¹ For the decree, first published in 1869, see Meiggs-Lewis, , GHI 20Google Scholar. The earlier colony decree, more recently discovered (Meiggs-Lewis 13), does not seem to be relevant to our discussion.

² Griech. Gesch. ii ⁶ (1888), pp. 168ff., 172ffGoogle Scholar. (fitting in the colony decree, not yet known at the time of his first edition, 1861). He does not explain how, on his view, Tolmides actually captured Naupactus: unless we assume a previous periplous (improbable and unreported), it is difficult to see how he could have got there.

³ GdA iii (1901), p. 597Google Scholar; unchanged in Stier's edition, iv.1 (1944), p. 563.

⁴ For Meyer's chronology see p. 591 of the first edition. He basically accepts Diodorus, to whom in general he gives little credit: in this instance, his dates happen, by accident (‘Es ist Zufall’), to be essentially correct. In a characteristic piece of imaginative fiction, he adds to his reconstruction that the Naupactians may have begged the Ozolians to take the city over, to defend it against the Aetolians.

⁵ Oldfather argues for both a strategic and a commercial purpose in the Athenian decision to occupy Naupactus: the main purpose, in fact, was the protection of Athenian trading interests against both Locrian and Aetolian piracy.

⁶ Meiggs and Lewis (p. 37), who also belong to this school, consistently date the law ‘Certainly earlier than the seizure of Naupaktos by Athens … c. 460 … (Thuc. i. 103.3); how much earlier, we cannot be sure.’ They follow Jeffery, , LSAG 108Google Scholar, in dating it ‘(?) 500–475’, to fit in with this view. They would probably admit that Jeffery's date is almost a pure guess, and that on epigraphic grounds a date even decades later could hardly be excluded.

⁷ Although in his general discussion of his method in using Thucydides' evidence as against other sources, he merely excludes acceptance of any later source where it contradicts Thucydides (84f.), he here inexplicably claims that Diodorus' report ‘deserves no credence’ even though it in no way contradicts Thucydides, except in chronology, as arrived at by emending Thucydides' well-attested ‘tenth year’ (i.103.1) out of the way. See further n. 13 below.

⁸ As will be clear, I am not denying that there are instances of the use of ἔχειν that do support Steup's view. But it must be stressed that mere reference to passages, without scrutiny and detailed analysis, is inadequate and will often turn out to be misleading.

⁹ I have tried to compile that list and discuss Thucydides' attitude in ‘Thucydides and the Outbreak of the Peloponnesian War: a Historian's Brief’, in Allison, June W. (ed.), Conflict, Antithesis and the Ancient Historian (Columbus, 1990)Google Scholar.

¹⁰ To toss off the suggestion that Thucydides made a mistake, as Gomme does (see p. 2 above), is too easy and explains nothing.

¹¹ Co-operation with the Spartan campaign (and, consequently, with the Phocians) is frequently stated as a fact, but is derived entirely from this Athenian action. It ought perhaps to be mentioned that Oldfather's reference to Athenian action at Eretria as parallel to that at Opus is unacceptable as such. Hesychius, s.v. Ἐρετριακ⋯ς κατ⋯λογος, records a psephisma of 442/1 [Oldfather misstates the date as 444], calling for the sons of the wealthiest Eretrians to be taken as hostages. To take the sons of leading citizens as hostages was, of course, standard practice. It is very different from taking the leading men themselves. Moreover, Hesychius does not give a number, although the decree must have specified one. (We know nothing about the background to this – further evidence as to the inadequacy of our information on the Pentekontaetia.)

¹² For my own views on the chronology of these actions, see EMC 23 (1988), 318f.Google Scholar, with notes 42, 43, 46.1 there saw no reason to doubt Tolmides' settlement of the Messenians at Naupactus, but was not sure whether to accept Diodorus' report that he actually captured the place on that occasion. At the time, I had not yet thought about the story of the Locrians. I would also now add that the periplous as such acquires a more serious purpose, commensurate with the effort, if planned not merely to show Athenian power and inflict some damage en route, but to capture a strategic site that could not be reached in any other way. (It would clearly be logistically impossible for Athenian forces to take Naupactus by an overland expedition.) To phrase it differently: if we were asked to conjecture how Naupactus was in fact taken (which is what we must do if we confine ourselves to Thucydides' evidence), it would be hard to avoid the answer that it must have been as a result of a periplous of Peloponnese; and that of Tolmides is the only relevant one recorded. Even if the Atthidographer on whom Diodorus' information is ultimately based was only guessing (which need not be assumed), his guess was the only rational one. [After completing this article, I was, through the kindness of Professor D. M. Lewis and Dr Hornblower, shown an advance copy of the treatment of these points in the new Cambridge Ancient History. I am happy to see that Diodorus' report that Tolmides settled the Messenians at Naupactus is there ‘provisionally’ accepted. The view that Tolmides actually captured Naupactus on that occasion is described (without any argument in the portion shown to me) as certainly contradicting Thucydides: I do not know what alternative occasion is proposed for the capture of Naupactus, but the issue will no doubt be faced and discussed.]

¹³ See n. 7 above with text [and compare what I have seen of the treatment in the new CAH: see preceding note]. It is interesting to compare an earlier scholar who set an example of rational reliance on Thucydides which Gomme might well have considered. Mitford, William, History of Greece ii (1814), p. 394 n. 12Google Scholar, set down his principles as follows: ‘I have been very cautious of following any other writer, in relating the transactions of these times, when not in some degree supported by [Thucydides].’ He goes on to denigrate Diodorus, comparing his relationship to Thucydides to that of Livy to Polybius. Yet he here thinks Diodorus' account ‘in itself probable, and consistent with every authenticated fact’, and he accepts Tolmides' settlement of the Messenians at Naupactus and the capture of Naupactus by Tolmides. (It is not clear how he chronologically correlates these two events.) He also accepts 455 as the date, no doubt from Diodorus (who gives 456/5, the date of the Atthis for the periplous): ibid. 397.

¹⁴ Peaceful accommodation may already have been revealed by Meiggs-Lewis 74, though that document is too late to be confidently used as evidence for the original relationship. Meiggs-Lewis, I.c., referred to an unpublished inscription confirming this interpretation. [I see from the new CAH that it is still unpublished!] This was noted by Mastrokostas, E. in Arch. Delt. 19 B 2 (1964 [1966]), p. 295Google Scholaras ‘very important’, and as describing the foundation of a joint community by the Messenians and the Naupactians. It is perhaps worth making clear that the ‘Naupactians’, both in that text and in Meiggs-Lewis 74, should be taken to be Thucydides' Ozolians at Naupactus, not any putative pre-Locrian population, of whom we have no record and whom we cannot even further define. The foundation document for the Locrian colony (see n. 1 above) does not mention any accommodation with those inhabitants and we must assume they were expelled. That the Locrian colonists, like colonists elsewhere, called themselves citizens of the colony (in this case ‘Naupactians’) was in any case to be expected and is clear from the foundation document. The Messenians were probably not sufficiently numerous to fill the large city [for its size, see now the careful description, CAH, I.c.] by themselves and would naturally want to share it with the ‘Naupactians’. That the two communities, although jointly protected by Athena Polias, remained notionally and perhaps politically separate is, however, shown by Meiggs-Lewis 74.

¹⁵ See the essay cited n. 9 above. I should like to thank Dr Simon Hornblower for helping to eliminate some errors in this article and for enabling me to obtain permission to see a page (with notes) of the forthcoming CAH. Any remaining errors are my own fault.