquila, Symmachus, and Theodotion are collectively known as the recentiores or simply “the Three”.) Some books of the Bible, especially those in verse, were represented in three other (anonymous) translations, the so-called Quinta, Sexta, and Septima. No evidence survives for how these three were integrated on the pages of the Hexapla. In addition, it appears likely that the books of the Pentateuch included a Samaritan version (169). Again, there is no evidence for what this looked like on the page.

G-W assess each surviving testimonium for this model: Eusebius ( HE 6.16), Jerome ( Comm. in epist. ad Titum 3.9), Epiphanius of Salamis ( Pan. 64.3.5; De mens. et pond. 510-535), and Rufinus of Aquileia ( HE 6.16.4). These authors are all rather early and each had a vested interest in the literary history of Origen and his Hexapla. Among the many notable details of these reports is Epiphanius’ claim that the whole point of the Hexapla was to prove correct the text of the Septuagint, for which reason (he says) Origen placed it “in the middle”(93-94).6 Adam Kamesar has likewise emphasized Origen’s commitment to the Septuagint in his magnificent study of the tradition of Jerome’s Bible.7

The two surviving fragments of the Hexapla (Cairo Genizah and Ambrosiana) answer certain questions while at the same time provoking more difficulties concerning the original format (96-102). Their textual traditions are formally unrelated, yet they share important features: 1) both are palimpsests, 2) both are texts of the Psalms, 3) both include the Greek transliteration, Aquila, Symmachus, the Septuagint, and Theodotion, and 4) both normally deal with just one Hebrew word per line, each presenting forty lines per page. Nevertheless, there are important differences: 1) the Genizah fragment is significantly shorter than the Ambrosiana; and 2) the Ambrosiana seems never to have had a Hebrew column. Overall, the testimonia and fragments provide a very incomplete picture of the Hexaplan text, but they do connect the Hexapla to trends in late antique book culture (102-132). The most significant trend is certainly the exploitation of the codex form to its furthest extent: the Hexapla comprised 40 codices of 800 pages each, requiring the incredible sum of c. 150,000 denarii for both the copying and the parchment of one edition (103-106, based on Diocletian’s price edict).

The rest of chapter 2 concerns the feasibility of contemporary speculations on Origen’s project.8 For instance, how much rabbinic influence was there on the formation of the Hexapla? In particular, who wrote the Greek transliteration of the Hebrew and for what purpose? If it was Jewish in origin, then perhaps Origen began his work on the Hexapla while in Alexandria. More likely, according to γ the work began in Caesarea, where there were thriving Jewish and Samaritan communities (111). Finally, the thorny issue of Origen’s own knowledge of Hebrew is raised. In the end, G-W conclude that the Hexapla makes little sense if Origen did not know any Hebrew at all (112-113), but many questions remain concerning his sources and inspiration for this massive book. It remains unclear, for example, how all this effort Origen expended on the Hebrew connects to his preference for the Septuagint: does he merely want to show the Septuagint to be philologically preeminent (93-94), or is he using the Hexapla as an apologetic tool against the Jews themselves, who he says had corrupted the biblical text after the Septuagint was made (121; 127-128)? G-W are right to conclude that these issues make the Hexapla “more, rather than less, perplexing” (122).

