Thucydides and his Predecessors

Tim Rood (The Queen's College, Oxford)

Nonetheless, no apology is needed for making another contribution to this topic: by drawing together and examining some of the recent explorations of Thucydidean intertextuality, I hope to establish more firmly how Thucydides alluded to his predecessors; and by looking beyond the worlds of epic and Herodotus that have dominated recent discussions, I hope to present a more rounded image of the literary milieu of the early Greek historians.

The Sicilian narrative is the section of Thucydides' work on which I will focus. It is also the section that has, more than any other, been mined for intertextual allusion in recent years.[5] This intensity of focus is a reflection of the particular richness of the Sicilian narrative as a whole. But after criticizing some of these recent treatments, I will analyse in detail a passage that seems unique in the richness of its allusions to different genres: Thucydides' closing assessment of the importance of the Sicilian expedition at 7. 87. 5-6. Scholars have seen allusions in this one paragraph not just to Homer and Herodotus, but also to tragedy and to epigrams commemorating the Persian Wars. I shall examine further possible links with epigrams, and suggest that Thucydides may have been alluding to Simonides' lost poem on the battle of Salamis.

Examining Thucydides' allusions to earlier writers raises several interesting questions. Firstly, one must ask about the point of alleged allusions. When Hornblower notes 'the density of Homeric echoes in the epic Sicilian books 6 and 7',[6] he is implying that the manner and frequency of Thucydides' allusions are conditioned by Thucydides' perception of his subject-matter. (Hornblower has also pointed out how Thucydides uses an unusual number of epic words in his description of Brasidas, the 'new Achilles'.[7]) These echoes evidently strengthen the traditional view that Thucydides' account of the Sicilian expedition is marked out as epic by the breadth of his coverage and by his inclusion of such features as a catalogue of the opposing forces.[8] Other scholars have taken the further step of stressing the context of alleged allusions: Allison, for instance, suggests that Thucydides' use of a particular word may gain an ironic resonance from the use of that word in Homer. At a broader level, attempts to spot epic allusions in Thucydides raise questions about attitudes in fifth-century Athens to Homeric epic and to the 'mythical' past represented in Homeric epic, and about the status and self-definition of historiography at an early stage in its development.

I shall answer these various questions - or at least discuss the issues that they raise - by turning to explore some specific passages. I start with isolated allusions, move on to broader parallels of story-pattern, and finally return to the abundance of isolated allusions that can be detected at the very end of the Sicilian narrative.

1 'Addressing him by his father's name'

What is the significance of the echo? Hornblower simply takes the passages as evidence for a shared genealogical interest in Homer and Thucydides (and fifth-century culture more broadly).[11] Some scholars, however, stress Agamemnon's involvement in the Homeric passage. Zadorojnyi thinks that the echo is but one of a number of links between Nikias and Agamemnon (see further below),[12] while Allison sees the identification of Nikias with Agamemnon - who 'despite his bumbling, eventually succeeded' - as 'ironic'.[13] But there is an important difference between the Homeric and Thucydidean passages which tells against this 'identification': Nikias is using patronymics himself, Agamemnon is merely instructing someone else to use patronymics (rather as Xenophon instructs his troops to call on each other by name, onomasti: Anab. 6. 5. 24). The 'identification' of the two men would be more convincing if Agamemnon were himself calling on his troops in the midst of battle. (One could press for an ironic identification of Nikias and Menelaus - but Homer does not describe Menelaus carrying out Agamemnon's instructions.)

Lateiner finds the fact that a Homeric echo is attached to Nikias more revealing than the context of the Homeric passage: 'the Homeric reminiscence underlines the obsolete quality of Nicias' efforts.'[14] (Compare Connor's argument that an allusion to Aeschylus Persae 402-5 later in Nikias' speech 'emphasizes how old-fashioned Nicias' approach is'.[15]) It is true that Thucydides does say that Nikias' appeals showed that he was not 'guarding against appearing to speak in platitudes [archaiologein - lit. 'to speak in an old-fashioned way']'. But this refers to the content of Nikias' appeals rather than to the way he addresses the trierarchs. To take the Homeric echo as an indication that Nikias' rhetoric was obsolete is to underestimate the continued cultural significance of the Homeric poems in the fifth century: is Melesippos to be branded anachronistic for his remark that 'this day will be the beginning of great misfortunes for the Hellenes' (2. 12. 3) - an echo of two Iliadic passages (5. 63, 11. 604)?

To brand Nikias as anachronistic, it would be better to cite [Arist.] Ath. Pol. 21. 4: Kleisthenes 'made the men living in each deme fellow-demesmen of one another, so that they should not use their fathers' names (patrothen) and make it obvious who were the new citizens but should be named after their demes: this is why the Athenians still call themselves after their demes' (trans. P. J. Rhodes). There are, however, good reasons to doubt the historicity of this alleged attempt to stamp down on patronymics - not least the abundant evidence for their continued use.[16] Thucydides himself often names people patrothen, and never by their demotic.[17]

The suggestion of an anachronistic Nikias is further undermined by the fact that he addresses the trierarchs not just by patronymic (as Agamemnon tells Menelaus to do), but also by tribe. This is one of the few references in Thucydides to the ten tribes established at Athens by Kleisthenes in 508/7, the basis of military organization at Athens.[18] Zadorojnyi, however, does find a Homeric archetype for Nikias' appeal: Nestor's advice to Agamemnon to 'divide your men by tribes and by clans, so that clan can support clan and tribe help tribe' (Il. 2. 362-3: note kata phula and phula de phulois).[19] But 'tribes are wholly foreign to the [Homeric] poems. phule, the standard word for tribes as subdivisions of the Greek states, is not used at all, and Homer's phulon is either wider . . . or more particular'; Il. 2. 362-3 is in fact 'the only passage in Homer where phulon must have a sense something like that of the classical phule',[20] and there is no Homeric parallel at all for naming by tribe. Nikias' appeal is not a loaded Homeric allusion, but telling evidence of the emotional pull of the Kleisthenic tribes less than a century after their institution. It is significant that Thucydides alludes to the tribes elsewhere in his description of the annual public burial of the Athenian war-dead (2. 34. 3) - and that the tribes help to bring out 'the democratic or "isonomic" ("equally-sharing") character of the institution'.[21]

Nikias' appeal to the trierarchs differs in two important respects, then, from the Homeric passage alleged as a parallel: he is making an actual appeal, not enjoining someone else to make an appeal, and he appeals to people by the name of their tribe as well as by their own and their father's name. These differences, I have suggested, tell against efforts to identify Nikias with Agamemnon or to depict his appeal as old-fashioned. We should rather see the Homeric reminiscence - like the unique appeal to tribe names - as underlining the seriousness and increasing the emotional impact of Nikias' appeals.[22]

It is worth asking whether Thucydides' epic use of patrothen is found in Herodotus too: if so, this would make readings which stress the particular context in Homer even less attractive. Herodotus does use patrothen three times - and once in the very phrase patrothen onomazon (3. 1. 4 - a nicely ironic use, because the patronymic is wrong: Amasis has deceived Kambyses by sending as his future bride not his own daughter, but the daughter of the previous king of Egypt). Perhaps Thucydides' phrase is simply an epic usage mediated through Herodotus.

Herodotus' other two uses of the word patrothen cast doubt on the very idea that Thucydides' phrase is much of a Homeric reminiscence at all. The word is used of the naming of individuals in inscriptions (the monument in Samos devoted to the trierarchs who did not desert the Greek cause at the battle of Lade, 6. 14. 3) or in other forms of writing (a list of the trierarchs in the Persian fleet who were seen by Xerxes performing courageously at Salamis, 8. 90. 4: note that the list combines the patronymic with mention of their cities - that is, the personal is not at the expense of the civic). The contexts in these two Herodotean passages are close to that in Thucydides - where it is also trierarchs who are being called by their fathers' names.[23] This could be seen as another point of similarity between Thucydides' account of the final battle at Syracuse and Herodotus' account of Salamis.[24] But it is more plausible to see it simply as a sign of how commonplace the term is. Indeed, it is frequently found in inscriptions with the same verb, anagraphein, that is used in the two Herodotean passages, in instructions for names, onomata, to be written up patrothen kai tou demou or kata demous - 'by father's name and by deme'.[25]

Thucydides' use of patrothen should not be isolated from the word's use in epigraphic and other day-to-day contexts.[26] Allison, though perhaps right to call it a 'formal word', is surely wrong to suspect that it had 'an archaic ring when found in a literary context'.[27] Indeed, the fact that the word only occurs once in Homer makes it a very weak epic usage. The Thucydidean scholiast doubtless did well to remember the one passage where it does occur - but this may tell us more about education and scholarship in the centuries after Thucydides than about Thucydides himself.[28]

2 False Withdrawals: Nikias and Agamemnon

Zadorojnyi's bold analysis seems to come unstuck on some important points of difference between Agamemnon's speech and Nikias' letter. For one thing, the audiences are different: Agamemnon is addressing the suffering troops at Troy, Nikias the Athenians at home. More importantly, Nikias explicitly presents the Athenians with an alternative: 'it is necessary either to recall the troops in Sicily or to send over just as many to reinforce them' (7. 15. 1). If he is repeating Agamemnon's trick, at least he does not dissemble. Even more telling is the fact that Nikias has been presented as hostile to the expedition's aims: why should his proposal to withdraw not be sincere? It is only when the reinforcements have arrived that he changes his tune slightly (7. 48 - but note that even here he is also concerned about the practicalities of withdrawal). Zadorojnyi's reading of Nikias' motivation destroys a paradox: it is to some extent the mere fact that reinforcements have been sent that sways Nikias' view of the chances of success and increases the force of his appeal to the threat of punishment by the Athenians at home.

I prefer, then, to take the similarities Zadorojnyi detects as a sign of the realities of naval warfare and distant campaigns (ship-timber does get wet) and of Thucydides' depicting those realities in an 'epic' manner. It is, rather, Nikias' second speech in the assembly at Athens (6. 20-3) that offers a good parallel to Agamemnon's disingenuous rhetoric: just as Agamemnon tries to stiffen the Achaians' resolve to capture Troy by a meek suggestion that they withdraw, so too Nikias tries to deter the Athenians from attacking Sicily by suggesting that they increase the size of the force they are due to send.[31]

It is not just the shared rhetoric of Nikias and Agamemnon, but also the frustrated prospect of an early withdrawal, that contributes to the epic resonance of Thucydides' Sicilian narrative. The motif of an early withdrawal is prominent elsewhere in epic [32] - notably when a despondent Agamemnon repeats his proposal to withdraw, now in earnest (Il. 9. 17-28, 14. 65-81). The obvious difference is that whereas the Achaians are saved from a dishonourable retreat, the Athenians are in fact magnifying their eventual disaster. But it is the similarity which underlies this difference that in both cases increases readers' emotional engagement with the story (an engagement magnified by our knowledge of the different endings): we are encouraged to imagine how close the Achaians came to throwing away all their toil at Troy,[33] and how easily the Athenians could have avoided disaster.

The occurrence of the motif of an early withdrawal in the epic cycle may point to an enlightening parallel with another scene in Thucydides. Proclus' summary of the Cypria reports a scene in which 'Achilles restrains the Achaeans when they are eager to return home' (Arg. l. 61 Bernabé). This may bring to mind Thucydides' description of what he calls the first good deed that Alkibiades did Athens: 'when the Athenians at Samos were eager to sail against their own people [i.e. against the oligarchs at Athens], in which case it is absolutely clear that the enemy would immediately have seized Ionia and the Hellespont, he prevented it. And at that moment not one other man would have had the power to restrain the crowd' (8. 86. 4-5).[34] Unfortunately nothing is known of the context of Achilles' similar deed in the Cypria. Proclus does report it immediately after a scene in which Achilles met Helen in person, and some scholars have been tempted to see a connection between the two scenes.[35] A selfish Achilles would seem to contrast with an Alkibiades who is performing his first good service for his country. But it might further hint at the underlying self-interestedness of Alkibiades' actions. It is rash, however, to infer a link with the Helen scene from Proclus' concise summary. But it is still tempting to see an allusion to Achilles' masterful leadership - an allusion that adds some nuance to Thucydides' presentation of Alkibiades' enigmatic and destructive brilliance.

One of the grounds for withdrawal suggested in Nikias' letter further hints at how the shaping of Thucydides' narrative is rooted in epic: the claim that the Athenians are now besieged rather than besiegers (7. 11. 4).[36] This reversal (which is also found in Thucydides at 4. 29. 2) is common in later historians.[37] And it is evidently grounded in the conditions of ancient warfare - the difficulty of taking walled cities by siege and the difficulty of securing supplies. The theme goes back to the Iliad:[38] with Achilles' withdrawal from the fighting, the Trojans' successes lead to their camping out in the plain and pressing to destroy the Greek ships with fire. The parallel is particularly close when Nestor addresses the Achaians in terms more appropriate to people whose land is being attacked than to invaders: 'Be men, my friends, and put pride in your heart for the regard of others. And think, every one of you, of your children and your wives, your property and your parents' (Il. 15. 661-3). So too Nikias appeals to the Athenian army in Sicily with the thought that they are fighting for their fatherland (7. 61. 1).

3 An Odyssey in Reverse?

Mackie's thesis is certainly ingenious; and he is surely right to stress that the explicit allusions to Charybdis, the Cyclopes, and the Laistrygonians do mould the reader's impression of Sicily. But it is important to note that the allusion to the strait between Rhegion and Messene as the 'Charybdis where Odysseus is said to have sailed through' occurs only in the narrative of Athens' first intervention in Sicily (4. 24. 4); and that the connection of Corcyra with the Homeric Scheria does not occur in a Sicilian context at all. It is slightly disturbing, too, that Thucydides implies some doubt as to the historicity of the Cyclopes and the Laistrygones with a tone that has often been felt to be dismissive (he adds at 6. 2. 1 that 'I have no idea what their race was, where they came from, or where they went; what has been said by the poets and what any person may know of them must suffice'). The other detailed correspondences are also strained: they seem to diminish (rather than to draw out through contrast) the pained sense of unique and unrelenting actuality that makes Thucydides' narrative of events in Sicily so pathetic.[40]

The allusions to the world of Odysseus, I would argue, should not encourage us to draw detailed comparisons between Thucydides' narrative and the Odyssey. Instead, they define Sicily as a distant land with a mythical aura - an aura that explains the 'longing for faraway sights' felt by the young men at Athens when they voted to invade Sicily. This 'longing' (pothos) for a spectacle associated with myth recalls Xerxes' 'desire' (himeros) to see Troy (Hdt. 7. 43. 1), and anticipates Alexander's 'longing' (pothos) to see everything (but in particular places associated with the divine and with his heroic predecessors, Dionysos and Herakles).[41] And it is the contrast between the mythical projection of Sicily and the reality that is so effective.[42]

This reading of Thucydides' allusions to the Odyssey makes even less attractive Mackie's search for a detailed intertextual relationship. Similarly strained is his comparison between Alkibiades' escaping from Sicily in a single ship at Th. 6. 61. 6 and Odysseus' escaping in a single ship at Od. 10. 95-6 and 131-2.[43] One could just as easily stress differences. Odysseus escapes when all the other ships are lost, and he is the only one of his companions to return home. Alkibiades flees at an early stage in the expedition, escaping a danger from his own side; he does not seek to return home in his single ship; and at the close of the expedition some of the Athenians do return home. That the seductive habit of drawing comparisons and contrasts between patterns in epic and patterns in Thucydides runs the danger of proving rather facile is suggested by the fact that Frangoulidis is able to point to an allegedly ironic contrast between Odysseus and two other Athenian generals in Sicily, Demosthenes and Nikias (Odysseus survives, the Athenian generals die).[44]

What has emerged is the need for caution in the search for Homeric archetypes for scenes and phrases, and the danger that claims of implausibly detailed correspondences may in fact distract attention from the subtlety of Thucydides' exploitation of his literary predecessors. I turn away now from the narrative as such to Thucydides' closing assessment of the Sicilian expedition. I shall argue that this passage has a range of register and allusion that adds depth to Thucydides' contextualization of the Sicilian expedition. I shall start by pursuing further the theme of epic reminiscence, and by looking at how Thucydides evokes the return home that eluded most of the Athenian force.

