¹ See, e.g., the statement that ‘The Areopagus was no doubt implicated in Cleisthenes’ exile, but, nonetheless, this body may have been the “boule” that headed the resistance [to Cleomenes]’ (CP lx [1965], 99). A recent article refers [JHS xc [1970], 182) to the ‘evenness of the political struggle [from Cleisthenes’ reforms] until after Marathon’—a period for which, as far as I am aware, precisely one political fact is known (cf. n. 44 below)—as an accepted truth.

² Ath. pol. 22. 5. I shall not refer to the author of this treatise as Aristotle.

³ All in chapter 22. On the interpretation of the author’s chronology, I have nothing to add to Raubitschek, Historia viii (1959), 127f. (table p. 128), which seems to solve all the internal puzzles of the text, without the usual violence. But I cannot accept his attempt to insert the Nicomedes of Ath. pol. in 482/1 after Dionysius’ Nicodemus (483/2). He acknowledges that these names are often confused in the texts, and Kenyon read ‘Nicodemus’ in the Berlin fragment of the Ath. pol. itself. I think it should be accepted that they are identical and that the difficilior lectio ‘Nicomedes’ is correct. Though it is technically easy to reconcile Dionysius and Ath. pol. as Raubitschek does, we must remember that there is another clash between them two decades earlier, over an archon for whom see pp. 3–4. The reconciliation of Plutarch’s datum on Aristides’ return with the archon date for his ostracism has been attempted along Raubitschek’s lines before, but will not readily persuade: Plutarch firmly puts the decree on the return of the ostracized in the third year after the ostracism. It is clear that the author of Ath. pol. did not have a chronological table set out by archons in front of him while writing: the vaguenesses and actual errors would in that case be quite inexplicable. As ancient writers often did, in view of the difficulty of actually using reference books as one wrote, he probably worked from his own notes, taken some time before, without taking any further trouble where they were incomplete or faulty. This seems the only way to explain such phenomena as the nature of chapter 22, where apparent precision is found hopelessly mixed with vagueness and error. This method of work was, of course, common enough in antiquity: we need only recall the massive notes kept by the elder Pliny—as reported by his nephew (Ep. iii 5)—and their frequently confused end product. There is certainly no reason to prefer our author’s testimony to that of Dionysius where they conflict. (On the attested merits of Dionysius, see Frost, , CSCA 1, 114, n. 46.)Google Scholar However, one must always try to make sense of an author on his own premises and find out what he meant to say, whether it turns out to be right or wrong.

⁴ As rightly pointed out by Sumner (CQ,xi [1961], 35 f.). His own suggestion of assuming a lacuna is, however, a measure of despair, which closer study of the text shows to be unnecessary. (He abandoned it BICS xi [1964], 84.) Not enough attention has perhaps been paid to the author’s apparent method of working (see last note).

⁵ On Fornara, , CQ 8 (1963), 104.Google Scholar

⁶ This has often been noted. See Appendix A.

⁷ See Day, and Chambers, , Aristotle’s History of Athenian Democracy (Berkeley, 1962)— instructive even where the authors’ views are not accepted in detail.Google Scholar

⁸ This, first pointed out a long time ago (see Ath. pol. 3.3), has recently been stressed by Sealey, , Essays in Greek Politics (New York, n.d. [1967?]), p. 20.Google Scholar

⁹ See n. 3 above.

¹⁰ Buck, , CP 60 (1965), 96 f.Google Scholar

¹¹ That this last part is false is conclusively shown by Hammond, , (CQ 19 [1969], 3 f.).CrossRefGoogle Scholar

¹² CAH iv (1926), p. 266. Whatley’s comment (JHS lxxxiv [1964], 128—but written 1920) on Munro’s ‘Sherlock Holmes method’ seems applicable to his colleague Walker.

¹³ Needless to say, Plutarch no more ‘confirms’ Herodotus than Thucydides ‘refutes’ him. But see Plut. Them.4.2 for his own view, polemically put. Plutarch knows Thucydides; but, of course, there may have been others who said the same.

¹⁴ That paradoxical assumption is frequently met with. It was properly castigated by Whatley, op. cit. (n. 12). For a sane view, at present still isolated, I can now refer readers to the work of Frost, referred to in the introductory note above.