Chapter 3, “Eusebius’ Chronicle : History Made Visible”, begins the second half of the book, which concerns the immediate reception of Origen’s work in the oeuvre of Eusebius. Eusebius, G-W claim, went beyond Origen in the application of book technologies, notably in his use of the multi-columned format for historiography, as well as in the advance of institutional, even imperial, patronage (133). Like Origen, Eusebius was a biblical commentator, a role which manifested itself in many of his scholarly endeavors, not least the Chronicle (c. 300). The Chronicle made it possible, in G-W’s words, “to fix a whole world on paper” (136), by aligning data from various strands of biblical and Near Eastern historiography. The first modern historian to suggest the influence of the Hexapla on the Chronicle was Timothy D. Barnes, whom G-W laud for this crucial observation (142).9 The Chronicle is divided into two parts, the Chronography and the Canons. The former is a tabular list of synchronisms of Greek, Roman, and Jewish history; the Canons, which is in many ways more interesting, is a systematic chronicle of world history and follows nineteen ancient states down through time, culminating in one column representing the Roman empire (141-142). In the preparation of the Chronicle Eusebius could rely on Julius Africanus, though in several ways he disagreed with his predecessor (148-168). Among these disagreements were 1) Eusebius’ rejection of the idea that a Christian could use the tools of chronography to predict the Second Coming; and 2) his decision to include information on the multi-millennial histories of Egypt (Manetho) and Bablyon (Berossus) even when it could not be convincingly reconciled with the biblical text (152-168). Commenting on the Chronography, G-W write, “[Eusebius] resisted all temptations to exclude discordant ingredients from the rich crazy salad of his first book” (167). This discordant blend of historical material was, according to γ suggested to him by Origen’s Hexapla. The value of the Hexapla for Eusebius is that it trained him to read parallel texts “word by word”, comparing them closely and allowing the discrepancies to remain (169-170). The rest of the chapter is taken up with comparanda for the Canons, from historiographical tables mentioned by Herodotus and Thucydides, to the Parian Chronicle, and the Liber annalis of T. Pomponius Atticus (170-177). This survey is more evocative than source-critical and (notably) does not include a discussion of Greek and Roman calendrical tables, which could have served (even indirectly) to inspire later chronography.10

The fourth and final chapter, “Eusebius at Caesarea: A Christian Impresario of the Codex”, concerns Caesarean editorial work on the Bible and begins with a survey of what we know about Eusebius’ mentor, Pamphilus (178-194). G-W conclude on the basis of recent scholarship that there was no institutional library passed down from Origen to Pamphilus (179-180). Rather, Pamphilus had a personal devotion to Origen and, like his master, a penchant for book-collecting and collating (182). Some of the most fascinating vignettes in the book are from biblical colophons that G-W quote in this chapter: these colophons name Pamphilus as corrector and annotator of the autographs (184-185). Eusebius continued the editorial tradition of his mentor while at the same time experimenting with literary technologies. Once again, G-W see Eusebius’ biblical tools, the Gospel Canon and Psalm Tables as dependent on Origen’s experimental Hexapla, and as the culmination of three generations of Caesarean biblical scholarship (195-208).

But Eusebius’ book technologies were not his work alone. First of all, he depended on a massive personal library, built-up by Pamphilus, and expanded and maintained by himself and collaborators.11 Secondly, Eusebius employed a number of scribes and assistants, just as Origen had —though we can imagine that by Eusebius’ day their number had swollen considerably. Eusebius also visited archives in the region, such as at Jerusalem (210), as well as perhaps at Edessa ( HE 1.13.5). He also drew up lists of books, pinakes, according to individual authors. One of these, of Philo, survives from an eleventh-century manuscript in Vienna (211).

All of these supplementary factors demonstrate that, as G-W note, “Eusebius’ workplace must have become a substantial research institution, at once an archive, a library, and a scriptorium” (215). Eusebius and Caesarea eventually received imperial support in the form of Constantine’s request (after 335) for fifty ornamental bibles to be prepared for Constantinople (216-221). It is possible that Codex Vaticanus and Sinaiticus both came from this order (220-221). The Onomasticon and Life of Constantine offer further proof of Eusebius’ bookish and documentary manner and prove the effect of the Caesarean library on the emergence of new Christian genres (221-225). The conclusion to chapter 4 returns to previously cited models for Eusebius’ overarching endeavor, namely the editorial work of Porphyry on the Plotinian corpus. And Justinian’s Digest, one of the few (ancient) post-fourth-century examples in the book, is also cited for comparison, particularly in its mode of compilation (228).

