The 16th and last satire breaks off mid-flow. The ending is lost, therefore, or perhaps was not written.
Ancient books were written on rolls. One modern author theorized that the end of a text ought to be safer than the start, because it should be inside the rolled up scroll. He seems to think that a roll would normally be stored ready to read.
But it seems to me, in my ignorance, that the reverse is the case. The average ancient reader would get to the end of his reading, and find his roll almost fully rolled-up. It is possible, of course, that some readers would then unwind the whole roll and roll it back up the correct way. But human nature being what it is, surely most of the time the reader will just pop the roll back in its cylindrical case. A reader who takes up a roll to read and finds it is back-to-front has an incentive to rewind it. A reader who wants his lunch has none.
I suggest, therefore, that as a rule most rolls were stored with the end hanging out. This would explain quite simply why so many ancient texts are mutilated at the end, without requirement for the hypothesis that they were written in codex form.
Fascinating insight. Quite analagous to the videotapes we all rented from video stores a decade or two ago. You pop it into your tape reader only to discover that that the previous viewer never bothered to rewind.
Pulling an old dusty copy off of a shelf somewhere, you’d be foolish to assume that your movie would start at the beginning.
A very good analogy — thank you!
This may have occasionally been the case (or perhaps in some societies I’m not aware of, the norm), but I believe the evidence points to the accepted practice being to “be kind, rewind”. Herculaneum being a library which “survives” from antiquity probably gives the best argument for this – when they first started hacking up the scrolls, they found that the less fragile center (“midollo”) of the scroll usually contained the end of a work. To protect against the kind of outer-damage you’re wary of, a blank or minimally-inscribed long sheet (“protokollon”) was used at the beginning of the text to be wound around it. There was also, however, often a “subscriptio” sheet at the very end giving the author and title (in addition to the sillybos tags), just in case the inconsiderate reader before you had not rewound, though to my knowledge the vast majority of rolls from Herculaneum are wound the “right” way.
This actually has important consequences for the reconstruction of texts from Herculaneum, due to the varied and fragmentary ways they have come down to us. One of the popular methods of reading involved cutting the scroll on two sides to remove the “midollo” for “unrolling”, leaving two (or more) sets of “scorza” – the scorched outside husks containing the beginning of the text. These were read by having Greek-illiterate artists make drawings of what they saw on a given layer, then scraping it away and repeating. Of course, these were numbered sequentially and between the drawing and the reading, this was forgotten. So until around the late 1980’s when Obbink and Delattre realized this independent of one another, readings made from these drawings (“disegni”) would be backwards, using the increasing number sequence (from inside to out) as the sequence of the text.
A good (inexpensive) introduction to Herculaneum and some of these aspects can be found in David Sider’s “The library of the Villa dei Papiri at Herculaneum”, but I don’t have my copy at hand. But in Google Books preview I also found Richard Janko’s article on the “Philodemus Translation Project” in “Proceedings of the 20th International Congress of Papyrologists, Copenhagen” pp. 370-371 to contain a good overview of this. Also Parkinson & Quirke’s “Papyrus” pp. 38-39 seems to indicate that something similar to the “protokollon” was also used in Egyptian rolls. They illustrate with a demotic example from BM EA 10380, showing that with a right-to-left script the blank was on the right – so it seems the same practice may have been used, but with a “counter-clockwise” instead of “clockwise” roll.
Of course, it could well be the case that after the explosion in popularity of the codex, the last of the scroll-readers did not particularly care or know about these old customs, and Herculaneum would hardly reflect this!
This is very interesting, and thank you for making it. I had not thought of the evidence available from Herculaneum.
Of course one could argue that the owner of that library probably had a slave to look after it, whose sole duty would be to reroll the scrolls. But then again, so would most libraries; and it would be libraries that would tend to furnish the exemplars of copies transmitted to us, rather than wild copies.
Thank you for those references — I will take a look.