¹⁵ Frost, op cit. 105, nn. 1–3, collects evidence showing some extraordinary views, which will serve to illustrate my points here. I should add, in fairness, that Hignett is well aware of Themistocles’ birth and station. The decisive evidence is in Plut. Them. 1.4. We do not know whether his father was actually a Lycomid, or how close the connection was. Hignett overstates the case.

¹⁶ There is no book more remote from the CAH in spirit and purpose than Thomson, Aeschylus and Athens. The view here described will be found there in Chapter XIII. See also Wade-Gery, , Essays in Greek History (Oxford, Blackwell, 1958), pp. 165, 177 f.Google Scholar

¹⁷ More Essays in Greek History and Literature (Oxford, Blackwell, 1962), p. 21 (from AfPh lxv [1944]).

¹⁸ Plut. Arist. 5. 4 stresses the rivalry, but does not call him strategus (while Aristides is so called and—significantly—takes a prominent part in the council of the generals). This is almost decisive. It is confirmed by the fact that the whole story is not mentioned in the Themistocles. He appears as a general even in so sagacious a book as Burn, , Persia and the Greek (London, 1962), p. 250.Google Scholar

¹⁹ D.H. vi 34. That the biographic tradition did not pick the office up and embroider it shows its relative unimportance and the absurdity of trying to build up his tenure of it into a historic occasion with conspicuous activity on his part. But unless and until a homonym can be produced, the fact of the archonship must be accepted. (See below.)

²⁰ FgrHist 328 F 40:

(Philochorus’ reference [Book v] reads like a flashback to the original fortification.) Gomme (Hist. Comm. i, p. 262) rightly says that we could not have helped hearing, if Themistocles had been at the head of the list of dedicants. He adds that, had the coast wall been anywhere near completion at the time of Xerxes’ invasion, ‘we should have expected it to play some part in the story’. However, I would not follow him in his denial of Themistocles’ archonship: the name is rare, at this period, and—unlike the case of Aristides—no other bearer of the name is attested, even in the recent ostraca. If one ever turns up, Gomme will be vindicated. In the meantime, we must continue to accept the archonship. Nevertheless, Gomme’s sagacious treatment of Themistocles’ beginnings is (despite his error in describing him as a nouus homo) a welcome contrast to the standard modern fiction. Acceptance of the archonship does not imply acceptance of the building activity in it. As we shall see, it would be an oddity in the history of the archonship at that period; and, of course, Thucydides can use arche for any office. Gomme may well be right in suggesting a special office held nearer the time of the shipbuilding. But the ‘annual’ () arche may well be the strategia, perhaps conferred on him as early as 472, after the ostracism of Aristides.

²¹ On this I am happy to say that Frost (op. cit.) has anticipated some of my conclusions, so that this part of my argument can be shortened. What I have to say here will, I hope, supplement rather than duplicate his arguments. The general point is suggested by Sealey, op. cit. (n. 8), 20 (originally published in 1960).

²² But note that one of the most recent and independent-minded of all, Hammond, still accepts the disastrous influence of the reform on the calibre of the archons and the standing of the archonship as ‘given’.

²³ Raphael Sealey points out to me that there is no positive evidence for a Solonian restriction to the first class. (Since that class was probably only instituted by Solon, it would have to be a positive restriction.) However, while admitting this state of the evidence, I still think that such a restriction is implied in the account of Ath. pol. (reading 7.3 together with 8.1; in the former the archons are mentioned before the treasurers, in the latter the restriction of the treasurers to the pentacosiomedimnoi is asserted); and I see no argument against accepting the author on this. Demetrius of Phalerum thought that this still applied after Cleisthenes (see n. 25 below); but I would not stake much on that. Certainly, Ath. pol. (again with no argument against acceptance) states that the first two classes were eligible just before the reform of 457. The tale in Plut. Arist. 22. 1 is not evidence for anything except the nature of later fiction.

²⁴ Dated by our author with his usual incompetence. He first gives a figure of six years after Ephialtes’ death (which could not on any account be earlier than 457/6) and then says that Mnesitheides (known as archon 457/6) was the first zeugites to be elected under the new law (Ath. pol. 26.2). Cf. note 28.

²⁵ Quoted Plut. Arist. 1. 2 f. Plutarch contradicts him on other points, but not on this one. Presumably he found no other firm evidence on the matter, any more than we can.

²⁶ From Solon to Socrates (London, 1968), p. 142.