4 'And this Hellenic event . . .': Thucydides 7. 87. 5-6

'And this Hellenic event turned out to be the greatest connected with this war and, at least in my opinion, of Hellenic events we have heard of, the most splendid for those who won and the most wretched for those who were ruined. For after having been completely defeated in every respect and suffering no little misery at every point in, as the saying is, total destruction, army and navy, nothing was not lost, and few out of many returned home. This was what happened concerning Sicily.' (7. 87. 5-6)

4. 1 Epic: 'Few out of many returned home'

Allison takes the further step of arguing that the contexts of Homer's six uses of the verb aponostein help us to appreciate Thucydides' own use of the word. The word is used once in Homer of a successful return (Od. 13. 4: of Odysseus): Allison argues that here 'the intertextual message of the History is gloomy: the Athenians will not realize an Odyssean nostos'. In the other five Homeric passages she sees a 'tension between the hope of return and the reality of impending death'; where the verb is used of Patroklos' failure to return to the Greek camp (Il. 17. 406), for instance, she argues that 'it is the foreboding sense of imminent loss . . . now predicted by Achilles, that Thucydides seems to impart into his account'.[48]

It is hard to feel satisfied with Allison's 'no-lose' model of literary allusion: if Homer uses aponostein of a successful return, then Thucydides is drawing a contrast; if Homer uses it of an unsuccessful return, then Thucydides is suggesting a parallel. And further objections can be made: at Il. 17. 406, for instance, there is no 'foreboding sense of imminent loss' in Homer; rather, Homer is drawing out the pathos of the contrast between Achilles' expectation that Patroklos would return and the reality that Patroklos is already dead.[49] So too there is no 'foreboding sense of imminent loss' in Thucydides: the main destruction of Athenian troops has already occurred; the men who return home are 'few' out of the 'many' who set off from the Piraeus, not few out of the men who actually left Sicily or even few out of those who were sold into slavery in Sicily. A further problem with seeking explicit allusion is that three of the Homeric uses of aponostein refer not (as is required for the parallel with the Athenians in Sicily to hold) to the return of the Greeks from Troy, but to the return of warriors from the battlefield to the city of Troy (Il. 8. 499, 12. 115) or to the Greek camp (Il. 17. 406).

Even if one allows that Thucydides is alluding to nostos as a general theme, the allusion does not seem primarily to be to the Odyssey or to the poem in the epic cycle called Nostoi. Those poems described the difficulties experienced by some of the Achaian leaders at Troy during their return journeys across the sea, and after their arrival back home; the Nostoi also related how some of the heroes returned by land. Thucydides does not say anything about how the survivors from the Athenian force returned to Athens. In the Anabasis, by contrast, Xenophon does implicitly assimilate the return of the Ten Thousand to Greece to the return of Odysseus to Ithaca (see the references to the Lotus-eaters at 3. 2. 25 and to Odysseus at 5. 1. 2); but he only uses nostos-words once, to refer to a report that no one at all returned from a Persian army which attacked the mountain-dwelling Kardouchoi (3. 5. 16, ouden' aponostesai).

Though it seems unhelpful to see Thucydides' use of apenostesan as alluding either to specific Homeric uses of aponostein or to the genre of epic nostoi, we should not dismiss the Homeric resonance altogether. At one level, it reinforces the sense created by Thucydides' other epic words that the Athenian expedition, and Thucydides' representation of it, are epic in scale and ambition.[50] More particularly, it evokes the emotive force that the idea of nostos has in epic: the bitterness of separation from home, the longing for the day of return, the nostimon emar, and the joy when that day arrives. Thucydides' evocation of the nostos theme seems the more poignant because the Athenians had been so sure they would come home in safety (6. 24. 3), and because they (or at least most of them) do not just fail to return, they do not even achieve anything to compensate for their loss of nostos. In the Iliad, by contrast, the idea of return is regularly linked with the sacking of Troy: what Agamemnon (and Menelaos) have been promised by the gods, and by their troops, is that they will return after sacking well-walled Troy (Ilion ekpersant' euteicheon aponeesthai: Il. 2. 113, 288, 5. 716, 9. 20).[51]

A brief look at Herodotus' more plentiful uses of the verb aponostein [52] will offer an interesting comparison with Thucydides' technique of epic allusion. Herodotus evokes the epic associations of the nostos theme in a broader, but equally general way. At one point, for instance, he describes the sufferings undergone by some of the heroes from Troy after their return home: 'the Cretans proved themselves by no means the most despicable champions of Menelaus; their reward for this service on their return home (aponostesasi) was famine and plague for both men and cattle' (7. 171. 2); this recalls one strand in the epic nostoi (see e.g. Nostoi, Arg. ll. 17-18 Bernabé for the murder of Agamemnon). Particularly rich in its evocation of this strand is Herodotus' account of the fate of Pheretima, ruler of Cyrene: 'No sooner had she returned to Egypt after her revenge upon the people of Barca, than she died a horrible death . . . So true it is that excess in taking revenge draws down on humans the anger of the gods' (4. 205). In epic, too, it seems that the excesses committed by the Achaians in the sack of Troy (notably Aias' rape of Kassandra and Neoptolemos' killing of Priam at the altar of Zeus) were responsible for their sufferings during and after their return home. Another sign of Herodotus' closeness to epic is that he does recount the sufferings of the Persians on their return home; Thucydides, as we saw, glosses over the details of the Athenians' return from Sicily.[53] Again, Herodotus' pathetic focus on the clash of expectation and reality in the story of Adrastus and Atys seems close to the nostos theme as it is elaborated in the Iliad (the main epic model, I have argued, on which Thucydides drew): Adrastos promises Croesus that his son Atys will return home safe from the boar hunt (aponostesein - last word of the speech, 1. 42. 2) - when it will be Adrastos himself who unwittingly kills him. At a more general level, it is the expected delight of the soldier's nostos that makes more startling the two stories in Herodotus which illuminate the typically Spartan shame of survival (1. 82, the suicide of the sole Spartan survivor of the Battle of the Champions; cf. 7. 229, on the two Spartans who miss the battle of Thermopylai).

While Herodotus' canvas has more room than Thucydides' for the full range of association that the epic theme of nostos conveys, it would evidently be as rash to claim for Herodotus as it would for Thucydides that the force of his nostos words ever depends on the reader's recollection of any single passage. Indeed, if we are to claim a narrower range of allusion for Thucydides' use of apenostesan, it is in Herodotus, not in epic, that this should be sought. Herodotus uses the word several times to express how only a small number of an original group survived (6. 27. 2: 'the Chians sent a chorus of one hundred young men to Delphi; only two of them returned home' - note the pathetic juxtaposition of hekaton duo; 6. 92. 3: out of 1,000 Argive volunteers who went to Aigina, 'most did not return home' (hoi pleones ouk apenostesan)). And once his expression is almost exactly the same as Thucydides' (4. 159. 6 : 'few of them returned home to Egypt', oligoi tines auton apenostesan es Aigupton): all this is lacking here is the explicit apo pollon. So while Herodotus draws on epic, Thucydides draws not just on epic, but also on epic as its themes are reflected in, and transformed by, Herodotus' History.[54]

Both Herodotus and Thucydides, moreover, seem to be using a mode of rhetoric already found in tragedy: 'All who survived and won to safety, when they had made their way through Thrace, as they best could, with grievous hardships, escaped and reached - and few they were indeed (ou polloi tines) - the land of hearth and home; so that the city of the Persians may well make lament in regret for the best beloved youth of the land' (Aesch. Pers. 508-12, trans. Weir Smyth). Since there are other echoes of Aeschylus' depiction of Salamis in the Persae in the closing sections of the Sicilian narrative, Thucydides' oligoi apo pollon can plausibly be read as an allusion to this very passage of the Persae. These allusions reinforce the broader pattern of allusions to Herodotus' depiction of the Persian Wars: I turn now to explore how that pattern is itself reinforced by Herodotean elements in Thucydides' grand closure to the Sicilian narrative.

4. 2 Herodotus

What is the force of the phrase at 7. 87. 5? Marincola reasonably states that 'the enormity of the claim probably motivates the qualification'.[56] But we must also consider the phrase in the light of other Herodotean phrases in their context.

