Listno.3,writtenontheversoofP.Oxy. 2660, which is a Greek-to-Latin
glossary dating from the first or second century, was carefully prepared,
arranged, and copied, as can be seen in figure 10.1. It gives every impression
of being a formal booklist from a library of some size. The writers mentioned
are all authors of comedy, beginning with Amipsias, whose name must
be supplied at the upper left, then continuing down to Aristophanes,
and after a lacuna resuming at the top of column 2 with Archippus and
continuing down to Epicharmus. Authors and titles are arranged in single-
letter alphabetical order, and the text is written clearly and carefully.Itseems
reasonable to assume that the list continued through the alphabet, providing
similar coverage for other comic writers and their works, and on that as-
sumption it is very unlikely that this was a list ofdesiderataor of assigned
readings: it is simply too long and complete for that. Nor is it likely to be a
scholar’s bibliography on, say, the subject of comedy, because it omits some
well-known works, among them theWomen of LemnosandPelargoiof
Aristophanes, and the title given specifically asPlutus Ashows that the
compiler was well aware that there was aPlutus Bas well, yet he omits
that title from his list.
23
In addition, there appear to be some repetitions
among the plays of Epicharmus in column 2: theHarpagaiat lines 15 and 17,
Dionysoiat lines 21 and 22, andEpiniciusat lines 24 and 25. The repetition of
these titles could be due to scribal error,^24 but they may well represent
duplicate copies of works in the collection.^25 Taken together, the possible
duplicates, the omissions, and the careful presentation of the text all argue
for this being the surviving part of an actual library booklist.
If we now look at our no. 3 as a list of the books in a library,^26 we are
struck first of all by how little information it provides. We find authors and
titles only: no dates (of composition, say, or of when this copy was made,
- These omissions, as well as that of theIchthuesof Archippus, were all noted by
John Rea, who edited the text forThe Oxyrhynchus PapyriXXXIII. He concluded that
the fragment is quite probably ‘‘the catalogue of some provincial library or a reading list.’’
I believe its length rules out the latter possibility. - Rea,ad P.Oxy. 2659, took the first pair as due to scribal error. Only the first six or
seven letters of each title survive, and Rea accordingly assumed that the second titles in the
second and third pairs were different, and previously unattested, works. The matter is quite
uncertain. It is unsatisfying to explain a problem by assuming a previously unattested work,
but on the other hand it is disconcerting to find these duplicate copies, if that is what they are,
all concentrated within the space of ten lines in a list that otherwise has no repetitions at all. - Otranto 2000, 37, took them as duplicates. We know from the Villa of the Papyri at
Herculaneum that libraries could and sometimes did have duplicate copies: Gigante 1979,
59, provides a list of the works present in the Villa collection in two or more copies. - There is no way to know if this was purely a private collection or a ‘‘public’’ library.
Nor is there likely to be much difference: any owner of books was likely to make his
collection, or specific books in it, available to his friends and to scholars. We see this both in
Egypt (P.Oxy. 2192 Otranto 11, cf. the discussion by Johnson forthcoming;P.GettyMus.
acc. 76.AI.57 Otranto 4; andP.Mil.Vogliano11 Otranto 5) and in Italy, where Lucullus
threw open his library to his friends and to Greek scholars (Plut.Luc. 42.1 2). That is, any
such library could be made available, though privately owned, to much of the reading public.
Papyrological Evidence for Book Collections and Libraries 241