Epilogue
have to reflect the grammar and spelling of archaic Latin or of early Greek
documents, where relevant, or of papyri or rustic inscriptions, and even—
to be fully purist—to represent variant letter forms.
Of course, in the real world the level both of collective effort, in deciding
on the principles to be followed in a document-based dictionary or lexicon,
and in reading and analysing the vocabulary, grammar, and script found in
millions of surviving items of evidence, is completely inconceivable, as is the
investment required for any such enterprise. So, in reality, we must reconcile
ourselves to making do with guides to the classical languages which are not
based on the vast evidence now available for Greek and Latin as they were
actually written in the ancient world.
This unattainable prospect is even more tantalising when we consider that
we have at our disposal a very significant range of, in one sense, literary
texts for which the manuscript evidence does in fact date from the ancient
world. For this purpose I suggest, as noted above, that the moment of the
Islamic conquests in the Levant in the seventh century can serve as a rea-
sonable terminal point for ‘‘antiquity.’’ This definition would leave us with,
as genuinely ‘‘ancient’’ ancient texts, not only large parts of Homer, as pre-
served on hundreds of papyri, but a mass of other Greek poetry, and some
Latin, also known from papyrus, not to speak of scraps of theAeneidfrom
Masada, in one corner of the Empire, and from Vindolanda, in the diametri-
cally opposite one;^10 but also (for example) the complete surviving text of the
AristotelianAthenaion Politeia, also from a papyrus, some fragments of Gaius’
Institutes, the late Roman illustrated manuscripts of Vergil,^11 virtually the
complete Greek Bible, from the fourth-centuryCodex SinaiticusandCodex
Vaticanusand the fifth-centuryCodex Alexandrinus;^12 and ( just about) theDi-
. For the quotation ofAeneid. from Vindolanda, see A. K. Bowman and J. D.
Thomas, eds.,TheVindolandaWriting-Tablets(TabulaeVindolandensesII, ), , no. ; and
for that ofAeneid. from Masada, see H. M. Cotton and J. Geiger,MasadaII:The Latin
and Greek Documents(), , no. .
. See D. H. Wright,The Vatican Vergil: A Masterpiece of Late Antique Art(); and
The Roman Vergil and the Origins of Medieval Book Design(), with the challenging re-
view article by Alan Cameron, ‘‘Vergil Illustrated between Pagans and Christians,’’JRA
(): .
. It is relevant, at least in principle, that two of these greatCodicesare available in
facsimile editions. For theCodex Sinaiticus,H.LakeandK.Lake,Codex Sinaiticus Petropoli-
tanus et Frederico-Augustanus Lipsiensis(), for the New Testament, and (), for the Old
Testament; alsoThe Codex Alexandrinus in Reduced Photographic Facsimile, Old TestamentI–
III (–), published by the British Museum. For a convenient account of both, see the
pamphletThe Codex Sinaiticus and Codex Alexandrinus(), published by the British Mu-