The book ends with a “Coda”, which, after briefly summarizing the conclusions of its chapters, looks forward to the reception of Caesarean scholarship in the works of Jerome (235-237).12 Here, G-W also consider the Christian Hebraism and scholarly paleography of the Renaissance. Erasmus is shown to have anticipated G-W’s argument that the Hexapla was seminal in the history of Christian biblical scholarship (238-239). The Chronicle too presaged the early modern trend in ecclesiastical history towards collection and compilation. And the connection between new forms of learning and technical advances in book technology is further illustrated by early modern print culture and its adoption among both Protestant and Catholic scholars (242-243).


a) Cognition

G-W are sensitive to the fact that they are considering questions of immense importance for the history of the organization of knowledge: how did the coming of Christianity change access to the word on the page, and vice-versa? Their title, Christianity and the Transformation of the Book, attests to this awareness. Nevertheless, the implications of this title are not clearly defined. In particular, what is missing is a sense of how a Christian reader’s interaction with knowledge would have been cognitively changed by Origen and Eusebius’ scholarly contributions. The cognitive element is implied in their title but never explicitly discussed.

A comparable volume, though very different in goal and scope, is James J. O’Donnell’s Avatars of the Word: From Papyrus to Cyberspace (Cambridge, Mass., 1998).13 The remit of O’Donnell’s book, signified by its subtitle, is to explore the book culture of the ancient world in light of our evolving knowledge-technologies today (or in 1998). This approach has the benefit of recognizing the essential continuity of human interaction with the written word.14 By contrast, G-W only rarely acknowledge the universality of cognitive issues, such as reader response, memory, and cognitive adaptability (cf. 6). On one hand, what has been so well elucidated in G-W’s work is the emergence of the Reference Book in codex form, yet the question of how a reader would go about engaging such a book (i.e., via discrete chunks of knowledge in somewhat random order) is not sufficiently addressed (cf. 211). O’Donnell’s remit is laudatory in this regard, if only because it is so easy (for any scholar) to get bogged down in the details of paleography and book production. O’Donnell stays true to the conviction that Late Antiquity is largely characterized (throughout the oikoumene) by the appropriation and reception of these new book technologies.

The phrase “mise-en-page”, repeated often throughout G-W’s book, seems to stand in for the cognitive question (e.g., 214). But arrangement on the page is only half the phenomenon. What about usability? The Suda retains an evocative entry about the Justinianic writer Stephanos of Byzantium and his alphabetical Ethnika. This massive book (in 60 volumes) was epitomated down to one handy volume (perhaps even in Stephanos’ lifetime) by his student Hermolaos.15 This example speaks both to the grandiose plans of these polygraphoi and to the physical difficulties of using their massive compilations. The alphabetization further highlights the desire (endemic to the period) to reference a text at random. Late antique scholars went to great lengths to make knowledge readily accessible. While G-W briefly mention the transmission of law codes and Ptolemaic geography — both important late antique comparanda (135-136, 228-230) — they make no concerted effort to elucidate how Origen and Eusebius’ innovations were connected to modes of reading and cognitive interaction in the broader Christian realm.

b) Personnel

Important as it is to emphasize the financial and physical help Origen enjoyed in the production of the Hexapla (e.g., 106), G-W do not point out that the scribes and assistants whom Origen had at his disposal were probably slaves. A letter from Libanius, writing in the generation after Eusebius, makes this plain: “[your scribe] looked quite pale, so I asked him if he was sick. He said no, but that the endless work was to blame. Constant writing had depleted him. I praised him for this and am delighted for you, that your slave ( oiketes) is not lazy!”16 Ausonius commented on a runaway slave whom he recaptured, “he’s as slow a runner as he is a writer,” and Gregory of Nazianzus freed his slave notarius in his will.17 There was no ecclesiastical scribal class in the third and early fourth centuries, at least none that compares to what we associate with the Middle Ages or which was prevalent in Jewish antiquity.18 Some indication by G-W that the production of forty hefty volumes of the Hexapla was likely enmeshed with the oppressive side of Roman society would have been salutary. Further, if we accept their thesis that Origen knew only a little Hebrew and required “assistants” to perform the collation and copying (112), it is possible even to imagine that these assistants were actually Jewish slaves who knew enough Hebrew for the copying work. The triumph of Christian scholarship in the fourth century begins to lose some of its sheen when we try to reckon with the manpower (potentially coerced) that these enormous and complex endeavors required.

