Skip to content

The Codex and Early Christians: Clarification & Corrections

September 16, 2014

The prominence of the codex bookform in early Christianity is a well-known phenomenon.  But there remain continuing questions, and also what I regard as confusion on some matters that make for mischief in historical analysis.  I’ve treated the topic rather extensively in my book, The Earliest Christian Artifacts:  Manuscripts and Christian Origins (Eerdmans, 2006), 43-93.  Nevertheless, in light of a couple of subsequent publications, I think it necessary to reiterate and clarify a couple of things.

I start with reference to a recent essay:  Stanley E. Porter, “What Do We Know and How Do We Know It? Reconstructing Early Christianity From Its Manuscripts,” in Christian Origins and Greco-Roman Culture: Social and Literary Contexts for the New Testament, edited by Stanley E. Porter and Andrew Pitts (Leiden: Brill, 2012), 41-70.  At one point in this otherwise very helpful and impressively informed discussion, Porter deals briefly with the codex in early Christianity (p. 49).  Referring to Roger Bagnall (Early Christian Books in Egypt, Princeton University Press, 2009, esp. pp. 70-90), Porter claims (p. 49) that Bagnall showed that the rate of adoption/usage of the codex in early Christian circles was no greater than in the larger culture of the time (2nd-3rd centuries CE).  In a footnote (p. 49, n. 26) he claims that Bagnall “corrected” my own discussion, showing “that the Christian uptake of the codex was no greater than in secular literature.”  But I must protest and correct matters.

Granted, Bagnall referred to my discussion as a recent example of what he calls “partly misleading” statements of matters (p. 77).  But, as I pointed out in my review of Bagnall’s book (Review of Biblical Literature, 01/2010, available here), it is actually Bagnall who has given a bit of a misleading impression of things.  My own key positions are these:  (1) Christians of the 2nd/3rd centuries preferred the codex over the roll to a remarkable degree in comparison with the general book-culture of their time, and (2) this preference was especially strong for copies of those texts that they most highly prized and treated/read as scriptures.  The data behind these two positions are entirely clear and easily verified.  Moreover, Bagnall actually granted  and underscored these two points in his own discussion.

Bagnall offered figures (pp. 72-74) comparing the number of non-Christian and Christian codices from Egypt datable to the early centuries, giving also the percentages of Christian codices of the total.  His own data show, e.g., that Christian codices amount to somewhere between 22-34%  of the total for the 2nd-3rd centuries CE.  Yet Christian books overall amount to only ca. 2% of the total number of books (codices and rolls) of these centuries.  Of course, there are more non-Christian codices, but the first point to note is that Christian codices comprise a vastly disproportionate percentage of the total number of codices in this period.

So it is in fact not at all “misleading” to say so, and Bagnall has not “corrected” me on these matters.  Neither I nor others familiar with the data claim that Christians invented the codex or were the only ones to use this bookform.  Nor have I claimed that the more widespread preference for the codex in 4th century CE and thereafter was caused by Christianity (although a few others have suggested this).  But the very data provided by Bagnall clearly show that Christians invested in the codex far more than is reflected in the larger book-culture of the time.  That is, the early Christian preference for the codex is undeniable, and this preference is quite distinctive in that period.  And Bagnall actually reached the same judgement, stating “Christian books in these centuries [2nd/3rd] are far more likely to be codices than rolls, quite the reverse of what we find with classical literature.” (p. 74)

My second point also stands, and is agreed by Bagnall:  The early Christian preference for the codex seems to have been especially keen when it came to making copies of texts used as scripture (i.e., read in corporate worship).  E.g., some 95+% of Christian copies of OT writings are in codex form.  As for the writings that came to form the NT, they’re all in codex form except for a very few instances of NT writings copied on the back of a re-used roll (which were likely informal and personal copies made by/for readers who couldn’t afford a copy on unused writing material).  And here again, Bagnall actually grants the same conclusion, judging that “the Christians adopted the codex as the normative format of deliberately produced public copies of scriptural texts” (p. 78), but were ready to use rolls for other texts (76).

As to the reason(s) that Christian preferred the codex (especially for their scriptures), that’s a matter less clear or agreed.  Bagnall expressed an interest in Kurt Treu’s speculation that early Christians inherited use of the codex from Jewish scribal tradition, but this proposal suffers from a complete lack of any evidence of pre-Christian Jewish use of the codex for literary texts.  In fact, all pre-Christian Jewish literary texts are in bookroll form.

In a lengthy footnote (p. 49 n. 27), Porter offers several frequently made assertions about the supposed superiority of the codex over the bookroll, claiming that these were what attracted Christians to the codex.  But I’ve discussed these matters extensively (The Earliest Christian Artifacts, 61-83), and I remain unpersuaded.   To mention one counter-argument, if the codex was already such a supposedly superior bookform in the 2nd-3rd centuries, why didn’t everyone recognize this at the time?  Why did most everybody continue to prefer the bookroll?  Were the early Christians the only ones perceptive enough to see the supposed advantages of the codex?  That seems just a wee bit counter-intuitive.

So, to correct in this instance Porter, Bagnall didn’t really correct my own discussion of early Christian preference for the codex, but essentially wound up agreeing with it.  That Christian preference was unprecedented and unparalleled in the 2nd-3rd centuries CE.