²⁷ Hammond provides the most recent instance (e.g. CQ. xix [1969], 140: ‘The prestige of all the archonships was severely injured in 487/6 by a reform whereby the nine archons and their [sic] secretary were selected by lot … ’). But see again Frost, 114. For the evidence here discussed, see Appendix Kiechle, B., (HZ cciv [1967], 294 f.) has brief comments on some of the archons after Telesinus.Google Scholar

²⁸ We must bear in mind (but it does not affect the argument much) that the reform may in fact have been passed in the year before Telesinus and he may have been the first archon who gained office under the new law. (See n. 24.) Our author simply cannot be trusted for such details.

²⁹ Plut. Arist. 1. 8; cf. 5. 9. It is an interesting study in the growth of biographic myth, and an instructive example of Plutarch at his most critical (but not quite critical enough).

³⁰ Hesperia viii (1939), 59f. I have seen no good argument advanced against his interpretation.

³¹ The argument (Arist. 1.6) as put by Panaetius is an example of Hellenistic scholarship at its best—and remarkably close to the way in which we still operate. Aristides son of Xenophilus has now turned up on a Ceramicus ostracon: see Willemsen, , Arch. Deltion 23 (1968), 28 f.Google Scholar This report must be used with great caution, as Willemsen unfortunately is not familiar with this material and, since (in this preliminary report) he gives no details of what was actually on most of the ostraca, his identifications (i.e. the way in which he correlates bare personal names, names and patronymics, names and demotics, and all three, and allows for errors on the part of the inscriber) cannot be checked. Thus, in the case of Aristides, he treats us to the statement that there are 32 ostraca with the name of Aristides son of Lysimachus ek Koiles. That would be a major shock: a homonym of the great Aristides discovered and well attested. In fact, various sources confirm to me that there is precisely one sherd with this inscription—obviously a mere error for the great Aristides (who is from Alopece); the other 31 omit the deme. Unfortunately the reader cannot tell how much more misinformation of this kind is hidden in that short list.

³² Frost has made the attractive suggestion that archons, at this period, may have been young men and inexperienced before appointment (see n. 21). He does not notice that the case of Aristides (general already at Marathon) contradicts this. I am happy to strengthen what seems to me a good case by removing the contradiction.

³³ Willemsen, op. cit.

³⁴ See (conveniently) Kirchner, PA ii 53 (with stemma facing).

³⁵ The interval has been shortened by the emergence of the fact that he shares a few ostraca made from the same pot with Megacles (Willemsen, l.e. with Plate 19a)—obviously not after the year of Megacles’ ostracism, 487/6 (if that is the right date: see Raubitschek, cited n. 3 above).

³⁶ An Archias on one of the new ostraca may or may not be identical.

³⁷ Willemsen, l.c. This would make him an ancestor of the man attested in 306/5 (PA 3865).

³⁸ My information on the new ostraca was, at the time of writing, based on private communications from friends, supplementing Willemsen, op. cit. I regret to say that Professor Willemsen did not reply to enquiries. Fortunately we now have E. Vanderpool’s delightful Semple Lectures on Ostracism in Athens (1970), fully illustrating all aspects of ostracism, including the one here in point.

³⁹ See the list in Hands, JHS lxxix (1959), 77; and cf. now Vanderpool, op. cit., pp. 17 f.

⁴⁰ I have no information on his date. But the really massive finds (Willemsen lists 700 sherds for him) all appear to belong to the earliest period, the 480s. Thus Megacles the Alcmaeonid has over 2,200, Themistocles about 950. These are the only figures higher than Callias’. He is followed, at a long distance, by Menon of Gargettus with only 235—if indeed he has been properly identified by the editor in all cases (see n. 31). See now Vanderpool, op. cit. pp. 21 f., conjecturally identifying Callias with the unknown victim of the ostracism of 485.

⁴¹ For this man see Hdt. vi 114: the only general who died. Polyzelus and Cynegirus (Plut. Mor. 305 b) cannot be accepted as authentic. Herodotus makes it clear that Cynegirus was not a general. Indeed, he probably mentions the man only because he was Aeschylus’ brother.

⁴² Hdt. v 97.3; Hesperia, Suppl. viii (1949), 400, 411.

