Over on his blog, Larry Hurtado has posted a link to his list of “Christian Literary Texts in Manuscripts of Second & Third Centuries.” When he first published a version of this list in his 2006 book, The Earliest Christian Artifacts, it provided a much needed succinct listing of the earliest surviving Christian texts and made it a lot easier to think about this entire corpus of material. That said, there are some features of the list that are worth pondering further. Whenever we try to distil information in a list like this, we make all kinds of choices that have benefits and drawbacks (for my own forthcoming book, I had to think through a number of these issues, so I’m especially attuned to these complexities right now). None of what follows here will be news to Larry, I suspect. It’s just a matter of points we choose to emphasize differently.
Texts vs. Artifacts
Larry is clear that his inventory is a list of texts and not artifacts. My own preference is to think in terms of complete manuscripts rather than isolated texts, especially when we’re talking about numbers. An inattentive reader of Larry’s list might get the mistaken impression that we have 257 Christian manuscripts surviving from the second and third centuries. So, a list of artifacts would look a little different (i.e., considerably shorter). To take just a couple examples, items 105, 106, 109, 131, and 132, are all part of a single book, Codex I of the Chester Beatty Biblical Papyri (P45). So also, Codex II of the Beatty Biblical Papyri (P46) accounts for 9 individual items in the list (139, 145, 147, 148, 151, 152, 153, 155, and 161).
“Christian” and “Jewish”
When these papyri started turning up in relatively large numbers in the late nineteenth century and early twentieth century, it was already difficult to identify some of the Greek copies of the Hebrew scriptures as either “Christian” or “Jewish.” This task has only become harder in recent years as we have come to appreciate the variety and complexity of these ancient identities. So, the “Jewish” or “Christian” character of some of these items is really challenging to determine with any certainty. There are, however, a number of items on the list that almost no scholar would designate as Christian, namely, the Qumran material (items 14, 19, 20, 25, 31, and 84) and pieces clearly datable to the period prior to the first century CE (such as item 28). Again, I hasten to add that Larry is well aware of this, makes the point explicitly in the introduction to his list, and clearly identifies the almost certainly “Jewish” pieces in his list with an asterisk *. I only want to emphasize that had an alternative decision been made (i.e. just leaving such items out of a list titled “Christian Texts”), the list would look different. Again, for one thing, it would be shorter.
The fluctuation of dates
As Larry notes in his aforementioned book, The Earliest Christian Artifacts (p. 16), palaeography is an imprecise undertaking. Thus, almost none of the items in the list have a date that we are really certain about. And scholarly opinion is always changing. So, for instance, item 245 (P.Oxy. 60.4009), a papyrus fragment of an unknown gospel from Oxyrhynchus appears on Larry’s list with a date of “AD2.” This is a fair statement of scholarly opinion. But, a visitor to the entry for this piece on Leuven Database today will find that it has now been assigned by the reputable palaeographer Pasquale Orsini to the fourth century, rather than the second. Because of the subjective nature palaeographic decisions, these dates are always subject to change, sometimes dramatically.
The ranges of proposed dates
For most of the items in the list, the date range given is representative of the breadth of scholarly opinion. For some items, though, that range is actually quite a bit wider. So, for instance, items 83, 87, and 171 are all part of what we now call Schøyen MS 193, a small codex of Coptic literature. Larry’s list represents the date of this codex as AD3. In fact, a spectrum of dates has been proposed: late second or early third century (C.H. Roberts), third or fourth century (Eric Turner), fifth or sixth century (tentatively by Tito Orlandi). All these dates are reported in J.E. Goehring, The Crosby-Schøyen Codex MS 193 (Louvain: Peeters, 1993), p. xxxiii. So also item 175 (P.Ant. 1.12), a parchment fragment containing 2 John. On the list, it is dated “AD3?” The range of opinion is broader: “not much later than the middle of the [third] century” (C.H. Roberts, in the editio princeps) to “first half or middle of fifth century” (Cavallo and Maehler, in their palaeographic handbook).
The bottom line is that a list like this is a good starting point, but it’s not the last word (nor would Larry claim that it is). Gathering the data gives us a basis for further discussion in an effort to reach the most reasonable conclusions.
Variant Readings 
Pingback: My List of Second/Third Century Manuscripts | Larry Hurtado's Blog
Brent, why do you think there are no NT manuscripts listed in Larry’s book from before the second /third century?
Well, almost no (reputable) scholar would place any of our known manuscripts containing NT texts in the first century. Occasionally such claims are made (I wrote about one here: https://brentnongbri.com/2017/07/21/a-first-century-papyrus-of-mark-probably-not-the-one-you-think/ ), but for the most part, nobody takes such claims seriously.
Brent, you haven’t answered my question. The reason I asked it, is because the evidence (the lack of first century manuscripts or copies) points to the church fathers of Caesarea being the manufacturers of the Christian religion.
Most scholars would agree that our lack of surviving manuscripts from the first century doesn’t indicate that such manuscripts never existed. Rather, the manuscripts we have reflect the growth and geographic spread of Christian communities in the third and fourth centuries. Again, I would refer you to Bagnall’s Early Christian Books in Egypt for more detailed argumentation.
Brent, it seems very strange to me that there are no manuscripts from the first century and then suddenly a large number appear in the second/third century. One would have thought that statistically, there would be a gradual transition. Do you have an explanation?
We don’t actually know with certainty how many second and third century Christian manuscripts we have. There are actually very few pieces that can be dated on non-subjective grounds to the period before Constantine. There is a good discussion of the problems in Roger Bagnall’s Early Christian Books in Egypt.