c) Bilingualism

The phenomenon of bilingualism is a major aspect of this book. The two languages at issue for Caesarean scholarship are Greek and Hebrew. G-W repeatedly call Hebrew a “barbarian” language (80-81), presumably from the point of view of Greek speakers in the Roman empire. Eusebius says that Origen “learned the Hebrew tongue thoroughly” ( HE 6.16), and G-W comment on this passage that Eusebius “can probably be trusted” (89). However, elsewhere G-W claim that Origen’s Hebrew knowledge was minimal and he “needed considerable assistance” in accessing the Hebrew Bible (112). In Adam Kamesar’s view, the Hexapla was not really about the Hebrew at all; it was about the Septuagint.19 The question that Origen’s late antique successors debated was not whether Origen knew his Hebrew but whether he preferred the original “pure” Septuagint or a “corrected” (Hexaplaric) Septuagint. The latter served to take account of what was missing or different in the Hebrew, variances which were perhaps indicated by the (admittedly redundant) asterisk and obelus in the margins of the Hexapla’s fifth column.

G-W end up concluding that the Hexapla was not made for correcting the Septuagint but rather was part of a “suite of scholarly tools” prepared over decades, including a self-standing Hexaplaric Septuagint (114-118). This stance serves to further the argument that the Hexapla was more about the Hebrew than not (a la Pierre Nautin). However, one still has to wrestle with Origen’s exegesis, and, as with other Christian scholars in the period, there is no systematic application of the Hebrew. A different Eusebius, that of Emesa (c.300-359), provides a useful test case, being the only writer prior to Jerome who seems truly to have given exegetical priority to the Hebrew text ( ho hebraios or to hebraikon, as he calls it).20 However, recent research has shown that Eusebius’ personal, direct knowledge of Hebrew was very limited.21 Like many writers from Edessa (see below), Eusebius was certainly bilingual in Greek and Syriac — not an insignificant fact — but his engagement with the Hebrew appears dependent, like Origen’s, on local informants, as well as on what he could glean from comparing the Septuagint and Syriac translations.22

The big picture easily gets lost in these debates. As with the question of cognition above, a broader view of late antique bilingualism is needed. The most obvious precursor to Caesarean bilingualism in the East is the intellectual world of Edessa in the late second and early third centuries AD. Greek and Syriac were mutually used and understood, and writings from Edessa in this period all share this bilingual flair: the works of Bardaisan, for instance, or the elegant Odes of Solomon seem to have emerged in both Greek and Syriac simultaneously.23 Biblical scholarship was a part of this mix as well. The “Assyrian” Tatian, a disciple of Justin Martyr who was comfortable in both Syriac and Greek (and possibly Latin), constructed his Diatessaron (c. 160) on precisely the opposite principle of Origen’s Hexapla and Eusebius’ Gospel Canon. Instead of seeking out variants between texts, he wove the four Gospels into a single seamless narrative.24 This was a major undertaking and had an influence on Bibles throughout the Christian world. The Syriac writer Ephrem produced at Edessa an important commentary on the Diatessaron in the fourth century.25 Furthermore, the history of Syriac translations of the Bible provides a crucial parallel with Caesarea. The Old Testament Peshitta (1st-3rd cent. AD) was made directly from the Hebrew (partly by Jewish translators). Later Syriac translations, however, show a marked preference for the Greek Septuagint — this is the proper milieu of the Syro-Hexapla (cf. G-W 95-96).26 Thus, the central conundrums of Caesarea were undeniably shared by Christian scholars very close in region and time. Not only was Edessa one of the first centers of Christian learning in the Mediterranean (before Caesarea), but the specific elements of bilingualism, exegesis, translation, and experimentation with form were constituent parts of their intellectual legacy.27