⁴³ See also Berve ap. Frost, op cit. 115, regarding the archonship as politically important only for noui homines like Themistocles. That view—a change from the usual—is mistaken not only in regarding Themistocles as a nouus homo, but in envisaging the possibility that such men could rise to the archonship in Athens at this time: I know of no evidence for this.

⁴⁴ But it must be pointed out that the story—common enough in the textbooks and advanced by respectable scholars: see n. 16 above—is pure fiction, with nothing in any source to support it. The date of the production of Phrynichus’ Sack of Miletus is simply unknown: Herodotus tells the story of the play and the consequent conviction in vi 21, with no chronological indication whatever. What, however, makes attribution to 493 (the very year after the capture) highly unlikely is the verb he uses of the poet, who ‘reminded’ the Athenians of the event: you can only remind people of what they may be presumed to have forgotten, and it would take longer than a few months to forget the destruction of one of the greatest cities in the Greek world, if one felt ties of kinship with it. Let us remember that plays were not necessarily topical journalism. The same author’s Phoenissae, which he may have produced for Themistocles, is put in 476 by Plutarch (Them. 5.5—if we accept the identification of the play, which seems reasonable) : i.e. four years after Salamis. The Persae of Aeschylus, dealing with that very battle, was produced four years later still. There is no justification for the conventional date assumed for the Sack of Miletus. Indeed, the assumption (suggested by Herodotus’ verb) that it was a good deal later also makes better political sense. It was surely in the early years of the Delian League, or about the time of its founding, that it might be most painful to the Athenians—and politically embarrassing as well—to be ‘reminded’ how little they had done for their oikeioi at the time. Thus we lose one of the ‘events’ of Themistocles’ archonship. The other is only a little firmer. Again Herodotus gives no date. All he tells us is that Miltiades escaped (after a close shave at Imbros) to Athens at what appears to be the time of the Persian expedition of 493 (vi 41 f.); and (in a different context: vi 104) that, after he got to Athens, he was prosecuted and acquitted, and having been acquitted was elected strategus, apparently for the Marathon campaign. The acquittal may, but need not, be put in 493/2: it is as likely to have been in the following year. The ‘facts’ of Themistocles’ archonship dissolve under scrutiny.

⁴⁵ Clearly stated Ath. pol. 8.1.

⁴⁶ Forsch. ii 149 f. (The figure 2,400 on p. 164.)

⁴⁷ On this, see (e.g.) Population Studies xiv (1960–61), 192 f., from which Duvillard’s figures, here used, are also taken. Recent research on Egypt, kindly communicated to me by Professor A. E. Samuel, shows an attrition rate many times as high there; but again, we cannot tell whether conditions were comparable. Roman statistics, based on inscriptions, are entirely misleading: see K. Hopkins, Population Studies xx (1966–7), 245–64.

⁴⁸ The minimum age of 30 does not happen to be attested for the archonship, but is certain for the boule (Hignett, Ath. Const., p. 224) and must therefore be assumed, as it universally is, for what was then a much more responsible office. Whether at this time archons were eligible for a second term (perhaps some years after the first) is again not stated in any of our sources, though the prohibition is attested for the fourth century (Hignett, op. cit, p. 228). We are reduced to arguing from probability; and it seems unlikely that Cleisthenes would not have held another archonship after the flight of Isagoras and the acceptance of his reforms, had he been entitled to do so. See further n. 52 below.

⁴⁹ To be precise: one of the ten became secretary to the thesmothetae, which must have called for further complicated arrangements right from the beginning—allotment by tribes, and just possibly rotation. But this we can only guess. (See Hammond, , CQ 19 [1969], 131,CrossRefGoogle Scholar with a just appreciation of the post-Cleisthenic Athenian political ethos in this matter.) I have ignored all these complexities, since they make no difference to the main drift of my argument. (See Ath. pol. 55. 1.)

⁵⁰ See Taylor, L.R., Roman Voting Assemblies (Ann Arbor, 1966), pp. 72 f.Google Scholar She mentions the use of hydriae for drawing lots in Athenian private life, but knows of none from public life. However, since we have no evidence at all for this early period and it is unlikely that the Classical ‘voting machine’ was developed all at once, a primitive method, in use in private life, seems perfectly conceivable in the first years of the Cleisthenic constitution.

⁵¹ See n. 30 above. Election, of course, is attested Ath. pol. 22. 5.