Brent, what you wrote “There are actually very few pieces that can be dated on non-subjective grounds to the period before Constantine”, undermines the whole of Christianity as we understand it. Was it, for example, the work of the early Church Fathers? The same applies to early Roman history or the works of Josephus, as we understand it. The surviving manuscripts are late (from the fifth century and later) and invariably they are preserved by a monk in some monastery.
Hi Geoff, When I say there are few pieces that can be dated on non-subjective grounds to the period before Constantine, I don’t exclude the probability that _some_ of the manuscripts usually assigned to that period on more subjective grounds were actually produced then. My point is that it’s sometimes very hard to distinguish, based on handwriting alone, a manuscript produced in the third century from a manuscript produced in the fourth century.
Brent, I remember when I first heard Bagnall present on what is in that book at an SBL meeting some years ago (didn’t you present in that session too?), that some pieces do not fit well in his model. For example, how does Marcion’s relatively large enterprise fit in the mid-second century if the transmission of the writings in question was so marginal as he seem to think.
Hi Tommy, It’s not so much that I think the _transmission_ of the writings from the second century was minimal as I think the survival of the writings from that period may be less than we sometimes think, and thus the writings that have survived perhaps not quite as representative of what was circulating in the second century. To stick with your example, we seem not to have any writings of Marcion (despite his seemingly wide popularity) other than fragments passed down in the writings of the patristic authors.
Brent, do you think you should use an expression like “most scholars” in an argument?
Sure, it’s simply stating a fact. Among scholars, it’s an uncontroversial statement to say that the lack of surviving manuscripts of a given work in the century (or centuries) after its composition doesn’t mean that such manuscripts never existed. How different such assumed manuscripts may be from those that have survived is a different question.
Brent, it is not scholarly to say that because there are no manuscripts related to the NT from the first century, there must have been manuscripts in existence from which the second/third century manuscripts were derived. How do they explain the total absence of any first century manuscripts? The best that scholars can say is there may have been manuscripts in existence. I have to ask which scholars are you talking about? Importantly what is their background? Are they Christians for example? Have they considered alternatives such as is Christianity a complete fabrication of the early Church Fathers? Eusebius of Caesarea seems to have convinced the emperor Constantine of the rightness of Christianity, frequently quoting from the writings attributed to Josephus. Yet we know that those writings were written with access to Vespasian’s and Titus’s memoirs. Professor Helen Bond of Edinburgh University has written that she believes that the Testimoniun Flavium was a Christian interpolation. I believe that Antiquities 18 is an edited document, and that it was initially written by a Jew, probably Agrippa I. The editing was a cover story about Jesus that obscured another story about Caiaphas a rebel priest.
re: The fluctuation of dates: I agree there is a problem with the way paleogrpahers present their arguments and conclusions, although I would say the problem lies in the intuitive nature of these conclusions, not subjectivity per se. The basis for the conclusions in, say, the P.Oxy. series, are based on the collective experience of the editors, having dealt with many hundreds of dated / datable documents from the period in question. It is difficult to distill experience of this sort into an argument that would fit into the self-consciously terse style of the P.Oxy series, let alone illustrate it. The few parallels the editors allow themselves in the ed.pr. is exemplary, can’t really be definitive. Orsini’s conclusion in LDAB is hard to accept because because the argument is entirely implicit, and depends on a theory of seriation of the esthetic analysis of graphic features that is itself intuitive. WIth regard to P.Oxy. 60.4009 in particular, the hand would not stand out in a tray of 100 second century (non-cursive) documents. In a similar tray of third century (non-cursive) documents, it would be more probelmatic, but in the fourth century, it would stick out like a sore thumb.
When I first saw Orsini’s assignment of P.Oxy. 60.4009 to the fourth century, I was surprised, too. But once my eyes were open to the possibility that it might be that late, I started to see some reasonably good comparanda from that era (e.g. Nag Hammadi Codex IX).
I would feel a lot better about Orsini’s dates if he said what in particular about e.g. P.Oxy. 4009 is comparable to mid-late fourth hand. First, there is very little about P.Oxy. 4009 that appears stylized (i mean, different in form from what is normal in documents, except for eta in three strokes rather than two. And nothing in it would be out of place in a second century doc. Fourth century docs. written like this (non-cursively) look very different. I find a comparison with Nag Hammadi IX difficult ebcause it is so very stylized, and the pen width so different.
I would feel happier about a fourth century attribution if there a continuous chain of docs throughout the fourth written in a similar style, which is difficult for the third, but impossible for the fourth. But this is what comes of approaching the problem in the way I am, docs. > Lit. I am also beginning to feel the Alexandrian Stylistic Class theory, since it looks backward from a Byzantine type for its predecessors, inevitably will draw date ranges higher.
Brent, Larry lists (page 217 of Earliest Christian Artifacts) two references (91 and 92) to manuscript fragments of Philo. Do you know if these fragments (apparently from the third century) are the earliest that we have for Philo?
Yes, these manuscripts are generally assigned to the third century, but even if they were as late as the fourth century (which is possible), they would still be the earliest manuscripts we have of the works of Philo.
Brent, thank you. I think, for various literary reasons, that the Philo manuscripts referred to above, could have been altered or interfered with. Why is it that many of the manuscripts listed by Larry are dated to the second and third centuries, and as you say, they may be as late as the fourth century? These dates would have given ample opportunity for changes or editorials to the texts.