^{page 223 note 2} Patzer, H., Das Problem der Geschichtsschreibung des Thukydides und die thukydideische Frage (1937)Google Scholar, merely wished to get the ‘question’ out of the way; by multiplying late passages (in my belief, recklessly) and by arguing that there are no traces of a oherent ‘early plan’ in the work. J. H. finley, in ‘The Unity of Thucydides’ History’ (Athenian Studies presented to Ferguson, W. S., 1940, pp. 255–97Google Scholar) and in his book Thucydides (1942), concluded from the conistency of Thucydides' thought that the work must all have been written at the same period, therefore after 404. Romilly, J. de, Thucydide et I'impérialisme athénien (1947Google Scholar), did lot deny that early and late passages can be distinguished, but was sure Thucydides lever substantially changed his mind: her lespair over the state of the ‘question’ (p. 12) s perhaps unecessarily deep. Meyer, C., ‘Zetemata 10’, Die Vrhmden im Geschichtsverk des Thukydides (1955)Google Scholar, is a recent example of German work which ignores or combats the views of the earlier analysts.

^{page 223 note 3} We can be certain that there has been evision, because of discrepancies between such post-war comment as 2. 65 and the larrative to which it should refer: see the excellent demonstration of Gomme, , J.H.S. xxi (1951), 70–73Google Scholar. These would be unaccountable if comment and narrative had aeen written all in one sweep. Gomme's inalysis will not of course tell us the interval Jiat elapsed between the two.

^{page 224 note 1} Pohlenz, immediately afterwards, in G.G.N. 1919, pp. 97 ff., and many scholars since.

^{page 225 note 1} If we trace the development back to 478, as Thucydides invites us to do, then Kimon contributed as much as anyone. Thucydides' liking for abstractions like covers this up, and much besides.

For some further nuances in the meaning of see Sealey, R., C.Q., N.s. vii (1957)Google Scholar, 1 ff. But, as will be seen, I have not followed him in his inferences from the tense of , and I believe that 23. 6 and 88 expound substantially the same doctrine.

^{page 225 note 2} Recapitulation, surely, not a fresh point, as it is required to be for Hammond's scheme, in which Thucydides states his theses in one order and proves them in the inverse order (C.Q., N.s. ii [1952], 134–5). Arguments from the order of presentation are not in any case decisive for my present problem, since, if Thucydides made additions, he could make those additions conform to the general system of the book.

^{page 225 note 3} It is, for instance, set out very clearly by de Romilly (see p. 223, n. 2), pp. 22–37.

^{page 225 note 4} The question, what he was thinking of when he wrote them, is considered below, § 5.

^{page 226 note 1} As Westlake points out in C.Q., N.s. viii (1958), 103–4, this formula does not imply that he wished to deny the importance of the allies' speeches; only that he insists on the deeper importance of his .

^{page 226 note 2} This point was taken up in passing by E. Kapp in his review of Schadewaldt, Gnomon, vi (1930), 100: ‘schon das schliesst m.E. aus, dass diese Korintherrede gleichzeitig mit 1,88 entstanden wäre’. I do not remember having seen it elsewhere.

^{page 226 note 3} In the text I have, after some hesitation, taken as Gomme does, as a reference to the late 430's; the with which the next sentence begins is unequivocally 432. But as Gomme points out, ‘it might be said that the power of Athens was clear enough, and her aggression against Sparta's allies had begun, a good deal earlier’, and should refer to the Messenian Revolt and the Arcadian Wars for the date of the latter cf. Phoenix, vi [1952], 1–5), so that the latter part of the long sentence reads as if it were all leading up to the outbreak of the First Peloponnesian War. Possibly the sentence is incomplete (if the editor bears any responsibility, as Wade-Gery tentatively suggests in O.C.D., s.v. Thucydides, p. 903, it is in die negative sense that he refrained from filling a gap). But it can be understood as it stands, in the light of Thucydides' known belief that Athenian power was at its height just before the war: between them cover the whole of the Pentekontaetia, … can be taken as a parenthesis, after which … brings us back to the present, the crucial moment when it was realized that Athens’ use of her power was intolerable. If there is a gap, one of the direct supports of my argument is gone, but I think die argument may still stand.