Thucydides' qualification that the Sicilian expedition was 'the greatest . . . of Hellenic events that we have know of by report' (hon akoe(i) . . . ismen) also has a Herodotean ring. Herodotus often restricts superlative clauses by adding ton hemeis idmen: Pausanias at the battle of Plataia 'won the most splendid victory of all that we know of' (niken anaireetai kallisten apaseon ton hemeis idmen, 9. 64. 1[57]); the losses suffered by the peoples of Tarentum and Rhegion against the Iapygians of Messapia were 'the greatest Hellenic slaughter of all that we know of' (phonos Hellenikos megistos houtos de egeneto panton ton hemeis idmen, 7. 170. 3 [58]). As with dokei(n) (e)moi, it is in the Archaeology that most of Thucydides' uses of the Herodotean formula '[first/greatest] of those that we know of' are found (1. 4. 1, 13. 4, 18. 1: only the first of these has 'by report', akoe(i)).[59] The difference is that Herodotus does not combine the phrase with akoe(i): his claims presumably have a slightly smaller chronological reach than Thucydides'.[60] Herodotus does, however, offer a similar restriction to his claim that 'of all the armies we know of, [Xerxes' army that invaded Greece] was by far the largest': 'The army with which Darius attacked Scythia was tiny by comparison; so was the Scythian army which invaded Media on the heels of a Cimmerian force and succeeded in conquering and occupying almost the whole of inland Asia . . .; and so, by all accounts (kata ta legomena), was the army with which the Atreidai attacked Ilion, and the army the Mysians and Teukrians raised - this was before the Trojan War - with which they crossed the Bosporus . . .' (7. 20. 2 (trans. Waterfield) - though note that he concludes with an undiluted claim at 7. 21. 1).

Both dokein d' emoige and hon akoe(i) . . . ismen, it could be objected, are standard expressions for toning down authorial claims. The important point, however, is that it is only with regard to the Sicilian expedition that Thucydides uses these phrases to extend a claim about the uniqueness of an event in the Peloponnesian War to cover previous wars too. And he makes this extension in the same competitive spirit shown by Herodotus at 7. 20. 2. The expeditions which Herodotus compares with Xerxes' all involve crossings from Asia to Europe or from Europe to Asia. Thucydides adopts the same manner as Herodotus while shifting the focus of the claim to greatness from conflicts between Asia and Europe to a conflict of Greek against Greek. At the same time he uses as a criterion of greatness not size, as Herodotus had done, but the destruction done to the invading force and the glory won by the defenders.

Thucydides also lays a direct stress on the Greekness of the Sicilian expedition: 'this Hellenic event turned out to be the greatest connected with this war and, at least in my opinion, of Hellenic events'. The repetition of 'Hellenic' (ergon touto Hellenikon, hon akoe(i) Hellenikon ismen) has, however, been found strained. Arnold found the first Hellenikon 'unnecessary': 'for what great events took place in the Peloponnesian war in which Greeks were not the principal actors?'[61] Krüger first proposed deleting it, and almost all editors have followed him (Dover finds it 'stylistically objectionable'[62]). Steup is the exception:[63] he argues that there were some important activities undertaken by barbarians during the twenty-seven years of the Peloponnesian War (the expedition of Sitalkes, the Carthaginian invasions of Sicily) which could have prompted Thucydides to restrict his claim by adding Hellenikon.[64] But Thucydides' claim surely applies to the events, rather than to the period, of the Peloponnesian War.[65] And we may, in any case, doubt whether Thucydides would have felt compelled by the Carthaginian invasions to impose the apparent restriction he makes on the greatness claimed for the Athenian invasion.

'The repetition if authentic is obsessive.'[66] And it is perhaps the nature of Thucydides' obsessiveness which guarantees the authenticity of the repetition. It is not just that Thucydides uses Hellenikos several times in superlative phrases elsewhere to underline the Greekness of his war (5. 60. 3, 74. 1, 7. 75. 7; cf. the stress on Greekness in superlatives at 1. 50. 2 (quoted n. 67 below) and 138. 6; and cf. also Hdt. 7. 170. 3, quoted above). It is also that the obsessive repetition sets the great combat of Greek against Greek against earlier clashes of Greek and Persian.[67]

The Herodotean feel of Thucydides' use of 'Hellenic' seems to be confirmed by the Herodotean feel of the phrase it accompanies - 'greatest event' (megiston ergon). This form of commemoration is typical of Herodotus. Yet Immerwahr has claimed that Thucydides differs from Herodotus in his use of ergon: Thucydides, for instance, rejected Herodotus' concept of monuments as erga which are, as much as deeds, a criterion of greatness.[68] The problem comes with Immerwahr's further claim that Herodotus saw the two types of erga - deeds and monuments - as parallel: 'the conception of fame underlying both monuments and deeds is exactly the same'; 'Herodotus looks at a deed as if it were a monument'[69] - that is, he analysed the effects of deeds not in terms of their direct historical consequences but in terms of honour and fame. Immerwahr claims that in Thucydides, by contrast, 'ergon almost always refers to an activity rather than an achievement, and to a fact rather than a deed'. Especially relevant for 7. 87. 5 is his further claim that in Thucydides 'a mega ergon means simply "great trouble", or "an important event", and never "a great deed" in the Herodotean sense'.[70]

The distinction between Herodotus and Thucydides drawn by Immerwahr, though not without justification,[71] is too sharp. There are passages where Thucydides uses ergon in its 'Herodotean' sense (e.g. 2. 81. 4, where the Chaonians rush ahead of the Spartans in the advance against Stratos, 'thinking they would take the city at one blow and the deed (ergon) would be all theirs'). When Thucydides calls the Sicilian expedition the greatest ergon of the war, and perhaps of Greek history, he is not just referring to its historical importance: as the phrase 'most splendid (lamprotaton) for those who won' shows, he is thinking in terms of fame and glory. He is echoing, indeed, the assessment of the Syracusan speaker Hermokrates ('whether we go on to defeat them or send them away without achieving their objective . . . the outcome will provide us with a most glorious deed (kalliston de ergon)', 6. 33. 4), and the basis for that assessment - the fame the Athenians had won by defeating the Persians.[72]

I conclude that Thucydides' closing assessment of the Sicilian expedition has the sort of commemorative tone that we usually associate with Herodotus; that this tone is conveyed by phrases that we also associate with Herodotus; and that this Herodotean manner is itself part of the way in which Thucydides sets the greatness of the Sicilian expedition against the greatness of Xerxes' expedition.[73] I shall argue below, indeed, that the language of ergon . . . lamprotaton evokes the epigrams set up by the Greeks to commemorate their victory over Persia. It is enough here to conclude that Thucydides does regard the Sicilian expedition as the most important event of the war in historical terms, but that he also regards it as a source of great fame for the Syracusans.

It remains to consider the possibility that a word Thucydides uses later in our passage is a direct allusion to a specific passage in Herodotus. Thucydides calls the Athenian defeat 'as the saying is, total destruction' (panolethria(i) de to legomenon); Herodotus had described how Troy was doomed - 'the gods were arranging things so that in their total destruction (panolethrie(i)) the Trojans might make it completely clear to others that the severity of a crime is matched by the severity of the ensuing punishment at the gods' hands' (2. 120. 5, trans. Waterfield). The echo (which was already noted in the eighteenth century [74]) has been ignored by most commentators on Thucydides. What has more recently been claimed is that Thucydides is not just alluding to the Herodotean passage, but also injecting Herodotus' theodicy into his account of the Athenians' destruction in Sicily.[75]

How plausible is the allusion to Herodotus, the basis for the startling theological reading of Thucydides? The passage of Herodotus is the only place where the noun panolethria is found before Thucydides. But the argument from silence is dangerous when so much early literature is lost: Hornblower has raised the possibility that there was a common source in epic for panolethria.[76] And need Thucydides' use of to legomenon imply a specific allusion to the noun panolethria - as opposed, for instance, to the adjective panolethros? That adjective is common in tragedy;[77] it also appears in a mock-tragic passage at Ar. Birds 1239. In any case, to legomenon ('as the saying is') can plausibly be taken as an argument against the Herodotean allusion: it should flag a conventional phrase rather than a specific passage.[78] There is also a negative argument: elsewhere Thucydides does echo specific passages in Herodotus without signalling the echo explicitly.