⁵² Hdt. v 66 f.; Ath. pol. 20. 1 (based on Hdt.). These passages seem to imply a contest for the archonship, in which Isagoras defeated Cleisthenes. But Herodotus cannot have been aware of Cleisthenes’ earlier archonship, since (as is well known) he believes that the Alcmaeonidae did not return to Athens during the tyranny. Hence his interpretation of the events he reports may well be false in this respect. It is as likely that Cleisthenes wanted the archonship, not for himself (he did not hold it after his triumph), but for a friend or relative. (See n. 48 above.)

⁵³ Hignett, , Ath. Const., p. 171.Google Scholar

⁵⁴ Griffith, G.T., in Ancient Society and Institutions (=‘Ehrenberg Studies’) (Oxford: Blackwell, 1966), p. 116.Google Scholar

⁵⁵ Buck, , CP 60 (1965), 97.Google Scholar

⁵⁶ Buck, l.c. On Idomeneus’ statement on Aristides, see p. 12 above

⁵⁷ FgrHist 338 F 4, 8, 15.

⁵⁸ Ibid. F 11.

⁵⁹ Hammond (JHS lxxxviii [1968], 28, n. 75, et al.) often insists on this point. In actual truth, the fact of these recitations (Hammond does not give the evidence) is not well attested. The early Hellenistic historian Diyllus has a tale that Herodotus received ten talents as a reward from the Athenians, presumably for flattering them in his history, which (in that case) must have been published in some form. (Cf. Kirchner, PA 1321; Jacoby, RE Suppl. ii 226, for parallel references in Hieronymus-Eusebius.) That story is absurd, not only because the mover of the motion is supposed to have been Anytus (no doubt intended to be the politician of the late fifth and early fourth centuries: the well-known enemy of Socrates could effectively become the friend of Herodotus, since the honest critic easily contrasts with the supposed flatterer), but because of the very sum involved, which is inconceivable in a Classical city state. The real basis of the story of Herodotus’ recitations is Thucydides i 22. 4, the famous passage where Thucydides admits that his work will not please the listener as much as others and claims it will be an acquisition of permanent value rather than a prize composition written for the moment: it is generally supposed that he is here pointing a contrast with his great predecessor, and though we should not assume that he actually heard Herodotus (rightly Jacoby, RE Suppl. ii 242), this conjecture is in itself reasonable enough. However, though one should perhaps not lay as much stress on this as Hammond tends to do, it is in any case almost certain that parts of his work would at least circulate in Athens and receive criticism. Oddly enough Hammond is ready to disbelieve Herodotus on this particular point and resorts to making weak excuses for him (p. 50): ‘he may have been confused in regard to a constitutional matter in which he had little interest’ (an unsound reason for this disbelief in his n. 145). For this he was rightly taken to task by Burn, who points out (JHS lxxxix [1969], 119) what this would do to the credibility of Herodotus on other matters. The point is important, and Hammond’s audience of ex-participants (or even any educated survivor of the time who saw the account) could not fail to notice an error—hammered home with such emphasis—which was not one of mere military detail. (On such details, Hammond tends to believe Herodotus—unwisely. For the uncertainty of information on battles, gathered from participants high and low, see the salutary remarks of Whatley, op. cit [n. 12], 119 f. It is amazing that impassioned ‘reconstructors’ of Marathon and other battles, while aware of that work, still proceed as if it had not been written.)

⁶⁰ Hammond, op. cit. 28.

⁶¹ Hauvette-Besnault, , Les Stratèges athéniens (1885), p. 13,Google Scholar with reference to a work by Fustel which has not been accessible to me. As we have noted: the error—if error it was—was far from inconspicuous, and ought to have obtruded itself upon the attention of anyone who knew better.

⁶² For that story see the outline given in L. R. Taylor, op. cit. (n. 50), pp. 88 f. The procedure here suggested (election to the archonship with sortition of particular spheres of duty) is, of course, precisely that of the Roman magistracy. Hammond’s note 145 (op. cit. [n. 59] ) informs me that it was proposed by Oncken and ‘may serve as a compromise’. Although his failure to give a reference for Oncken’s suggestion (not even the precise work, not to mention the page!) prevents me from checking Oncken’s arguments, I gladly cede him priority on the strength of Hammond’s word. The point is whether the suggestion (not, as far as I know, made for generations) is true. It will on no account ‘serve as a compromise’—whatever that phrase may mean in a scholarly context. Compromises are possible in politics, or even in historical assessment; but a fact of this sort, posited by a scholar, is either true or false, and to speak of compromise in this respect shows lack of appreciation of what such arguments are about. I think it can be shown to be true—or, at least, its truth can be shown to be highly probable, since something of this sort is actually needed to explain certain facts on which I have commented—and it should be believed until the opposite can be shown to be more probable. But that this cannot be done by quoting Pausanias and Idomeneus against Herodotus should have been obvious long ago.

⁶³ I cannot accept (or quite understand) Mabel Lang’s elaborate scheme in Historia viii (1959), 87 f.

⁶⁴ Whether Ath. pol. knew this, we cannot tell, and it is in any case irrelevant (despite Hammond, op. cit. [n. 59] 50, n. 145). Nor can we tell (or does it matter) whether the fourth-century Atthidographers ever discovered it. Since it was a short-lived process, superseded after a decade by one that remained in force (essentially) until their own day, it is quite possible that they could not recover it. Most probably they, like us, had only odd phrases in Herodotus (and possibly elsewhere) to deduce it from: no one ever seems to have written a detailed account of the ‘Cleisthenic constitution’ while the details were still in people’s minds. Possibly, just like us, the Atthidographers failed to fit the evidence together, or discarded information that did not agree with their theories as erroneous. (This can be shown to be so in other cases.) Herodotus must have known the facts. But he did not choose to write an analysis of the ‘Cleisthenic constitution’, and so he had no occasion to tell us, except by mentioning details as they became relevant to his story. This is precisely what he has done here.

⁶⁵ We cannot guess at the details of the process. Did all ten men draw lots together? Was the Secretary chosen first, and did the thesmothetes and the three major archons draw their lots separately? These and similar questions remain unanswerable.

⁶⁶ See, e.g., Livy xxii 41. 3.

⁶⁷ As will be seen, I accept much of Hammond’s analysis, even though I cannot accept all the details, including the construction of a concept of hegemonia that this is supposed to exemplify. There was never one concept of hegemonia as opposed to one of strategia—only different rules and relations of fact. See Appendix A.

⁶⁸ See Hammond, CQ,xix (1969), m f. His strategoi as standing officers before Cleisthenes cannot be firmly established. One man (after Cleisthenes, but long before Marathon) of whose appointment we happen to have a precise account is Melanthius (Hdt. v 97); and he is chosen, as a citizen of high standing, on an ad hoc basis; precisely what Hammond considers ‘unlikely’ (113—not referring to Melanthius, nor producing a counter-example). His only argument is an (I suspect) excessively ‘bureaucratic’ view of the Solonian state as a whole. I see nothing unlikely, in Solonian Athens, in commanders’ being appointed for overseas expeditions when needed (which was hardly ever): thus already Hauvette-Besnault, op. cit. (n. 61) 5 f.; and see Sealey, op. cit. (n. 8) 26. I consider it far more unlikely that such commanders were regularly elected, even when (as was practically always the case) there was nothing for them to do. No evidence suggests this. For home defence, the army commanded by the polemarch, no doubt with each tribe under its own commander, would suffice. Hammond’s interpretation of in Hdt. vi 109. 2 (p. 122) as referring to a period before 490 is far from proved. He arrives at this from his ‘acquaintance with the style of Herodotus’; but most other scholars, from their (sometimes equal) acquaintance with that style, have believed the opposite. And he admits that in one very obvious parallel (the Samian ships in iii 58. 2) the explanatory clearly refers to the time of the action itself. The in the phrase we are considering seems to me to prove the communis opinio right: Herodotus is explaining why there was ‘an eleventh man with a vote’, namely the one who had drawn the lot of polemarch. The explanation is, as has always been thought, that ‘formerly [i.e. at that time] the Athenians used to give the polemarch an equal vote with the strategoi’. This makes perfect sense, and just the sense required. Hammond paraphrases his own ‘explanation’ as follows: ‘and the polemarch was the eleventh voter because in early times [i.e. no longer at that time] the Athenians continually made the polemarch equal in vote to the generals’. I cannot understand the logic of that supposed statement. How does the fact that the polemarch had at one time (not then) had an equal vote with the generals explain his having an eleventh vote on that occasion? Readers will have to judge. That the usual interpretation implies that must be reduced to ‘a casting vote only’ (Hammond, I.e.) is false. There is no reason to think that Callimachus had a casting vote only, though, as it happened, he used his vote as a casting vote, after (it seems) not voting the first time round. It does not follow that he was not there the first time (as Hammond continually asserts): we have all known conscientious or cautious chairmen refraining from using their own vote to swing an important issue. As hegemon (whatever else the word implies), Callimachus must have presided at the meeting of the generals. It was only when Miltiades came to him later and used persuasion that he changed his mind about abstaining. See further Appendix A.

⁶⁹ Ath. pol. 58. 1; see Hammond, op. cit. (n. 68), 117 f.

⁷⁰ Kierdorf, , Erlebnis und Darstellung der Perserkriege (Göttingen, 1966), pp. 83 f.CrossRefGoogle Scholar

⁷¹ Hammond 118 (where, for [Thucydides] ‘1.34.1’ read ‘2.34.1’). Hammond regards the arrangements reported by Diodorus as a ‘devolution of a function’ of the polemarch, due to his no longer (after the reform of Telesinus) being ‘outstanding in the requisite qualities’. (Had the polemarch traditionally been required to be a qualified orator?) There is no sign of devolution: he is put in charge of a new ceremony of major importance. There is no reason to think that the speech over the dead, if there traditionally was one, had ever been his duty. Thucydides’ statement on the patrios nomos is demonstrably inaccurate and must be regarded with suspicion: the historian is here not speaking of what he knows, but going by tradition. (See Gomme, , Hist. Comm. 2 94 f.Google Scholar—still too unwilling to convict Thucydides of error, but aware of the problem.)

⁷² Hammond refers to the story of Cimon’s performing the ‘customary’ sacrifice to Dionysus on his return after the victory of the Eurymedon (Phit. Cim. 8) as significant in this connection. I cannot see how the polemarch comes into this story (though I admit it is very interesting in showing the rise of the strategia): we know nothing about this ‘customary’ sacrifice: not even who had ‘customarily’ performed it before the strategoi, and for how long. Pickard-Cambridge and his editor (Dramatic Festivals of Athens², p. 96) admit complete puzzlement.

⁷³ I quote this phrase from Burn, Persia and the Greeks, p. 284, precisely because Burn’s interpretation of the reform of Telesinus is by far the most sensible and moderate I have read, and I would agree with much of it. But he still overstresses it, through (I think) failure to consider the post-Cleisthenic archons and the difference that Cleisthenes may have made. It is Cleisthenes, not Telesinus, who made the ‘drastic and revolutionary’ changes in Athens.

⁷⁴ We do not know who commanded the contingents of the old (Solonian) tribes: certainly not strategoi in Hammond’s sense, of equal commanders of the whole army; though they may nevertheless have had that title, or another. On Hammond’s idea that strategoi in his sense were a standing institution in ‘Solonian’ Athens, see n. 68. It was only in 501/0, as we have seen, that the Cleisthenic tribes were first connected with the military system—whether or not Cleisthenes himself had planned this connection, we simply cannot tell. If anyone chooses to maintain that it only occurred to the Athenians after the establishment of the boule had set the civil side of the system up in working order, this (I think) cannot be refuted. But it seems to me that, whatever the old military system, it must have been based on the Solonian tribes; and it appears to have been Cleisthenes’ intention to deprive them of all but residual religious functions. (See Hignett, , Ath. Const., p. 143.)Google Scholar If this is so, then he must at least have conceived the general idea of using the new tribes for the levy, even if not the details of command.

⁷⁵ have no doubt that Hauvette-Besnault was right in thus interpreting the sources even before we had the Ath. pol. (op. cit. [n. 61] 16). That work confirms his deduction. (See also now Hammond, , JHS 88 [1968], 49 f.CrossRefGoogle Scholar

⁷⁶ See Burn, , JHS 89 (1969), 119, interpreting the details we know of the Stoa Poecile.CrossRefGoogle Scholar

⁷⁷ Ath. pol. 23. 1 f. In the state of our evidence on that period we have no right to reject this, at least in its main lines.

⁷⁸ See n. 72 above. It is not the sacrifice, but the right of judging, that is marked out as significant in our source. It was (of course) an exceptional honour, but showing a trend.