A third group of dialogues can be identiWed by a set of common features:
(1) they are short; (2) Socrates appears as an inquirer, not an instructor; (3)
the Theory of Ideas is not presented; and (4) stylometrically they are at the
greatest remove from the late group Wrst identiWed. This group includes
Crito, Charmides, Laches, L ysis , Ion, Euthydemus, and Hippias Minor. These dialogues
are commonly accepted as those most likely to be presentations of the
philosophical views of the historical Socrates. Here too belongs the Apology,
in which Socrates is the sole speaker, on trial for his life, and which in
philosophical content and stylometric features resembles the other dia-
logues of the group. The Wrst book of the Republic, too, in both content and
style, resembles this group more than it resembles the remaining books of
the dialogue: some scholars suppose, with good reason, that it Wrst existed
as a separate dialogue, perhaps under the title Thrasymachus.ItisdiYcult to
assign a chronology within this early group, though some authors place
the Lysis Wrst and assign it before 399, on the basis of an ancient anecdote
that it was read to Socrates himself, who said, ‘what a load of lies this young
man tells about me’ (D.L. 3. 35).
In my view there is good reason to accept the general consensus that
thus divides the Platonic dialogues into three groups, early, middle, and
late. The division results from the striking coincidence of three independ-
ent sets of criteria, dramatic, philosophical, and stylometric. Whether we
focus on the dramatic role given to Socrates, or the philosophical content
of the dialogues, or tell-tale details of style and idiom, we reach the same
threefold grouping. Twentieth-century developments in stylometry, with
much more reWned statistical techniques, and with vast amounts of new
data obtained from computerised texts, have essentially done little more
than conWrm the consensus achieved in the late nineteenth and early
twentieth century.20
A number of dialogues, however, do not fall clearly into one of the three
groups, because the three criteria do not so happily coincide: the most
important such cases are Cratylus, Euthyphro, Gorgias, Meno , Phaedrus, Parmenides,
Protagoras, Theaetetus. Here more recent stylometric studies have thrown new
light on the problems.21 There is no space here to enter into the detailed
20 The consensus has been signiWcantly questioned only in respect of the Timaeus and its
appendix, the Critias. The debate here will be examined later when I discuss Plato’s Theory of Ideas.
21 See L. Brandwood, The Chronology of Plato’s Dialogues (Cambridge: Cambridge University Press,
1990); G. Ledger, Re-counting Plato: A Computer Analysis of Plato’s Style (Oxford: Clarendon Press, 1989);
PYTHAGORAS TO PLATO
40