It seems possible, nonetheless, that Thucydides' use of to legomenon signals the commonness of the phrase without removing the chance of a specific allusion. There is a possible parallel in Xenophon's use of the verb dekateuein ('tithe') as a euphemism for 'destroy'. The word is twice used by speakers in the Hellenica, both times with the phrase to legomenon, alluding to an oath to tithe Thebes allegedly made during the Persian Wars.[79] Herodotus tells of an oath made at Thermopylai to 'tithe' medizing states (7. 132. 2). The phrase is also found in fourth-century versions which present an oath at Plataia (see Tod ii. 204; Lyc. 1. 81; cf. Diod. 11. 29. 2): historians argue that it took on new life after 374, when the Spartans and Athenians began co-operating against Thebes, and when Thebes' medism could be resented all the more because of Theban agitation for the destruction of Athens in 404.[80] So the association with the Persian Wars is certain: but how important it is for our interpretation of Xenophon that it occurs in the work of an earlier historian is far harder to determine.

With Thucydides' apparent allusion to Herodotus we are on slightly surer ground. But the question of the significance of the allusion remains open. The theological interpretation suggested for the destruction of the Athenian force in Sicily would be without parallel in the History. If it is right to stress the specific allusion to Herodotus' theodicy, it would be more attractive to see Thucydides as commenting sardonically on the failure of the divine plan that the Trojans should 'make it completely clear to others that the severity of a crime is matched by the severity of the ensuing punishment at the gods' hands'. But there is another way of reading the allusion that is also attractive: 'Thucydides' choice of words proclaims the parity of his subject, and his treatment of it, with both Herodotos and "Homeric" epic.'[81] This reading is perhaps supported not just by the epic features in Thucydides discussed above, but also by Thucydides' broader patterning of the Sicilian disaster after the Persian invasion of Greece: he could easily have been swayed by the way that the Persian invasion was often linked with the Trojan War in imaginative constructions.[82] The specific allusion to the fall of Troy is all the more portentous if we recall Troy's status in tragedy as 'a permanent reminder of the fact that cities are mortal'.[83] Perhaps the allusion reinforces Thucydides' repeated suggestions that Athens' defeat in Sicily foreshadows the fall of the city in 404.[84]

This interpretation of Athens' 'total destruction' appears to be supported by the stress on the mortality of cities in later historians - notably in Appian's (and perhaps Polybius') account of the destruction of Carthage:

'Scipio, when he looked upon the city as it was utterly perishing and in the last throes of complete destruction (panolethrian eschaten), is said to have shed tears and wept openly for his enemies. After being wrapped in thought for long, and realizing that all cities, nations, and authorities must, like men, meet their doom; that this happened to [85] Troy, once a prosperous city, to the empires of Assyria, Media, and Persia, the greatest of their time, and to Macedonia itself, the brilliance of which was so recent, . . . he said: 'A day will come when sacred Troy shall perish, and Priam and his people shall be slain' [Il. 6. 448-9]. And when Polybius . . . asked him what he meant by the words, they say that without any attempt at concealment he named his own country, for which he feared when he reflected on the fate of all things human.' (Appian, Punica 132 = Polyb. 38. 22. 1, trans. Paton)

Thucydides, I conclude, is dwelling on destruction rather than on the divine. Indeed, even if it is felt (reasonably enough) that the commonness of the adjective panolethros in tragedy militates against the interpretation I have proposed, and that Thucydides is simply evoking the atmosphere of tragedy, it is still on destruction that he is dwelling.

4. 3 'And the most wretched for those who were ruined': Tragedy

Dover rightly stresses that Thucydides uses tragic vocabulary as well as a tragic metre: words with the root dustuch- are 'abundant in Euripides but not so common in prose'.[88] Indeed, Thucydides' dustuchestaton picks up his comment that the Athenian general Nikias was 'the least worthy (axios) of the Greeks of my day to come to such misfortune (dustuchias)' (7. 86. 5): a tragic refrain.[89]

Thucydides' specific evocation of the form of Athenian tragedy - as well as his broader construction of the Sicilian expedition along tragic lines - does not mean that the expedition is in a closed sense a 'tragedy'.[90] Scholars are right to point out that Thucydides' perception of the significance of the Athenian disaster in Sicily must reflect the broader perception of contemporaries in Athens.[91] Even so, there is always more than one story to tell - and not just because there is no natural way of separating off beginnings, middles, and ends, but also because there are different ways of reading stories with the same beginning and end. The event that is for the Athenians 'most wretched' is for the Syracusans 'most splendid' (lamprotaton).

The splendour of the Syracusans' achievement also adds to our sense of the Athenians' tragedy. Thucydides had described the Athenian force when it set off from the Piraeus as 'celebrated for its splendour (lamproteti) to look upon' (6. 31. 6); and he had noted when it set off in retreat from Syracuse how 'from such splendour (lamprotetos) and vaunting at first they had reached such an end in humiliation' (7. 75. 6). The shift of focus from the Athenians' physical splendour to the Syracusans' symbolic splendour furthers Thucydides' suggestion that the Athenian defeat in Sicily is a reversal; and this pattern of reversal is something that we associate with tragedy. The suggestion of reversal would be even stronger if Thucydides' phrase ergon . . . lamprotaton were an allusion to Simonides' description of the battle of Salamis - a (slight) possibility that I shall now raise.

4. 4 'The most splendid event': Simonides and Salamis

Then the rest, put on an equality in numbers with their foes, because the barbarians had to attack them by detachments in the narrow strait and so ran foul of one another, routed them, though they resisted till the evening drew on, as Simonides says, bearing away that fair and notorious victory, than which there has been performed upon the sea no deed more splendid (hes outh' Hellesin oute barbarois enalion ergon eirgastai lamproteron), either by Greeks or barbarians, through the manly valour and common ardour of all who fought their ships, but through the clever judgement of Themistokles. (15. 4, trans. adapted from Perrin)

How are we to interpret these links? They could be regarded simply as a sign that Plutarch shared Thucydides' taste for amplification, and a sign of the frequency of certain motifs (land and sea, Greek and barbarian) in such rhetorical contexts. More ambitiously, they could be taken as a sign that Plutarch had picked up Thucydides' modelling of the Sicilian expedition on the pattern of the Persian Wars. Plutarch would then be retrospectively anticipating Thucydides (so to speak) by applying to Salamis the claims Thucydides had made for the Sicilian expedition.[92] It would be more interesting for our interpretation of Thucydides if Plutarch's assessment were derived from the source he mentions: Simonides' poem on the battle of Salamis. In that case we could say that Thucydides boosted a pattern he had already established (the modelling of the Sicilian expedition on the Persian invasion of Greece) by a forceful closural allusion to a famous poetic treatment of the Persian Wars.

To determine whether Thucydides could have been alluding to Simonides' poem on Salamis, we must first try to determine the scope and precision of Plutarch's citation in the Themistokles passage. An immediate problem is that scarcely anything is known about Simonides' poem: there is uncertainty about its form; and the passage of Plutarch quoted above is in fact the only certain citation.[93] Translators and editors of Plutarch have tended to view the citation as extensive: unmodified, Perrin's translation reads 'though they resisted till the evening drew on, and thus "bore away", as Simonides says, "that fair and notorious victory . . . through the clever judgment of Themistocles."'[94] The optimism of Plutarch's translators has been shared by various scholars: the passage of Plutarch has been taken as evidence that Themistokles' role at Salamis was seen as decisive as early as Simonides;[95] as a sign of a friendly - and perhaps even politically fruitful - relationship between Themistokles and Simonides;[96] and as evidence that Simonides (unlike Aeschylus in the Persae) mentioned Themistokles by name.[97] Editors of Simonides have been more sceptical: West, for instance, prints the whole paragraph of Plutarch in the latest edition of Iambi et Elegi Graeci (1992), but in earlier editions he cut it off at 'more splendid' (lamproteron) and at 'as Simonides says'.[98]

Two things about Plutarch's citation are at least clear: it has no discernible metre, and it is not presented as precise (in the de Herodoti Malignitate, by contrast, he introduces verbatim quotations of Simonidean epigrams with tade [99]). Modern translators who use quotation marks are misleading. But scholars have still been tempted to see some of Plutarch's phrases as accurate quotations: 'deed upon the sea' (enalion ergon) [100] or 'deed more splendid' (ergon lamproteron), for instance.[101] These interpretations would support the view that Thucydides may have been imitating Simonides. But perhaps the most common view among students of lyric is that only the phrase 'holding out until evening' (mechri deiles antischontas) is Simonidean;[102] this view is supported by its placement directly before the phrase 'as Simonides says'.