^{page 227 note 1} Not that Corcyra's treatment of Epidamnos is presented as admirable (e.g. 24. 6–7) or that Thucydides overlooks the sanction given by Delphi to Epidamnos' appeal to Corinth; but though this gave Corinth some justification () Thucydides' emphasis is all on Corinth's underlying hatred, which he analyses at some length.

^{page 227 note 2} Her reasons for rejecting the first proposals of Corcyra are presented as plausible (28. 4), but the second Corcyrean proposal, die offer of an armistice pending arbitration, seems more practicable dian the Corindiian alternative, and no case at all is put up for the final Corinthian refusal, (29. 1). Cf. also 39. 1–2.

^{page 228 note 1} The controversy over the meaning of (see Gomme, ad loc, and Steup: both perhaps analyse the position too formally, and I do not much like the idea of ‘private oaths’ at 5. 30. 2) does not affect the ques tion: whether ‘separately’ or ‘unofficially’, the Corinthians are the active party and other Peloponnesians are involved only to the extent that some of their citizens had joined Aristeus' expedition. Hammond's somewhat different picture is obtained by citing from 56 and 66 only those clauses which speak of rather than (C.Q., N.s. ii [1952], 135) and by neglecting in 58. 1 (C.Q. xxxiv [1940], 151).

^{page 228 note 2} There has been some misunderstanding of 58. , for which see Busolt-Swoboda, , Gr. Staatsk. ii. 687Google Scholar with n. 4, and the passages there cited, means the ‘competent authority’, so that the precise reference varies with the context: where only the assembly is competent it can mean the as sembly, as it does in Xen. Hell. 3. 2. 23, and one would suppose that the assembly was the competent authority here. In fact the Corinthians say to the Spartan assembly later (71.4), so the original promise was probably a quite regular decision of that body, and ‘unverbindlich’ (Schwartz 101) only in the realistic sense that there was no conceivable way of enforcing it.

The promise to Poteidaia must not be left out of account in assessing Thucydides'; beliefs about the causes of the war, but the measure of its practical importance is the fact that the Corinthians felt it necessary to continue their agitation. The Spartans were not committed beyond recall until an army had been sent out and had crossed the Attic border: even the decision in 87, which Thucydides does regard as critical, did not preclude further negotiation.

^{page 228 note 3} Patzer (see p. 223, n. 2), who wants everything to be late, suggests (57) that the Corcyreans' reference in 36. 2 to their commanding position on the route to and from the West was composed after the Sicilian expedition. The whole history of Athens' intermittent interest in the West shows that this is unnecessary, to say nothing of 2.7.2.

^{page 229 note 1} The record of Pausanias is enough basis for 77. 6, which has been taken as a reference to Lysander and his harmosts (e.g. Schwartz 115); and Archidamos, who in fact bequeathed the war to his son Agis at an early stage, might surely have enunciated the possibility in fact in 432 (81.6): see Gomme's notes. Only the Athenian speech gives rise to serious doubts, not because of particular anachronisms (and not, indeed, because the embassy could not have taken place) but because parts of it, especially 77, seem generally out of key widi their context (cf. Hammond, , C.Q_. xxxiv [1940], 150Google Scholar). Somediing may have been added here, though not necessarily after 404. The argument that the speeches are late because some of them use the idea of the is of course no more acceptable to me than to de Romilly (cf. p. 225, n. 3).

^{page 229 note 2} Cf. Hammond, , C.Q. xxxiv (1940), 151Google Scholar. ‘Willingness’ is not the only consideration: Spartan hesitancy is still an important factor, and it was quite uncertain whether or how soon the willingness of a majority might be translated into official action. 28. 1, where Spartan and Sikyonian envoys join Corcyra in 435 in trying to persuade Corinth to a settlement, is no evidence of a general desire for peace on Sparta's part. It was in no circumstances to her interest that Corcyra and Corinth should sink one another's ships: this is Athens’ interest, (44. 2).

^{page 230 note 1} The conclusion no doubt formed gradually in his mind, and was prepared by the gradual breakdown of Nikias' treaties—we may for instance assume that Thucydides began his inquiries into the battle of Mantineia at once, and expected to make use of them—but there could be no certainty till after the formal resumption of war.

^{page 230 note 2} Their general reaction to Alkibiades' speech is 6. 93. I. Cf. 7. 18. 2, .

^{page 231 note 1} Meyer, p. 393; Schwartz, pp. 141 ff., 235 ff.; and others since.

^{page 231 note 2} Plutarch, , Kim. 4. 2–4Google Scholar; Marcellinus 17, quoting Polemon .

^{page 231 note 3} Gr. Gesch. iii. 618–20. But cf. Strasburger's introduction to Horneffer's translation of Thucydides, p. xliv.

^{page 231 note 4} For a different view of the relations between Perikles and Kimon see Sealey, R., Hermes, lxxxiv (1956), 234–7Google Scholar.

^{page 232 note 1} Rev. Phil. 1929, p. 283; the stemma is also printed by Wade-Gery, , J.H.S. lii (1932Google Scholar), 210 = Essays in Greek History, p. 246: see his comments.

^{page 232 note 2} It is by no means common either, but in the fifth century we have one from Erechtheis (Tod 26. 30 = P.A. 7266), one (P.A. 7271), and one (P.A. 7272, whom Kirchner identifies with two others whose demes are not stated): that is, with the son of Melesias () and the historian (), from five different demes. The chances are against all these families being connected by marriage.

^{page 232 note 3} If Oloros' fadier was named Thucydides (which would mean, probably, that two sisters of Kimon married men of this name), and if the historian was an eldest son, then his name would be almost dictated by convention and would not be significant of any one's politics. In any other case the element of choice would be greater, and Oloros could not be unaware that he was giving his son a name with very particular associations.

^{page 232 note 4} J. H. Finley started an interesting speculation along these lines (Thucydides, pp. 20 ff.), but drew back (pp. 30–31) on the ground that the Kimonian connexion is not demonstrable and (more regrettably) that it is not important.

^{page 232 note 5} Cf. Croix, G. E. M. de Ste., Historia, iii (1954), iff.Google Scholar

^{page 232 note 6} Disregarding the Apollodoran estimate (244 F 7) which merely chooses the out break of the war for his , the only ancient statement is from Marcellinus (34), that he ended his life . This needs to be considered, if only because it is not based on an obvious calculation: if it has any basis in fact, we must take it, at least, that he was under 60 when he died, whenever that was. Among modern arguments, the most important is from his strategia in 424, best in the form which Steup gave it (introduction, pp. iii–iv), viz. that he could not speak as he does of Alkibiades in 420 when he was on the point of election (5. 43. 2), or as he allows Nikias to speak in 415 (6. 12. 2), if he himself had been elected at Alkibiades' age or younger. This would mean that he was well over 30 in 424, say about 35; which might be reconcil able widi Marcellinus so long as we do not extend his life long beyond 404.

^{page 233 note 1} For the elder Thucydides see Wade-Gery's article (p. 232, n. 1 above). There is some controversy over the extent to which he gained any real ascendancy in any part of this period: Wade-Gery thinks in terms of a temporary eclipse of Perikles, Gomme (pp. 386–7) and Ehrenberg (A.J.P. Ixix [1948], 160–3) doubt if matters went as far as this: the case cannot be argued here, involving as it does the whole question of the foundation of Thouria (on which see A.T.L. iii. 305 with nn. 19–20 for a modification of Wade-Gery's view). In any case the sources make it plain that Thucydides' main activities fall between 449 and 443, that he campaigned vigorously against Perikles in these years, but (I think) that he had no constructive policy of his own. This is enough for my present purpose.

^{page 234 note 1} The only directly attested date is that of Megara's complaint at Sparta, in the second half of 432 (Philochoros 328 F 121, but we knew this already from Thucydides). Gomme inclines to follow Steup in taking 42. 2 as a reference to the decree, which would mean it was passed before the Corcyra debate; but he notes the difficulty created by the word , and it seems to me that the phrase s a litde less uneasy if the reference is back:o c. 460, in which case the decree had probably not been passed when these words were, supposedly, spoken. Brunt, , A.J.P. lxxii (1951). 269 ff.Google Scholar, would date it well before 433, before the beginning even of the Epidamnos episode: this, if true, would remove it still more definitely from the context lere under discussion.

^{page 235 note 1} The Kallias decrees, as Meyer (p. 324) already noted, suggest that the assembly did by this time expect war.

^{page 235 note 2} Thucydides gives us but this one speech, and it has been surmised that it is in a sense composite, summing up what was said on various occasions. I mean no more than that Perikles is bound on some occasion, and likely on many, to have gone back thus into the past.

^{page 235 note 3} In Zahn, R., Die erste Periklesrede, p. 88, n.33.Google Scholar

^{page 235 note 4} Proc. Camb. Phil. Soc. 1952/1953, pp. 43–44Google Scholar. Jones assumes that the procedure would normally be along the lines described for 432 by Thucydides, and he is clearly right in saying that Sparta could not be committed to war without her consent: moreover, there was at least one precedent for the league's rejection of a Spartan proposal (Her. 5. 91–93). But it does not certainly follow that the Spartan assembly always took its decision first, and it is to be noted that a conference might be called by some other member, to judge from 67. 1.

^{page 236 note 1} Cf. Strasburger, , Hermes, Ixxxvi (1958), 17 ff.Google Scholar, on his attitude to the standard praises of Athens.

^{page 236 note 2} Cf. Meyer, p. 324.

^{page 236 note 3} There are, I think, two distinguishable problems here: (1) In large areas of his history, Thucydides confines himself to public statement and neither asserts nor speculates about concealed aims, although not only Perikles, but the leaders of the Spartan war party too, must have entertained wider designs about which they did not make speeches, and Thucydides can hardly have felt no curiosity about such matters. His reticence can be extremely frustrating in detail—we might be better off if we had been told his view of what Perikles expected to make of the capture of Epidauros (2. 56. 4) or Nikias of the Solygeia expedition (4. 42–45)—but it is by no means uniform: he tells us about plans of Demosthenes in 426 (3. 95. 1) and 424 (4. 76) which were certainly not made public in advance, and there is speculation enough about Tissaphernes' private mind (8. 46. 5, 87. 4). A survey of the cases might be helpful. The answer is possibly along the lines of Schadewaldt's distinction between his earlier and later conceptions of his task.

(2) The problem here before us, on the other hand, is to discover what Thucydides himself accepted from Perikles as the basis of his own conception of the nature of the war.

^{page 237 note 1} Thucydides keeps this point well in our view: cf. 44. 1, 45. 3, 49. 4, 53, 78. 4, 85. 2, 86. 3, 140. 2, 141. 1, 144. 2, and indeed 7. 18. 2. Sthenelaidas' rejection of arbitration is put in stark and shocking form, 86. 3 —an assertion of the right of self-help which would make all arbitration clauses useless for ever: even if Perikles' offers were throughout insincere, this style of intransigence could not be justified. The violent tone of this speech is not accidental, and the assembly which is swayed by it is not being treated as innocent.

^{page 238 note 1} In the same way I would argue that Thucydides' consciousness that he had something to retract is responsible for the unnecessary air of contradiction in 88 (p. 226 ibove).

^{page 239 note 1} A late date for these books was made the basis for Schadewaldt's view of a change in Thucydides' general aims; a view which seems to me extremely valuable even if it has to be put on another basis. The particular passages on which Schadewaldt relied could almost all have been written before the end of the war, and the crucial 6.15 is in part a later addition. I am glad to note that Westlake (p. 226, n. 1 above) prefers the earlier dating.

^{page 239 note 2} It is also a possibility that parts of book 4 were written comparatively late.

^{page 239 note 3} See the first chapter of Munch, H., Studien zu den Exkursen des Thukydides (1935).Google Scholar