and we thus should not ordinarily draw conclusions based on the absence of,
say, a given author in a given concentration.
There are problems of another sort, too, in dealing with this material,
leading to similar uncertainties. Once the papyri were placed in the
dump, they (or detached fragments of them) might blow from one spot
to another, thus contaminating or at least confusing our concentrations.
Nor do dumps provide the kind of stable stratigraphy, especially when the
archaeologist is concerned only to find papyri, that other kinds of sites do:
sections of material can slide down slopes, and other trash can be depos-
ited within a concentration before it is sealed.
45
Even so, such concentra-
tions provide, as we will see, remarkably coherent collections of material,
and if we work with what is there, rather than with what is missing, we
can draw some useful conclusions. The concentrations and other collec-
tions attested archaeologically that I have been able to identify so far are
presented in table 10.2.
Considered thus in summary form, these concentrations are reassuringly
similar to the collections represented by the lists in table 10.1. They range
laterindate,nos. 8, 9,and10all being later thanany of the lists intable 10.1,
but there is a clear preference for classical authors, as opposed to writers of
the Roman imperial period; we find both specialized libraries (nos. 1 and 6)
and moregeneral collections;theprobablenumberofvolumes varies widely
along a range from fewer than twenty (nos. 2, 3, and 10) to well over a
thousand (no. 1); and most collections, however small, include one or more
‘‘subliterary’’ items, such as grammatical works, commentaries, glossaries,
and author’s drafts (nos. 2, 4, 5, and 7, respectively, as examples). All of this
is similar to what we saw in the lists of books. The variations in types of text,
and the relatively high frequency of classical literature, are in both cases
simply a reflection of what we find in the papyrological record generally and
not at all surprising. The real question about these concentrations of papyri
can be simply put: what do we learn when we thus divide papyri into their
probable original collections? In what follows, I will suggest five things that
we can learn by doing this.
- The useful life of manuscripts. Within the concentrations of papyri, we
frequently find manuscripts that were two or more centuries old at the
time the collections to which they belonged were thrown out or
destroyed. A few examples will suffice. In no. 4 (Breccia 1932), which
was discarded sometime aroundA.D. 300,
46
there were three manuscripts
- I am grateful to Nikolaos Gonis for discussing with me the problems involved in
identifying the specific papyri that were found in any given concentration. - This collection was found intermingled with documents pertaining to a man named
Sarapion alias Apollonianus, two of which have dates ofA.D. 265 (PSI1249 and 1250). Even
if the books did not belong to Sarapion, it is quite clear that the books and the documents
were discarded at roughly the same time, in other words, probably about a generation after
the latest dated document, and so around 300.
248 Institutions and Communities