Early Christians & the Codex: A Correction/Clarification
An academic pal pointed me today to an essay published a couple of years ago (but I don’t read everything) by Alan Jacobs: “Christianity and the Future of the Book,” which appeared online in The New Atlantis, accessible here. I can’t comment on Jacobs’ reflections on modern technology, but I think it appropriate to correct his musings about the early Christian preference for the codex, as that’s a topic on which I’ve spent a good deal of time.
Jacobs refers to my book, The Earliest Christian Artifacts: Manuscripts and Christian Origins (Eerdmans, 2006), in which I devote a large chapter to the topic (pp. 43-93). Authors are always grateful for such notices, of course. But he seems to have ignored (or failed to grasp) a good deal of what I wrote there.
Jacobs posits four factors as disposing early Christians to prefer the codex: economy, portability, “integrity,” and “sequentiality.” But his proposal rests on a seriously faulty grasp of the relevant historical data. So, I shall elucidate them briefly.
First, as shown many years ago by T.C. Skeat, the putative savings of writing a text on a codex are not all that much, and (as Jacobs notes) with others I’ve pointed to the wide margins, wide line-spacing, and large lettering of most early Christian codices as indicating that their copyists weren’t particularly concerned to save on writing material. Moreover, the time and skills required to construct even a simple codex were additional to those required for a roll. In the latter case, the material came from the shop all ready (sheets of papyrus already glued together to form a running length of material), and all that was required was the skill to write a text in nice columns. For a codex, however, one had to learn a new means of estimating how much writing material would be needed, how to attach the sheets of writing material to one another, et alia.
Second, the only references to portability of codices in antiquity concern miniature codices (e.g., in the epigrams of the Roman poet Martial). But we also have examples of miniature rolls too, likewise intended for portability. So, small rolls could be just as portable as small codices, and there’s no firm indication that this was really a major factor in the preference for regular/larger-size codices among early Christians.
As to alleged “integrity”, by which jacobs seems to mean holding in one book physically various texts such as those that form a canon of scriptures, again, he misses the matter, and by a few centuries. He refers to the great 4th-century codices (e.g., Sinaiticus), which allowed a complete Bible of OT and NT writings in one set of covers. But in the initial period of Christian codex-preference (which goes back at least to the early 2nd century CE), the technology of codex-construction seems not yet developed to permit anything along these lines. The initial Christian codices seem to have been mainly (or perhaps all) single-text books: e.g., a codex of the Gospel of John or the Gospel of Matthew. Our earliest clear examples of multi-text Christian codices come from the early/mid-3rd century CE (e.g., the Pauline codex P46, and P45, which includes the four Gospels and Acts, and P75, which contained at least the Gospels of Luke and John). These are not likely the very first such experiments, of course; but they seem still to have been somewhat exceptional.
Indeed, well into the 3rd century CE it appears that Christians were still experimenting with how to make codices of sufficient size to accommodate multiple texts. The Chester Beatty papyri (housed in Dublin) include examples of various contruction techniques: “single-gathering” (numerous sheets stitched together in a single group), folded sheets stitched together concertina-style, and small numbers of sheets stitched together and then these in turn stitched together (the “multiple-gathering” codex form). The last technique proved best in the end, and modern leaf-books continue to be made this way (in what are traditionally called “quires”).
So “integrity” (i.e., accommodating multiple texts) likewise seems not to have been a strong factor in the initial preference for the codex by Christians.
And the same goes for “sequentiality” (by which Jacobs means that the codex permitted fixing an order of texts). I repeat: The initial Christian codices tended to be single-text books. So, no “sequentiality” of texts there. Moreover, even after Christians developed codex technology sufficiently to accommodate multiple texts, we have evidence of different sequences preferred (e.g., several different orderings of the four Gospels).
I thought I had made these matters sufficiently clear in my book, but it appears less successfully than I had hoped. As I judged in that discussion, the supposed “practical” reasons for Christian preferring the codex just don’t fit the data. And were the Christians the only people in the 2nd century to realize the supposedly “obvious” advantages of the codex? Just a bit counter-intuitive, yes?
Finally, there is one further interesting fact: The early Christians’ preference for the codex seems to have been markedly stronger with respect to texts that they treated as scriptures (e.g., OT texts, and those that came to form the NT), whereas they were somewhat less concerned about using rolls for other texts. This, it seems to me, indicates that their preference for the codex was expressive of a desire to mark their own books, especially their copies of scripture-texts, physically, distinguishing them and identifying them as Christian books. (I’ve referred to the Christian preference for the codex as signalling an emergent early Christian “material culture.”)
I’ve given a boat-load of the data for these matters in the relevant chapter of my Artifacts book. Whatever the future of the familiar leaf-book in these days of e-books/readers (and I personally hope that it has a future), the early Christians seem to have been ahead of their time in preferring the leaf-codex, when pretty much everyone else regarded the roll as a much-superior bookform for literary texts. Indeed, because they were ahead of their culture (and may have influenced it in this matter), they had to experiment themselves with developing the humble leaf-codex to serve the much more ambitious purposes that they (uniquely at the time) wished to pursue: e.g., using codices for large bodies of text and multiple texts.