d) Conclusion

Despite these reservations about the absence of late antique contextualization, Christianity and the Transformation of the Book performs with aplomb the Herculean task of narrating (in plain language) a story rife with lost texts and missing information. G-W relate the oft-told tale of Origen and Eusebius in a new and accessible way by connecting it to a range of library practices in the ancient world. They also demonstrate the unique contribution of the multi-columned codex for posterity, highlighting the importance of book design for the transmission of disparate streams of ancient knowledge. Their book will benefit scholars and students alike as an introduction to this critical phase in the Christianization of the Roman empire, a phase which included the founding of Christian academic habitus and the proliferation of Christian knowledge. Origen and Eusebius thus paved the way for literary phenomena as varied as Jerome’s Vulgate, Bede’s Ecclesiastical History, and the preservation of Aristotle in Syriac and Arabic. Christianity and the Transformation of the Book offers a sustained and dependable account of Caesarean scholarship and its contribution to intellectual history writ large.


1. On this, see now Mark Julian Edwards, Origen Against Plato (Aldershot, 2002).

2. This section is explicitly dependent on Gregory Snyder, Teachers and Texts in the Ancient World: Philosophers, Jews, and Christians (New York, 2000). See also Mark (Julian) Edwards, Culture and Philosophy in the Age of Plotinus (London, 2006).

3. The inclusion of an archaeological plan of the Villa on p. 47 beautifies the volume but is not referred to in the text of the book.

4. See the Observations below.

5. See Neuschäfer, pp. 86-103.

6. It might be observed that there is no center text in the Hexapla, since there is an equal number of three columns per side. In this passage ( De mens. et pond. 510-535) Epiphanius claims that, in a later version, there were actually eight texts in the book (the Octapla): the two Hebrew columns (incl. the transliteration) followed by six Greek versions (the two additional, anonymous fifth and sixth translations marked simply with a Greek number). Whether Hexapla or Octapla, the Septuagint can never be the center text, even when an Octapla brings it to the position just to the right of the gutter. What is most significant in Epiphanius’ testimonium is the fact that the arrangement of the Hexapla has immediate cognitive implications for its reader. These implications are largely not drawn out by G-W. The most noticeable effect of the Hexapla is to de-center the text, which is precisely the discomfort which Epiphanius describes: “Now some while reading these books … might think that Aquila translated [chronologically] first, then Symmachus, then the Seventy-Two, according to the order in which they are placed, which is not the case. Rather, Origen, considering the edition of the seventy-two to be accurate, placed it in the middle, so that he might refute the translations on either side,” (trans. γ p. 319, slightly adjusted).

7. Adam Kamesar, Jerome, Greek Scholarship, and the Hebrew Bible: A Study of the Quaestiones Hebraicae in Genesim (Oxford, 1993).

8. This section is deeply indebted to Ruth Clements, ” Peri Pascha : Passover and the Displacement of Jewish Interpretation within Origen’s Exegesis,” Th.D. dissertation, Harvard Divinity School, 1997.

9. Timothy D. Barnes, Constantine and Eusebius (Cambridge, Mass., 1981), p. 120.

10. See Robert Hannah, Greek and Roman Calendars: Constructions of Time in the Classical World (London, 2005).

11. Here G-W are dependent on the work of Andrew Carriker, The Library of Eusebius of Caesarea (Leiden, 2003).

12. On this, see also Megan Hale Williams, The Monk and the Book: Jerome and the Making of Christian Scholarship (Chicago, 2007).