One way of trying to establish more clearly the scope of Plutarch's citation is to consider the point of the citation. Doubtless translators have extended its scope because the mere detail 'holding out until evening' seems so banal. The grander claim about the greatness of Salamis seems more memorable, more worthy to be quoted. Aeschylus, too, presented Salamis as a battle that lasted 'until the eye of dark night took away [the slaughter]' (Pers. 428).[103] And battles lasting until evening are common in epic and in historiography.[104] Perhaps, then, it was the Persians' 'holding out' that Plutarch found significant; their resistance doubtless helps to justify the greatness claimed for the battle. In Aeschylus, by contrast, the Persians did resist at first, but what continued until night was not their resistance but their slaughter; the stress falls (from the viewpoint of the defeated) on the magnitude of the slaughter ('there never perished in a single day so great a multitude of men', 430-1). Herodotus, on the other hand, does say that the Persians fought well (8. 86), but not how long the battle continued.

Plutarch's citation is effective, therefore, even if it only covers the phrase 'holding out until evening'. He seems, too, to have been fond of quoting the great Simonides;[105] indeed, as Christopher Pelling suggests to me, his mere act of signalling that there was a poem on the battle of Salamis by Simonides promotes the glory of the battle. But there are reasons (other than the claim itself) why Plutarch might have cited Simonides for the grander claim, and for the role of Themistokles. The relationship of Themistokles and Simonides is noted by Plutarch elsewhere in the Themistokles. He relates two anecdotes in which they converse, and cites a poem of Simonides [106] which told of Themistokles' restoration and dedication of the telesterion (initiation-house) of the Lykomids at Phlya (Them. 1. 4: perhaps wrongly, he took this as evidence that Themistokles himself belonged to the Lykomid family). If the citation from Simonides does extend to the end of the paragraph, then it would be tempting to explain it through Simonides' link with Themistokles (and perhaps also through Themistokles' absence from Aeschylus' Persae).

A further Themistoklean link has been suggested as a basis for establishing the extent of Plutarch's use of Simonides. Plutarch relates that Lykomedes was the first Athenian to capture an enemy ship at Salamis, and that he dedicated the ensign to Apollo at Phlya (Them. 15. 3). Both the name Lykomedes and the association with Phlya suggest a connection with the Lykomid genos. We know that Simonides wrote about Themistocles' connection with the Lykomids: it has been suggested that Simonides was Plutarch's source for the detail about Lykomedes' feat at Salamis too.[107] But Herodotus reports that Lykomedes was the first Athenian to capture a ship at Artemision (8. 11. 2): 'it is easier to explain [Plutarch's] statement as a simple confusion of the two great battles rather than as a separate tradition based ultimately on Simonides.'[108]

It remains to consider whether there is anything distinctive about Plutarch's vocabulary which can help to determine his use of Simonides. Some words are certainly not poetic: in the phrase 'bearing away that fair and notorious victory', the word periboetos [109] is prosaic (it is found 40 times in Plutarch's Lives, and 59 times in his oeuvre as a whole). Indeed, the expression 'that fair and notorious victory' (ten kalen ekeinen kai periboeton) itself has the mark of a distant perspective (cf. Plut. Kim. 13. 4, 'that notorious peace', ten periboeton eirenen ekeinen). [110] But this distant perspective could be taken as a nod not just to a famous victory, but also to a description of that victory by a famous poet: that is, it might tell for, rather than against, the view that the quotation from Simonides is resumed in the clauses that follow. The phrase 'bearing away victory' (aramenoi niken) is quite high language (cf. Bacchylides 2. 5), but not exclusively so (cf. Strabo 3. 2. 13). The adjective enalion ('upon the sea') is poetic, but (as LSJ note) it is used in later prose (Plutarch uses it at Luc. 39. 3 and four times in the Moralia). What about lampros - the word common to Plutarch and Thucydides? It is commonly used for the brilliant light of the sun and stars, and commonly extended to brilliant deeds in celebratory poems.[111] But it is equally common in prose;[112] it is worth noting that Plutarch applies the epithets lampros and megas to the former deeds (erga) of Gylippos, Spartan commander in Sicily ('after adding a deed so disgraceful and ignoble as this to his previous great and splendid achievements', Lys. 17. 1: perhaps an echo of Th. 7. 87. 5, but cf. also Artax. 15. 4, 24. 9, Alex. 5. 4, for the combination).

While it is best to be cautious in reconstructing Simonides' poem, there are several reasons why it is at least worth considering the possibility of such an allusion. One reason is the light cast on the development of Greek historiography by the recent publication of Simonides' elegy on the battle of Plataia: the apparent suggestion in that poem of a continuity between the great deeds of epic and the Greek performance at Plataia is interesting in view of the more distant and competitive attitude to the Trojan War in Herodotus.[113] Doubtless Herodotus' account of Salamis - as well, perhaps, as Thucydides' account of Syracuse - would be equally illuminated by the discovery of Simonides' poem on the battle of Salamis. Plutarch's account strongly suggests that it would have contained the motif of ships confined in the narrows that is the principal link between Salamis and Syracuse. There may be other, more surprising, echoes: Stephen Harrison has noted that both Herodotus (8. 77) and Aeschylus (Pers. 420) stress the bloodying of the water at Salamis, and speculated that Horace's description of the bloodying of water during the sea-battles in the Roman civil war (Carm. 2. 1. 33-6) may be an allusion to Simonides' use of this motif in his poem on Salamis (the speculation is based on the reference to Simonides in the phrase 'Ceae . . . neniae' at line 38).[114] This is relevant because Thucydides too uses the motif - but he applies it not to the final battle at Syracuse, but to the final slaughter of thirsty Athenian troops in the river Assinaros.

To return to the 'most splendid event': either both Thucydides and Simonides/Plutarch are using a shared encomiastic vocabulary or there is a specific echo (whether it is Thucydides echoing Simonides or Plutarch echoing Thucydides). This problem returns when we consider the possibility that Thucydides was alluding to another possible sphere of literature - the epigrams inscribed on monuments commemorating the Greek success against Persia. And this further possibility will suggest that it does not matter that the problem is intractable: Thucydides can still be recognized as using language that recalls celebrations of the Persian Wars even if he is not held to be evoking specific examples of that language (so too with Xenophon's use of kalliston ergon at Anab. 6. 3. 17, where the stress on shared safety reproduces two other motifs familiar from Persian Wars contexts). The important thing is that he is using that language in the context of Athens' defeat in Sicily, and to suggest an inversion of the Persian Wars.

4. 5 Persian Wars Epigrams

Since the time when the sea first separated Europe from Asia and wild Ares controlled the cities of mortals, no such deed of earthly men was ever carried out on land and sea at the same time: these men destroyed many Medes on Cyprus and then on the sea captured a hundred ships of the Phoenicians with their full complement of men; and Asia groaned loudly when struck with both hands by them with the strength of war. (trans. D. A. Campbell)

Thucydides' dichotomy of land and sea is complicated by his presentation of the battles in the harbour at Syracuse as sea battles that are like land battles. This blurring is not just a way of linking Sicily with the fighting at Pylos, where the same confusion occurs; it also contrasts the confused fighting of the Peloponnesian War with the clear-cut battles of the Persian Wars - 'the greatest action of the past, yet it had a quick resolution in two battles on sea and two on land' (1. 23. 1)[121] - and of later stages in the fighting against Persia (the battles of Eurymedon and Cyprian Salamis, both battles on land and sea on the same days: 1. 100. 1, 112. 4). (Note how Thucydides consistently ignores the fighting on the island of Psyttaleia in his presentations of Salamis purely as a sea battle; Aeschylus, by contrast, links the two parts of the battle by having the men who have fought in the sea battle go on to slaughter the Persian nobles on the island.[122] Nor does he just simplify Salamis to increase the contrast between the Persian and Peloponnesian Wars: he further blurs the distinction of land and sea fighting in Sicily by postponing the motif of the bloodying of water - which, I have suggested, we might have expected in his account of the final sea battle at Syracuse - until his description of the slaughter of Athenian troops at the Assinaros.)