From → Uncategorized

  1. I wonder if Christians were simply more likely to want to “flip around” to specific passages in their New Testaments (easier in the codex than scroll) than other secular literature which was usually only read start-to-finish.

    • Jews as well as Christians used their scriptures similarly, searching for specific texts. And pagans searched their classical literature. I don’t think your proposal works.

  2. Paul J permalink

    I realise I’m coming rather late to this discussion, but could the relatively low take-up of the codex by non-Christians be purely a matter of cultural conservatism despite its superiority as an information medium?

    • Paul: That might be one factor, but I think it was more complex than that. People knew very well the codex format, which at the time was generally regarded a more informal, workaday type of bookform, typically for manuals of magic or medicine, astronomical tables, etc., and the roll was prized as a much more appropriate and useful form for literature proper. So, the early Christian preference for the codex, and especially for their most prized literary texts, involved a deliberate difference.

  3. Dr. Gaddie,
    There would be no practical way anyone, Church, government, etc, could completely censor book production. Books in antiquity were “published” and circulated through social networks by popular demand. The modern equivalentfor ccomparison would be to make an attempt at gathering up all printed news papers of the last 200 years and destroy them. This would be physically and practically impossible to do, even with todays technology.

  4. Steve Ulrich permalink

    What do you think of the prevalence of censorship, the burning and destruction of roll, codices, manuscripts, etc during the 2nd/3rd centuries? How would this effect the “evidence” of codex vs roll? Or am I way off base?

    • Er, I have no idea what you’re referring to. There were some sporadic efforts to destroy Christian books in the time of Emperor Decius (c. 250 CE), and in subsequent persecutions of Christians. But I don’t see that this is relevant.

  5. As a graduate-trained scholar in culture studies and historical method, historiography, I’d like to note here that the survival of an apparently disproportionately high percentage of Christian codices, might not reflect original production figures – but survival figures.

    That is to say, Christianity eventually caught on, and came to dominate western cultures. And? These later Christian scholars might well have tended to preserve more Christian texts, over others. Since Christian texts were regarded as holy, and far more worth preserving than others.

    Thus, the higher percentage today of more ancient Christian than pagan texts, does not necessarily reflect original production figures.

    • Your argument might work for the time after Christianity exercised any effects and control, i.e., after the 4th century. But the MSS that survive from the 2nd and 3rd centuries come from a time when Christianity was persecuted, and certainly not able to suppress production or preservation of pagan texts. So your argument doesn’t work for that period.

      • Even if Christians were only in control of culture after say 400 AD, for the next 1600 years, they were the gatekeepers; who decided which earlier, ancient texts survived. And in effect, their preferences would have edited history.

        A Christian in 600 AD, or in 1400 AD, could look over a pile of old books from 200 AD; and could decide which old texts were kept and copied; and which were thrown away. Leaving us with an edited view of 200 AD.

        Thus more modern Christians controlled and changed History in effect. According to their own preferences.

      • Dr. Gaddie: As a person “trained in historical method” (your self-description), you’ll know that the first responsibility of someone who wants to comment on something with any credibility is to study the data. The Leuven database of ancient books, for example lists thousands of copies of ancient texts for the first few centuries CE, among which Christian books (esp. in the first three centuries) are a tiny minority. So, obviously, Christians didn’t destroy the mass of pagan literature from this period.
        Moreover, these manuscripts were found where they were put in the 2nd and 3rd centuries (e.g., Oxyrhynchus trash heaps of the time), Christian and non-Christian copies side by side. So, no evidence of selective and later destruction.
        But the point of the statistics I cited is that it’s clear that christians preferred the codex in a time when most others preferred the bookroll. The data make that unquestionably clear to anyone who has studied them.

  6. Tim Reichmuth permalink

    Dr. H.,

    Obviously, we cannot know how faithful Porter was in checking sources, i.e. actually reading your work as oppossed to relying on Bagnall’s comments, but it appears to me that way too often scholars accept what other scholars have said without checking the primary source themselves. Often this self perpetuates and soon we have an accepted fact that is not factual at all. I for one wish all scholars were as diligent about their own work as you. Those of us who do not have the expertise nor resources to followup on footnotes often assume such items to be accurate. The primary avenue for us to be ‘corrected’ is the originating author continuing to be involved in his/her work and being willing to do so in a public forum such as this.

    All of that to say, thanks!


    • I for one intend no accusation that Porter didn’t read my own material carefully. I simply note that his characterization of Bagnall’s discussion as “correcting” my own is incorrect.

      • jfjoyner3 permalink

        From memory: when I read Bagnall’s comments (a while back) I got the gist that he was challenging your analysis and conclusion. I do NOT recall Bagnall’s comments, quoted above, surely due to my own carelessness, but these excerpts seem at odds with the gist of the essay as I remembered it. Giving my memory the benefit of the doubt, there must have been a “tone” in Bagnall’s essay that is at odds with these quoted commments. Possibly Porter was left with an (inaccurate?) impression of Bagnall’s gist, as I was, from reading carelessly, as I did? Whatever his reason, I remember thinking about Bagnall’s essay, Hurtado cannot be so easily dismissed on this … and here (above) is validation of my confidence in your assessment! I have pulled your book from the shelf, and his, because my interest in this topic is renewed. Thank you for “correcting the record.”

Comments are closed.

%d bloggers like this: