Some of you may recall my recent post where I mention my recent article that briefly disputes Brent Nongbri’s case for P18’s greater likelihood of being a codex. Following this, Brent concocted a twofold response, which is as eloquent as it is amicable. In Part 1, he engages with Charlesworth’s response published last year and in Part 2 with mine. I won’t be writing a rejoinder and would encourage our readers to read Brent’s posts for themselves. He summarises the points of contact very accurately and, since both his and my piece were written in more tentative terms, the debate pretty much ends up in weighing the probabilities. I for one remain unconvinced by the counterproposal, but, once again, I am impressed by how Brent is able to interact with opposing views with civility and fairness. (Another good example of this is Brent’s latest post concerning Ryan Kaufman’s counterproposal concerning the ending of John 20 in P66.)I might note, however, that I was quite intrigued by (parts of) Steven Goranson's comment under Brent second post quite interesting. Let me quote it in full:
I am undecided whether it is a roll or a codex, and excuse me if I missed something, but concerning the paragraph ‘Yes, but the specific point at issue is “fragments of what were once more extensive rolls” that preserve no more or less than a single column of writing on both sides. How common is that? (It’s an honest question–I don’t know the answer.) If our Exodus/Revelation papyrus is indeed a portion of a reused roll, it would seem to be a very happy coincidence indeed that this surviving portion of the roll preserves exactly a single column of text on each side.’ I would comment that if one column from the *middle* of a roll were preserved on the recto, then the chances of a nearly-matching column on the verso would be smaller than the chances of a match if indeed this piece were a roll end. That it holds the end of Exodus and the beginning of Revelation (assuming the top edge, with room for initial Rev. verses, is missing) might suggest (but not prove) the end of a roll. After all, if using a similar margin, one side of the column already aligns as a given (unlike in a mid-roll scenario), so if similar column width was used, the match of columns may not occasion much surprise. Also, though I may be on thin ice, let me go further: aren’t Exodus and Revelation both scroll-prominent books, and scribes knew that? So, given a choice…
Although I’m not sure I’d call Revelation a ‘scroll-prominent book’ (it does involve a comparatively higher proportion of papyri, the numbers are so small that it’s very hard to make any convincing generalisations), I think the main force of Goranson’s argument lies in the fact that the beginning followed by the ending of another work in a fragment like this would speak in favour of a re-used roll. To this I might add that portions of the roll closer to the centre (i.e. ending of the → side) are more likely to get preserved than outer parts—for obvious reasons. This would fit nicely with the situation in P18, provided that the roll wasn’t rolled up the opposite way upon re-use. Here I must confess that I don’t really know how this was done or whether there are any studies that deal with this. I would thus gladly echo Brent’s observation that ‘we really do need a thorough survey of reused rolls’. I couldn’t agree more. How’s that for a doctoral thesis topic?