My suggestion that the polar expression 'land-force and ships' (kai pezos kai nees) at 7. 87. 6 recalls depictions of the Greek victory over Persia as a victory on land and sea is strengthened not just by Thucydides' own depiction of the Persian Wars in terms of the land/sea dichotomy, but also by the agonistic spirit that seems to bind the various Persian Wars epigrams. We have seen that the Eurymedon - or Cyprus - epigram quoted by Diodorus included the claim that 'no such deed of earthly men was ever carried out': Wade-Gery took this as a sign that 'the poem deliberately invited comparison with Mykale';[123] indeed, he took the phrase as one reason why the epigram had to refer to Eurymedon (if it referred to the victory at Cyprian Salamis, the precedent of Eurymedon would make the claim invalid). Gomme, by contrast, took the boastfulness of 'its implied comparison with Salamis and Plataia' as an argument against Wade-Gery's view that the epigram dates from the time of Eurymedon (he aptly compared Plut. Kimon 13. 3: 'though he had surpassed the victory at Salamis with an infantry battle and that of Plataea with a naval battle', trans. Perrin).[124] Besides this, the use of the phrase pezoi te kai okuporon epi neon in both the Salamis (or Persian Wars) and the Eurymedon epigram, while doubtless a sign of the repetitiveness of epigrammatic language, could also be taken as another reflection of the tendency to present one battle in the light of another.[125] And the Salamis epigram itself has been (implausibly) linked with a 'battle of the battles': the addition to the monument on which the Salamis epigram was inscribed of an epigram which may refer to Marathon has been taken as the reaction of the Kimonian, hoplite, faction against the democratic exploitation of Salamis.[126]

The grand heightening which Thucydides gives to the Sicilian expedition at its close belongs in this tradition. But the effect of his evoking the past of the Persian (and perhaps also the Trojan) Wars is rather different from what we find in other contexts. In contexts such as funeral orations, or Simonides' poem on the battle of Plataia, or the paintings on the Stoa Poikile, or the epigram on the Eion herms,[127] past events are explicitly assimilated to the Greek resistance to Persia, as evidence of a continuity of bravery. The Persian Wars epigrams offers some evidence of a linking of battles, but also some rivalry. With Thucydides we have rivalry - but now no longer a continuity of bravery, but a contrast between success and total destruction.[128]

5 Conclusion

Some of the links which scholars have seen between Thucydides and earlier authors are instances of verbal repetition (patrothen, panolethria); others consist in shared plot-patterns (Nikias' letter and Agamemnon's speech, the Athenians' expedition to Sicily and Odysseus' return to Ithaca). My cautious treatment of some of the alleged links does not mean that I am hostile to the attempt to detect such potentially enriching patterns. I have argued, indeed, that many of the links are indeed plausible, but that in interpreting them we should not subject them to more weight than they can bear. Often a tempting similarity may be blurred by more telling differences, or our response to an apparent verbal echo may be modified by a broader view of a word's application. Indeed, the picture that has emerged from both parts of my paper is that it is by defining intertextuality in a broad and accommodating way that our understanding of Thucydides is most likely to be enhanced. I have, however, argued elsewhere for a detailed correspondence between Herodotus' narrative of the Persian invasion of Greece and Thucydides' narrative of the Sicilian expedition - a correspondence that is seen both in the choice of words and at a broader level of the structuring of the accounts, and indeed of the shaping of the works as a whole.[129] And in this paper I have offered slightly different readings of some of the alleged links with epic, proposing some new points of contact with epic (including the epic cycle), and broadened my previous analysis of Thucydides' use of Herodotus' account of the Persian Wars by looking for possible links between Thucydides and other accounts of the Persian Wars. This broader analysis gives a better view of the horizons of expectation of a fifth-century audience. Our tendency to place Herodotus and Thucydides at the beginning of the historiographical tradition runs the risk of obscuring the fact that they were writing in the context of - and to some extent competing with - other forms of commemorating the past. It is not that we should downplay the genuine striving for a new form of historical expression which we see in Herodotus and Thucydides; it is, rather, that we should see their receptiveness to other genres as part of the way they convey both the experience, and their own interpretations, of historical events.

NOTES

[1] OCD³ 1521 (the conclusion to a survey, published in 1996, of work on Thucydides since 1970, when the second edition of the OCD was published). For an excellent general account of the importance of Homer for historiography, see H. Strasburger, 'Homer und die Geschichtsschreibung', in Studien sur alten Geschichte, ii (Hildesheim and New York, 1982), 1057-97 (esp. 1065-6 on the agonal aspect and 1087 on the theme of suffering). Note also the recent PhD thesis of R. A. G. Williams, 'The Literary Affinities of Thucydides, with Particular Reference to the Influence of Epic' (University of London, 1993): he discusses specific points of contact not just with epic, but also (more speculatively) with logographers such as Charon of Lampsacus; R. L. Fowler, 'Herodotos and his Contemporaries', JHS 116 (1996), 62-87, helpfully discusses the links between some of these figures and Herodotus. For a good general discussion of intertextuality, see D. P. Fowler, 'On the Shoulders of Giants: Intertextuality and Classical Studies', MD 39 (1997), 13-34. [Return to text]

[2] See the good survey of S. Hornblower, 'Narratology and Narrative Techniques in Thucydides', in id. (ed.), Greek Historiography (Oxford, 1994), 131-66. [Return to text]

[3] See my paper 'Thucydides' Persian Wars', in C. S. Kraus (ed.), The Limits of Historiography: Genre and Narrative in Ancient Historical Texts (Leiden, 1999), 141-68, where I also cite earlier bibliography. [Return to text]

[4] See Hornblower, Comm. on Thuc., ii. 122-45. [Return to text]

[5] I will discuss: (§ 1) J. W. Allison, 'Homeric Allusions at the Close of Thucydides' Sicilian Narrative', AJPh 118 (1997), 499-516; (§ 2) A. V. Zadorojnyi, 'Thucydides' Nicias and Homer's Agamemnon', CQ 48 (1998), 298-303; (§ 3) S. A. Frangoulidis, 'A Pattern from Homer's Odyssey in the Sicilian Narrative of Thucydides', QUCC 44 (1993), 95-102, and C. J. Mackie, 'Homer and Thucydides: Corcyra and Sicily', CQ 46 (1996), 103-13 (independent of Frangoulidis). [Return to text]

[6] 'Introduction', in id. (ed.), Greek Historiography (Oxford, 1994), 67. On such usages, cf. C. F. Smith, Traces of Epic Usage in Thucydides', TAPhA 31 (1900), 69-81. [Return to text]

[7] Comm. on Thuc., ii. 38-61. See further G. Howie, 'Thucydides' Treatment of Brasidas and Cleon: A Homeric Approach', Histos (forthcoming). [Return to text]

[8] See e.g. E. H. Havelock, 'War as a Way of Life in Classical Culture', in E. Gareau (ed.), Classical Values in the Modern World (Ottawa, 1972), 19-78, who stresses the importance of Herodotus in mediating this epic influence. Havelock, 58 n. 76, criticizes the older view of F. M. Cornford, Thucydides Mythistoricus (London, 1907), esp. 201-20, that the Sicilian narrative is structured on tragic lines; note, though, that Cornford does stress the influence of Herodotus as well as of tragedy. [Return to text]

[9] Translations from Th. are based on the version of S. Lattimore (Indianapolis, 1998). [Return to text]

[10] Note that the Iliadic passage is from the Doloneia, a section of the Iliad often regarded as a later addition; but that it seems unlikely - to judge from his use of 'Homer's' Hymn to Apollo at 3. 104 - that Th. viewed it in such terms. The most similar passages in Homer are perhaps Il. 22. 415 (exonomakleden onomazon andra hekaston) and Od. 4. 278 (ek d' onomakleden Danaon onomazes aristous); but the use of patronymics in direct reports of characters' speeches is common. [Return to text]

[11] (n. 6), 9, 67. [Return to text]

[12] (n. 5), 301-2. [Return to text]

[13] (n. 5), 509-10. [Return to text]

