2011.11.09 | Roger S. Bagnall, Early Christian Books in Egypt. Princeton: Princeton University Press, 2009. Pp. xv + 109. ISBN: 9780691140261 (Cloth).

Reviewed by Dan Batovici, University of St Andrews.

RBECS would like to thank Princeton University Press for kindly providing us with a review copy. You can find RBECS on facebook, here.

A shorter version is now published in Sacra Scripta 9.2 (2011).

This is a stirring small volume from a prominent papyrologist, containing the published form of four lectures offered in May 2006 at the École Practique des Haute Études of Paris, which were published simultaneously in French (with Droz, see here). The four lectures form now the four chapters of the book. The notes of each chapter are rather few in number and placed at the end of the volume, which is also equipped with a select bibliography (pp. 99-104), a subject index (pp. 105-9), and an index of the papyrological texts discussed (p. 110).

1. The first chapter, The dating of the earliest Christian books in Egypt,  (as indeed the rest of the book) brings to the fore the state of research on the earliest Christianity, of which much is deemed rather narrow and “profoundly at odds with fundamental social realities of the ancient world and with basic probability” (p. 1). In this scholarship, who has been “for more than a century obsessed with questions about the nature of early Christianity, of Christian origins” (p. 2), a central piece is the Egyptian strand of Christianity, mainly because only there have the papyri survived (p. 3).

As a result, a pressure emerged “to produce usable information prom the papyri” (p. 4), both literary and documentary, in the context of the scarcity of sources on the pre-Demetrian Alexandria – Demetrius was bishop there between 189-231 – given that without the early dating of some Christian papyri, “we have no contemporary witnesses to pre-Demetrian Christianity” (p. 5). Since the documentary papyri, official records or private contracts do not reflect at all any Christianity in the 2nd century Egypt, private letters “that could be dated before the Tetrarchy (295-305)” (p. 8)have been sought, but with not much success (pp. 8-9).

The literary papyri are hence of critical importance, being “in fact, the last hope of finding a pre-Demetrian Christianity outside Alexandria” (p. 9). The lack of other evidence from the period before Demetrios has as result a correspondingly higher expectation from the literary papyri (p. 9). Yet, if there are not any such Christian literature manuscripts in Oxyrhynchus, Arsinoe, and Hermopolis, “we may be entitled to suspect, with Jakab, that the widespread presence of Christianity into the countryside before Demetrios is an illusion” (p. 9). The problem, in this case, lies with the issue of dating the papyri.

Bagnall then moves to the dating of the earliest Christian papyri, as they appear in the LDAB (p. 11): one papyrus dated to the late 1st, early 2nd century, and other six dated to the 2nd century. For six out of these seven, Bagnall shows that there are largely two groups of scholars, of which one (including Bell, Roberts, Skeat) would date them earlier (“clearly” 2^nd  century), while the other (including Turner, Thomas, but also Grenfell and Hunt) would date them to the 2nd, 3rd century (pp. 11-5). This is entirely a bibliographical discussion; one might miss a palaeographical treatment of those manuscripts.

After presenting this discrepancy, Bagnall sets out to answer the question of how many Christian manuscripts should we expect to have from the 2nd century. His answer is a probabilistic one, building upon a model proposed by Rodney Stark (The Rise of Christianity: A Sociologist Reconsiders History, Princeton: 1996) for assessing the rate of the Christian population growth, in which the latter works with the assumption that there were only a thousand in the whole empire in 40 CE and that they grew to an overwhelming majority by the late 4th century. Bagnall then uses the conventional number of 55 million for the population of the empire, of which the Egypt conventionally “amounted to about a tenth” (p. 19). With these dates, he computes that by the time Demetrios became a bishop in 189 CE “there were probably fewer than twenty thousand Christians in all of Egypt” (pp. 19-20), somewhere between 0.17 and 0.35 percent of Egypt’s population.

Adducing the assumption that the share of Christian papyri from the total number of surviving papyri – “fragments of books of all types” (p. 20) – should be directly proportional with the percent of Christian population from the total Egypt population, Bagnall calculates the number of probable Christian books in three time slices: 1st/2nd century, 2nd century, and 2nd/3rd century. The result is that we shouldn’t have any papyri from the 1st/2nd century – 0.056 probability, “only one chance in 18” (p. 21) –, only one or two from the 2nd (1.360 probability), and 12 in the 2nd/3rd. Bagnall rightfully notices that the number of Christian papyri is significantly higher than this estimation (p. 21).

