or of acquisition), no bibliographical information whatever. There is no
catalogue number of any sort, no reference to which cabinet the roll was
stored in. This is true of our other lists as well, and of the survivingsillyboi
(the tags or labels that were attached to papyrus rolls to identify them), on
which we regularly find author and title, but never any indication of where
the book was to be shelved or stored.^27
All of this is consistent with what we know of the organization of
Roman libraries generally. There is no evidence for an ancient equivalent
of the modern call number or catalogue.^28 Rather, as previous scholars
have argued, ancient booklists were very probably organized first by
literary genre (epic, say, or lyric) or, where literary genres did not apply,
by subject (medicine, astronomy), and then alphabetically by author.
29
Presumably this system derives ultimately from categories defined by
Aristotle and employed in the library at Alexandria, and it probably also
reflects the physical organization of the actual book collections them-
selves.
30
All of this is now generally accepted.
31
What has not been
And, on the other hand, any ‘‘public’’ library in a provincial town would almost certainly be
a gift from an individual to the town, like those in Ephesus and Timgad, and open to
virtually the same group of readers as a ‘‘private’’ library in that town would be. Libraries
in gymnasia might conceivably have had restricted access, but we have no clear evidence
on such libraries and cannot even prove they existed in Egypt. See Funghi and Messeri
Savorelli 1992a, 59 61, for a general discussion, and cf. van Minnen 1998, 106 8, on the
question of whether we can assume that high quality texts might have come from gymnasia.
- Onsillyboi, see Dorandi 1984, and add Stephens 1985, and Hanson 2004, 209 19.
Dorandi’s no. 1 can serve as an example. It reads in its entirety, ‘‘Hermarchus,Against
Empedocles, Book 9.’’ - Probably the closest we come is a statement in theHistoria Augusta. The author,
here called Flavius Vopiscus, claims that his reader could find a certain book in the ‘‘sixth
bookcase’’ (in armario sexto) in the Ulpian library at Rome (HA Tac. 8.1). This has been
taken (e.g., by Blanck 1992, 218) as an indication that the bookcases were numbered
implying a kind of elementary cataloguing system but it is not convincing evidence. This
whole passage in theHAis not to be trusted (Paschoud 2002, 276 7), and the author of this
section of theHAis quite willing to invent documents and bogus sources, even ones that
could readily be checked, so that we cannot even assume that he was striving for verisim
ilitude (Chastagnol 1994, cxxi cxxii). Given that no knownsillybosindicates the bookcase in
which the book was shelved, we are entitled to doubt that such a cataloguing system existed.
It should be noted, however, that the number of knownsillyboiis very small, and the
discovery of even a single example with a bookcase number could change the picture we
have substantially. - For a useful summary of the thesis, the ancient evidence on which it is based, and
references to some of the secondary literature, see Otranto 2000, XII XV. - Otranto 2000, XVI XVII, with a useful distinction between the booklist of the
library and thePinakescompiled by Callimachus. ThePinakes, it appears, did not include all
of the books in the Alexandrian library, and so were not the kind of booklist we are
concerned with. They were, rather, a scholarly bibliography, with biographies of authors as
well asincipitsand line counts of their various works, precisely the kind of information that
we do not find in, for example, our no. 3. - Cf. Blanck 1992, 217 8. For book retrieval in practice in second century Rome, see
Houston 2004.
Papyrological Evidence for Book Collections and Libraries 243