^{page 211 note 1} Hereafter all references are to the fourth book of Thucydides unless otherwise stated.

^{page 211 note 2} It is true that he does not supply any details about the defence of Pylos against attacks by land.

^{page 211 note 3} Cf. Grundy, G. B., Thucydides and the history of his age, ii (1948), 129: ‘Thucydides’ account of what happened after this is not very easy to understand.’Google Scholar

^{page 211 note 4} On the number see Gomme, A. W., Historical Commentary on Thucydides, iii (1956), 450.Google Scholar

^{page 211 note 5} See below, p. 215.

^{page 211 note 6} The term , which Thucydides uses frequently in his account of the episode, undoubtedly denotes the bay of Pylos (or Navarino), though it was far larger and deeper than other harbours known to him.

^{page 212 note 1} This point will be discussed below, p. 216. There is also implied criticism of their failure to implement their intention to block the two channels leading into the harbou. Undoubtedly they ought to have kept the Sikia channel continuously blocked so long as the Athenian fleet was in the vicinity, a simple operation which had obvious advantage. On the other hand, Thucydides is palpably mistaken in believing that the southern channel could be blocked by eight or nine ships (8. 5–6), as is acknowledged by most scholars, including Pritchett, W. K., Studies in Ancient Greek Topography, i (1965), 22 and 29, who considers that he is not guilty of any other topographical errors. This error provides a very strong argument, amounting almost to proof, in support of the generally accepted belief that his account of the operations at Pylos is not based on autopsy.Google Scholar

^{page 212 note 2} As was usually the case, triremes being rowed at speed enjoyed an enormous advantage over triremes moving slowly or notat all, cf. 2. 91. 4–92. I.

^{page 212 note 3} Such appears to be the meaning of . The ships did not necessarily suffer extensive damage. In the two other instances where Thucydides uses the verb in accounts of naval operations (8. 13; 8. 105. 3) almost all the ships subjected to this treatment are stated to have made their escape.

^{page 212 note 4} The familiar Thucydidean terms , and appear together in a single sentence in 14. 3 (though Gomme, Hist. Com. iii. 452, is justified in objecting to the sentiment expressed in this passage).

^{page 213 note 1} See below, p. 215.

^{page 213 note 2} Cf. Busolt, G., Gr. Gesch. iii. 2 (1904), 654Google Scholar n. I; Schwartz, E., Geschichtswerk des Thuhydides (1929), 293Google Scholar; J. H.Finley, Thucydides (1942), 188. Recently Silhanek, D. K., C.W., lxiv (1970), 10–13, has maintained that Demosthenes was the primary source for the entire account of the Pylos episode.Google Scholar

^{page 213 note 3} Cf. my Individuals in Thucydides (1968), 97.

^{page 213 note 4} Silhanek, op. cit. 11–12; but his suggestion that the final stage of the battle was fought directly beneath the crags of Pylos is not at all convincing.

^{page 214 note 1} It is true that there is no hint of any friction after the fortification of the Pylos peninsula.

^{page 214 note 2} Gomme, , Greek Attitude to Poetry and History (1954), 147–8, notes the contrast between the favourable treatment of Nicostratus and the unfavourable treatment of Eurymedon.Google Scholar

^{page 214 note 3} Cf. Gomme, , Essays in Greek History and Literature (1937), 125–31 and Hist. Com. iii. 485. Passages in which some of the detail was probably derived from Spartan prisoners are: 26. 5–8; 32. 1; 34. 2–3; 38. I (perhaps also the anecdote in 40. 2). The narrative does not seem to contain any information which must have been obtained from Peloponnesians after Thucydides became an exile (see below, pp. 224–5).Google Scholar

^{page 214 note 4} Apart from his error in stating that the southern entrance to the harbour could be blocked by eight or nine ships (see above, p. 212 n. 1).

^{page 214 note 5} It is 5 km. long and 4 km. wide at its widest point according to Philippson, A., Landschaften, Gr. (ed. E. Kirsten), iii (1959), 387Google Scholar, whose geographical description of this district (ibid. 386–90) is packed with valuable information.

^{page 215 note 1} Cf. Gomme, Hist. Com. iii. 451 and 482. Thucydides gives the length of Sphacteria as 75 stades (8. 6); the figure should be 24. This error perhaps lends some support to the view that he failed to appreciate the size of the harbour, but it may be the result of textual corruption.

^{page 215 note 2} Pritchett, op. cit. [above, p. 212 n.I], 24.

^{page 215 note 3} The problems of this disputed passage cannot be discussed here.

^{page 215 note 4} Some of these apparently arrived only after the naval battle, cf. 74. 5 .