This estimation is ingenious and offers a real feel and a very helpful heads-up on the matter, at least with regard to our expectations. At all points in the simulation, Bagnall strives to show that, if anything, these estimations are in favour of the number of Christians. He is most likely right to notice that “it is time to let go of the idea that Christian literature is somehow underrepresented in the papyri before the late third century” (p. 21). One thing Bagnall does not say explicitly is whether or not this estimation should be used as an a priori criterion in dating early Christian papyri. Many would quibble, with regard to these estimations, that statistics tend to work when high numbers are involved, and to be less relevant when the numbers are low, as is the case here at the heart of the argument.

2. The second chapter, Two case studies, offers two examples for the problem presented only in a general manner in the first chapter, that is, the “forces, both academic and religious, that have made the study and particularly the palaeographic dating of early Christian books in Egypt such a difficult and even distorted subject” (p. 25).

a\ The case of the late Carsten Peter Thiede is pretty straightforward and quite telling: the book that followed his 1995 Zeitschrift für Papyrologie und Epigraphik article reveals by itself an agenda underlining his dating in the 1st century of the Papyrus Magdalen Greek 17. The case is presented caustically – at times a presented more as a case of psychology of scholarship than one of “the sociology of scholarship” as announced on p. 26. Not only the popularising book is stripped of its transparent agenda, but also Thiede’s article in ZPE on the matter is exposed as rather mimicking proper scholarship: its arguments only “attempted to give the impression” of following accepted scholarly procedures (p. 27).

Bagnall’s presentation of this case is compelling, though one might wonder why the ZPE editors published the article (they might have missed the agenda, but they seem to at least have had accepted the validity of the arguments which Bagnall, in his turn, presents as academically weak arguments and altogether bad scholarship).

b\ The second case study, by contrast, deals with “good, even excellent, scholarship” (p. 40), focused on the Shepherd of Hermas, more precisely on the dating of the P. Iand. I 4. “But the issues raised by the material are nonetheless full of troubling and difficult aspects concerning the earliest stratum of Christian books” (p. 40). The first editor, who thought it a medical text, dated the manuscript to the 4th century, and so did one of the two scholars who independently identified it as a fragment of Hermas. Yet other scholars have argued for a significantly earlier 2nd century date.

Bagnall’s discussion is centred mainly on an article of Antonio Carlini from 1992, in which the latter presents a correspondence on this topic among several scholars, illustrating a diversity of opinion on the matter. The aim is to reconcile an early dating of P. Iand. I 4 with the testimony of the Muratorian Fragment who seems to date the Shepherd of Hermas between 142 and 155. This dating would allow for very little time for a realistic diffusion of the work in Egypt (from Rome).

Bagnall then presents Carlini’s solution to this dilemma – namely that parts of the Shepherd might have circulated independently, so that some of them might have had reasonable time to reach Egypt, before the publication of the complete work at the time mentioned by the Muratorian Fragment – and deems it methodologically “at best somewhat awkward” (p. 47). He further contends that “it will be obvious that this hypothesis requires us in a sense to interpret the Muratorian Canon tendentiously, in order to save the palaeographic date” (p.47).

Some may find this very interpretation a bit tendentious: the theories of successive redactional stages in the composition of the Shepherd of Hermas certainly did not start with Carlini’s article, nor have they appeared provoked by the dilemmas of dating the Hermas papyri. Having said that, the two century discrepancy between the two datings begs the question of whether there were non-palaeographical criteria or agenda involved in the matter.

3. The 3rd chapter, The economics of book production, is surely among the most interesting ones. The first topic discussed here is the cost of the book in antiquity; the very few sources mentioning book costs are revisited. Bagnall notes that even if these sources may be for various reasons inexact, their “internal consistency is impressive” (p. 51). These literary sources also not inconsistent with the book costs found in the Theban West Bank ostraka (p. 52). Further, the issue of the relative cost of papyrus and parchment is tackled: the parchment (pp. 52-5), the papyrus (p. 55), the cost of copying which would have been similar for the two media (pp. 55-7), with the results synthesised on a table on p. 57, subsequently discussed.