13. See Anthony Grafton’s concerned review, “Error Messages: Night Thoughts Inspired by James O’Donnell’s Avatars of the Word“, Boundary 2 28 (2001) 195-209.

14. O’Donnell, chapter 2. See also David R. Olson, The World on Paper: The Conceptual and Cognitive Implications of Writing and Reading (Cambridge, 1994). See now the collection entitled The Early Christian Book, William E. Klingshirn and Linda Safran, eds. (Washington, DC, 2007).

15. Suda, s.v. “Hermolaos” (no. 3048, ed. Adler 2.416). See also Aubrey Diller, “The Tradition of Stephanus Byzantius,” TAPA 69 (1938) 333-348.

16. Libanius Epistulae 131 (10.132, ed. Foerster); translated and contextualized by Kyle Harper, “Slavery in the Late Ancient Mediterranean”, Ph.D. dissertation, Harvard University, 2007, pp. 132-133.

17. Ausonius Epigrammata 14 (199, ed. Schenkl); see Harper, 132. For Gregory, see J.B. Pitra, Iuris Ecclesiastici Graecorum Historia et Monumenta (Rome, 1868) 2.157; see H.C. Teitler, Notarii and Exceptores: An Inquiry into the Role and Significance of Shorthand Writers in the Imperial and Ecclesiastical Bureaucracy of the Roman Empire, from the Early Principate to c. 450 AD (Amsterdam, 1985), p. 91.

18. See, generally, Teitler, pp. 86-94, who notes that slave notarii still fetched high prices under Justinian. On the Jewish scribal class, see the new book by Karel van der Toorn, Scribal Culture and the Making of the Hebrew Bible (Cambridge, Mass., 2007).

19. Kamesar, 10 and esp. 20: “Origen makes use of the Hebrew text not for its own sake, but for the sake of the LXX.”

20. Kamesar, 38-39, 126-158.

21. See R.B. ter Haar Romeny, A Syrian in Greek Dress: The Use of Greek, Hebrew, and Syriac Biblical Texts in Eusebius of Emesa’s Commentary on Genesis (Leuven, 1997).

22. Eusebius of Emesa certainly does not use the Hexaplaric version of the Septuagint; see ter Haar Romeny, Eusebius of Emesa, pp. 34-47.

23. See the venerable article by Han J.W. Drijvers, “East of Antioch: Forces and Structures in the Development of Early Syriac Theology”, in idem, East of Antioch: Studies in Early Syriac Christianity (London, 1984), sec. 1.

24. Since the publication in 1935 of a mid 3rd-century fragment in Greek, the consensus has been that the Diatessaron was originally composed in Greek, though, in the words of Drijvers, “what does ‘original’ mean in a thoroughly bilingual situation?” (“East of Antioch”, p. 7).

25. See Carmel McCarthy, Saint Ephrem’s Commentary on Tatian’s Diatessaron (Oxford, 1993) and W.L. Petersen, Tatian’s Diatessaron: Its Creation, Dissemination, Significance, and History in Scholarship (Leiden, 1994).

26. See in general Sebastian Brock, The Bible in Syriac Tradition (2nd rev. ed., Piscataway, NJ, 2006) and for the Hellenizing tendency, idem, “Aspects of Translation Technique in Antiquity” and “Syriac Attitudes to Greek Learning” both in idem, Syriac Perspectives on Late Antiquity (London, 1984), secs. 3 and 5.

27. What is more, the Caesarean legacy has a direct connection to Edessa through Julius Africanus, who met Bardaisan while on an imperial visit with Septimius Severus in AD 195. He admired primarily his skill at archery. See Africanus Cestorum Fragmenta 1.20.28 (183, ed. Vieillefond); and Han J.W. Drijvers, Bardaisan of Edessa (Assen, 1966), 167.