^{page 216 note 1} In 34. I and 2. 92. I the phrase is used of attackers who were already close to their adversaries.

^{page 216 note 2} There is, according to Casson, L., Ships and Seamanship in the Ancient World (1971), 279Google Scholar n. 37, ‘no information on how fast an ancient galley could spurt’. If, however, the somewhat speculative calculations of Rodgers, W., Greek and Roman Naval Warfare (1937), 516, are accepted and some allowance made for the fact that the rowers must not be allowed to exhaust themselves before the fighting began, the distance in this case can hardly have been covered in less than one and a half hours.Google Scholar

^{page 216 note 3} See above, p. 212.

^{page 216 note 4} Wilson, J. and Beardsworth, T., C.Q. xx (1970), 50, make this attractive suggestion. Their view that the Athenian fleet may have been thought to have ‘gone for good’ is less convincing. Even the most optimistic Spartans can hardly have imagined that Eurymedon and Sophocles would abandon the Athenians besieged on Pylos without making even the smallest effort to help them.Google Scholar

^{page 216 note 5} A possible explanation of Peloponnesian unreadiness is that, when the Athenians entered the harbour, the Spartan High Command assumed that their primary aim was to force a landing somewhere (cf. 8. 8), not to fight a naval action. If so, the task of repelling an attempted landing would best be entrusted initially to the troops on the shore, while the Peloponnesian fleet was held in reserve until the Athenians were fully committed and its intervention might prove most effective.

^{page 217 note 1} Abydos: Xen. Hell. 1. 1. 4–7; Diod. 13. 45–6. Cyzicus: Xen. Hel. 1. 16–18; Diod. 13. 50–1.

^{page 217 note 2} See above, p. 212.

^{page 217 note 3} Apparently no Athenian ship was lost or damaged beyond repair (23. 2), but some Athenians were wounded (14. 4).

^{page 217 note 4} Michell, H., Sparta (1952), 275. They contributed only ten ships to the fleet at Arginusae (Xen. Hell. 1. 6. 34).Google Scholar

^{page 217 note 5} They had at the outset sought to avoid the risk of a naval action (8. 8).

^{page 218 note 1} See below, pp. 220–1.

^{page 218 note 2} The motives of the Athenian generals, when the truce was negotiated, in insisting on the surrender of all the ships in Laconia as well as those at Pylos (16. 1) may have been similar. They may have suspected that during the truce an attempt might be made to smuggle out Spartans from Sphacteria on ships hitherto based in Laconia and thus to present the Athenians with a fait accompli. Only a small squadron would be required for this purpose. The Spartan troops who rushed into the sea realized the potential value of ships in saving their comrades on the island (see above, p. 212).

^{page 218 note 3} For example, Cornford, F. M., Thucydides Mythistoricus (1907), 87:Google Scholar ‘The Athenian fleet… sailed in and knocked them into bits’; Steup, n. on 14. 5: ‘Die Flotte, die so gut wie vernichtet war’; J. H. Finley, op. cit. [above, p. 213 n. 2], 192: ‘the Athenians[ shattered the enemy fleet.’ On the other hand, the attempt by Ferrabino, A., L'impero ateniese (1927), 157–9, to show that the Peloponnesians were more successful than the Athenians is unconvincing. He maintains that the Peloponnesian fleet was under orders to withdraw to the shore if attacked and thus suffered little damage; that Thucydides has based his account on a dispatch sent to Athens by Eurymedon and Sophocles which created the impression that they had won a victory; that the Spartans on Sphacteria were not at this time in any great danger. It is, however, difficult to believe (i) that Eurymedon and Sophocles can have made an essentially false report when its falsity was bound to be exposed by men who, for one reason or another, returned to Athens and did not accompany them to Sicily; (ii) that, even if a report of this kind was sent, Thucydides can have been misled by it when he obviously had information from eyewitnesses, probably including Demosthenes; (iii) that the Athenians who negotiated the truce can have been in a position to impose terms so unfavourable to the Spartans (16. 1–2) if their own fleet had accomplished almost nothing and if the troops on Sphacteria were not seriously threatened.Google Scholar

^{page 218 note 4} The closest parallel is 129. I, where the rebels on the peninsula of Pallene whom Brasidas considered himself to be unable to ‘assist’ () are the people of Scione, which was in danger of either being taken by storm or reduced to surrender by blockade (as it eventually was, 5. 32. 1).

^{page 219 note 1} J. de Romilly (Budė) translates ‘d'exercer des représailles pour libérer leurs hommes’, which may well be what Thucydides means but is an interpretation rather than a translation.