Bagnall then turns to the “question whether monastic labour lowered the cost of book production and gave monasteries an advantage in becoming dominant centers of book production” (p. 59). To put it in his words, “because the potential cost savings of monastic production have become a standard topic in such discussions, we need to understand as well as possible what these savings might have been” (p. 50). A distinction is proposed between the work the monks did for their own library and their work “on commission for external individuals or institutions” (p. 60), and for the latter Bagnall points out that we have “no evidence that monasteries actually charged less than anyone else” (p. 60).

Next issue tackled is what incomes would have allowed one to buy books. Assuming that many costumers would be Christian institutions, Bagnall present the scarce evidence we have about their revenues (pp. 60-6), in order to see who would have been able to buy books. Hence, for an αναγνώστης, a reader in the church, a complete bible would have cost him “half a year’s income”; a priest without a large private income would have had an institutional income multiplied by only 2.5, so a complete bible would be “a major expenditure”, yet “a copy of the gospels alone would still have amounted to a large sum, but it was not unthinkable.” Bagnall continues: “it is not really until we get to the level of bishops that it is possible to imagine the regular buying of books as an affordable activity” (all these quotes are from p. 62).

Given that this reasoning is based on sources from the 6th century, the situation in previous centuries could only have been worse (p. 63). As a result, “the inherent probability of finding many Christian books truly datable to the second century is very low” (p. 65). Bagnall then explains the larger than expected number – by external reasons – of manuscripts surviving from the end of the 2nd, beginning of the 3rd century by proposing an explicitly speculative model in which, along with the changes in the Egyptian administration in the severian period, favoured the emergence of hellenized Egyptians, mainly bilingual, “rich enough to make it into the councils” (p. 69), who could have afforded ordering and buying books even before the creation “of well-financed church institutions” (p. 69).

This is a very interesting and very informative chapter; its considerations are very important for our general understanding of book production in context in early Christian Egypt.

4. The last chapter deals with similarly interesting and important topic, namely The spread of the codex. Bagnall notes, with respect to the relation between Christianity and the codex, that although it is usually accepted that the former is not connected in any way with the origins of the latter, the acceptance of the codex is still related to the Christianity.

Compiling the data offered by the Leuven Database of Ancient Books, he shows that the percentage of Christian codices is constantly under forty percent of all codices for 7 temporal slices up to the 4th century. He further notes that “the Christians adopted the codex as the normative format of deliberately produced public copies of scriptural texts, but they did not generalize from this adoption to broader use for all books” (p. 78).

The current explanations for the question why Christians used the codex for their scriptures are presented (pp. 79-80) and refuted (pp. 80-1). The model proposed by Bagnall is based on the work of William Johnson, “who has shown that there is a close relationship between the format of the elegant Greek book-roll and the cognitive process by which these rolls were used. […] Johnson argues that this process is linked to the normal use of such rolls for performative oral reading in elite circles” (p. 82).

In that case, the rather narrow width of the book-roll column is precisely what “the eye is capable of taking in at a singular ocular fixation and processing before moving to a new point” (p. 82). As such, the move from the book-roll to the codex with its generally wider column is rather a cognitive paradigm-shift in the process of reading.

The Theban magical papyri are then offered as a controlled experiment to observe the shift from book-roll to codex (p. 83). In Bagnall’s words: “The shift in the magical texts from roll to codex marks not only the passage of time, but also the privatization of magic, the acquisition of the books by an individual or family, and the entry of the treatises on magic into the world of Greek book” (p. 86). Furthermore, “it is hard to suppose that those responsible for them switching from roll to codex because of the influence of Christianity” (p. 86).

The Roman source of the codex is then mentioned (pp. 86-7), and Bagnall suggests that “the wider, if gradual, adoption of the codex elsewhere is yet another manifestation of what for short we may still call Romanization” (p. 87).

5. The poignant tone of the polemic, especially in the first two chapters, produces a perhaps unusual presence of the author in his book, and this may well prove bewildering for the younger scholar, who is usually bound to be neutral in tone, although I’m sure it might be a treat for the experienced one with a taste for a spiced-up academic read.

This very informative volume offers, by all means, anything but a tedious read. The reader is urged to question the interference of non-academic criteria and agendas into the issue of dating the earliest Christian papyri, as well as in our general understanding of their testimony for those centuries. All in all, this surely is an unavoidable read for any scholar working on Early Christianity with an interest in the early Christian manuscripts.

Dan Batovici
University of St Andrews
db47 [ at ] st-andrews.ac.uk