^{page 219 note 2} Huart, P., Annales de la Faculte des Lettres de Nice, xi (1970), 31, cf. 44, infers from this passage that the Spartan authorities, with characteristic selfishness, considered only the safety of their own men, especially the Spartiates, and ignored the interests of their allies.Google Scholar

^{page 219 note 3} He sometimes uses in con texts where ships were neither sunk nor damaged beyond repair. The verb occurs four times (2. 90. 5; 91. I; 92. 2 and 5) when in the battle off Naupactus nine Athenian ships were trapped close to the shore by the enemy, although they were later recovered (92. 2), some of them doubtless in a damaged condition. Clearly, however, the verb could not be applied to ships lost through being handed over temporarily and not restored.

^{page 219 note 4} Cf. II. 4, where the exhortations of Brasidas during the unsuccessful attempts to force a landing on Pylos show that the Spartans expected sacrifices by their allies in return for benefits conferred upon them.

^{page 220 note 1} This problem was recognized by Grote, G., History of Greece (1888), v. 241, who refers briefly to it after analysing the military situation, but it has been strangely neglected since his time.Google Scholar

^{page 220 note 2} Gomme, Hist. Com. iii. 468, suggests that in windy weather, when the Athenian patrols would be less effective, an attempt might have been made to take the men off in the boats which brought their food. An operation of this kind would, one imagines, have appealed to Brasidas, but he had been wounded (12. 1).

^{page 220 note 3} Wilson, J. and Beardsworth, T., Phoenix xxiv (1970), 116–18, consider various methods of rescue which might have been attempted, but they conclude that the risks were too great in each case.Google Scholar

^{page 220 note 4} See above, p. 217.

^{page 220 note 5} Wilson and Beardsworth, ibid., discuss the prospects of rescue by the southern channel when only small boats were available. In their opinion, which may well be right, the Athenians were then in a position to prevent evacuation by this route.

^{page 221 note 1} This suggestion is based on autopsy. The operation could have been conducted, partly or wholly, under cover of darkness.

^{page 221 note 2} Ferrabino, op. cit. [above, p. 218 n. 3], 159, convincingly maintains that the Spar tans welcomed the opportunity of making peace.

^{page 221 note 3} The ensuing discussion of this speech will be limited to aspects of it which are relevant to the present investigation.

^{page 222 note 1} Cf. the admission by Brasidas in his speech at Acanthus (85. 2).

^{page 222 note 2} Brunt, P. A., Phoenix xix (1965), 260–3 with full references.Google Scholar

^{page 222 note 3} An order issued to the Spartan commander at Plataea in 427 (3. 52. 2) shows that at that time the government at least envisaged the possibility of a negotiated peace.

^{page 223 note 1} Cf. Busolt, op. cit. [above, p. 213 n. 2], 1079.

^{page 223 note 2} Cf. my observations in Ryl. Bull. liii (1970), 230.

^{page 223 note 3} The cautious treatment by Gomme, Hist. Com. ii. 391, is abundantly justified.

^{page 224 note 1} At that moment the Athenians expected to bring the operation to a successful conclusion within a few days (26. 4).

^{page 224 note 2} Erbse, H., Rh. Mar. xcvi (1953), 38–62, who points to 1. 56–9; 4. 80–1; and 7. 27–8 as the principal examples, cf. my Individuals in Thucydides, 192, where I note that the letter of Nicias from Syracuse pictures the situation not so much as it was at the time of writing but rather as it became six months later.Google Scholar

^{page 224 note 3} See above, p. 214.

^{page 225 note 1} Busolt, op. cit. [above, p. 213 n. 2], 659 n. 4; Brunt, op. cit. [above, p. 222 n. 2], 278–80. It is significant that the period on which Thucydides is most fully supplied with information from behind the scenes at Sparta is when Alcibiades was involved in discussions and intrigues leading to the formation of Spartan policy after the Athenian disaster in Sicily (8. 5–12).

^{page 225 note 2} Neither the names nor the status of the officials sent from Sparta are mentioned, though they doubtless included at least one of the ephors and possibly one of the kings. They are defined merely as (15. I).

^{page 225 note 3} See above, p. 214 with n. 3.

^{page 225 note 4} I have considered this point in The Speeches in Thucydides, edited by Philip A. Stadter (1973), 97–8.

^{page 225 note 5} Nicias established a reputation for consistent friendship with Sparta: 5. 16. 1, 43. 2, and 46. 1–4; 6. 89. 2; 7. 86. 3; Plut. Nic. 9. 6, cf. 27. 5–6.

^{page 225 note 6} See above, pp. 221–4.