[14] 'Nicias' Inadequate Encouragement (Thuc. 7. 69. 2)', CPh 80 (1985), 201-13, at 203 n. 5; so too Zadorojnyi (n. 5), 301 - though he also asks what alternative Nikias had, and sees sympathy for Nikias' earnestness (302 n. 22). I criticize other aspects of Lateiner's discussion of this passage in my book Thucydides: Narrative and Explanation (Oxford, 1998), 194-5. Note that, though Hornblower (n. 6), 9, refers to 'Nikias' "old-fashioned" appeal to his troops by their father's names, patrothen, and by the fame of their ancestors', he elsewhere rejects Lateiner's view that Nikias is criticized at 7. 69. 2 (Thucydides² (London, 1994), 193). [Return to text]

[15] W. R. Connor, Thucydides (Princeton, 1984), 201 n. 43. [Return to text]

[16] See Rhodes, CAAP, 254-5. D. Whitehead, however, claims that Ath. Pol. 21. 4 does reflect Kleisthenes' intentions, and that 'an individual's choice of demotic or patronymic could become an issue and an expression of class, status, and political values' (The Demes of Attica, 508/7 - ca. 250 BC: A Political and Social Study (Princeton, 1986), 71); cf. E. Vanderpool, 'Ostracism at Athens', in E. Sjöqvist and C. G. Boulter (eds.), Lectures in Memory of Louise Taft Semple 2 (Norman, Oklahoma, 1973), 215-70, at 222. The question B. Hainsworth, The Iliad: A Commentary, iii: Books 9-12 (Cambridge, 1993), 164, asks of Il. 10. 68 is relevant here too: 'would this be formal courtesy or, like the patronymics of Russian, a claim to intimacy and friendship?' [Return to text]

[17] The standard treatment of Th.'s use of patronymics is G. T. Griffith, 'Some Habits of Thucydides when Introducing Persons', PCPhS 187 (1961), 21-33; see also Hornblower (n. 2), 161-2, who also discusses Th.'s avoidance of (parochial) demotics (cf. also id. (n. 14), 97 n. 98); cf. also Williams (n. 1), 70 and n. 16, who notes that demotics are unhomeric (one might compare Th.'s lack of attention to the Athenian boule), and also sees Th. as anti-Kleisthenic. [Return to text]

[18] Both Marchant and Dover ad loc. note that it is the tribe's military significance that explains why Nikias names the men by tribe rather than by deme. [Return to text]

[19] (n. 5), 301 (he calls phulen 'a loaded word'). [Return to text]

[20] A. Andrewes, 'Phratries in Homer', Hermes 89 (1961), 129-40, at 132-3; he argues that Il. 2. 362-3 is an anachronism, reflecting the conditions of Homer's own day. The only other possible allusion to tribes is kataphuladon at Il. 2. 668. [Return to text]

[21] R. Parker, Athenian Religion: A History (Oxford, 1996), 131 (though he adds that Th. 'does not stress' this aspect). The stress on the tribes undermines Crane's reading of this passage: 'Nikias reverts to the appeal of aristocrat to aristocrat. . . . [he] is calling upon the elite members of society separately from their fellows. . . . The focus on the old rhetoric, on family and ancestry rather than on the polis, is for Thucydides not a mark of eloquence but a symptom of Nikias' own limitations' (G. Crane, The Blinded Eye: Thucydides and the New Written Word (Lanham, MD, 1996), 107). (The Kleisthenic tribes are also mentioned in military contexts at 6. 98. 2 and 8. 92. 4 (accepting phulen, the reading of C, rather than phulaken); for tribal arrangements in other states, cf. 3. 90. 2, 6. 100. 1.) The locus classicus for invocation of the tribes in a military, and strongly democratic, context is Dem. 60. 27-31. [Return to text]

[22] Cf. N. Richardson, The Iliad: A Commentary, vi: Books 21-24 (Cambridge, 1993), 150, on Il. 22. 415, quoted n. 10 above: 'The insistence on Priam's naming of each person individually stresses the desperation of his appeal'; also Andrewes (n. 20), 129-30, on references to military organization in the Iliad (such as Nestor's suggestion about tribes at 2. 362-3) as an ad hoc technique for emphasizing the particular occasion. [Return to text]

[23] Note, though, that the term trierarchos in Herodotus probably simply means 'captain of a ship', without the institutional significance that it has in Th.'s Athenian context, where the trierarch is performing a liturgy (Rhodes, OCD³, s.v. 'trierarchy'). [Return to text]

[24] See my paper cited n. 3 above for other, more important, links between the two accounts. [Return to text]

[25] e.g. IG i³. 59 a. 6, d. 37; ii². 1237. 119-20; see further Whitehead (n. 16), 72, who also quotes Raubitschek's view that this formula arose in the aftermath of Perikles' citizenship law of 451 - though it is not clear to me why a law which added having an Athenian mother to the requirements for Athenian citizenship should lead to a stress on paternity (distinguishing between homonyms is one reason for having additional forms of identification on inscriptions, as John Ma reminds me). [Return to text]

[26] Note e.g. 'naming patrothen' at Xen. Oec. 7. 3; Pl. Lys. 204e4, Leg. 753c3 (including naming by tribe and deme); Lys. fr. 345 (significantly of orphans). [Return to text]

[27] (n. 5), 509; she herself notes the inscriptional use (509 n. 25), as well as some other fifth-century literary uses. (But note that her claim that the use at Soph. Aj. 547 - where Ajax says that his son will not shudder at the sight of blood 'if indeed he is truly mine ta patrothen' - 'shows patrothen's connection to Homer, as Ajax' address to his son recalls Hector's talk with his' is scarcely credible if we remember that the word is only found once in Homer, and that Ajax' is a far more portentous (because redundant) use, stressing a deeper sense of connection between father and son: cf. Aesch. Ag. 1507 and Sept. 841, where the word is used of family curses.) [Return to text]

[28] Note that Smith did not discuss patrothen in his article on epic usage in Th. (n. 6). [Return to text]

[29] (n. 5), 301. He backs this claim by noting some specific parallels - e.g. the shared stress on the poor condition of the ships: Il. 2. 135, Th. 7. 12. 3; as he notes, the scholia cite the parallel, though here there is not even any verbal echoing (Homer has sesepe, Th. the less serious diabrochoi). [Return to text]

[30] ibid.; the point is that Nikias' strategy is successful, but its success is unfortunate, because withdrawal would have been better for the Athenians, while Agamemnon's strategy misfires because it does make the troops long to return home (though Zadorojnyi stresses rather that a major military setback follows for the Achaians: this complicates the matter further, because, as he notes, the Achaians do at last succeed). [Return to text]

[31] Another interesting parallel to Iliad 2 is Xenophon's ambiguous dream (matching the deceitful dream that Zeus sends Agamemnon) at Anab. 3. 1: see W. Rinner, 'Zur Darstellungswise bei Xenophon, Anabasis III 1-2', Philologus 122 (1978), 144-9. Cf. also Anab. 1. 3, where Klearchos deliberately stresses the danger of the Greeks' position, and has an associate suggest that the army returns home, as a way of pointing up the impossibility of a return and of stiffening the army's resolve: a successful variant of Agamemnon's tactic; and Ap. Rhod. Argon. 2. 622-37, where Jason stages a successful repetition of Agamemnon's feigned despair. [Return to text]

[32] Cf. M. J. Anderson, The Fall of Troy in Early Greek Poetry and Art (Oxford, 1997), 76. [Return to text]

[33] Note the typically Homeric counterfactual at Il. 2. 155-6: 'Then the Argives would have made a homecoming beyond what was fated, if Hera had not spoken to Athene . . .' [Return to text]

[34] Note that here Th. does bring out what would have happened: cf. n. 33. Note too that while it is undoubtedly a mere coincidence that Proclus' summary uses the same terms as Th. (hormemenous . . . katechei ~ hormemenon . . . kataschein), Th.'s kataschein (which harks back to kateiche at 2. 65. 8, where it is applied to Perikles) does seem to be an allusion to the political poems of Solon: cf. A. Szegedy-Maszak, 'Thucydides' Solonian Reflections', in C. Dougherty and L. Kurke (eds.), Cultural Poetics in Archaic Greece (Cambridge, 1993), 201-14, at 207-8; Hornblower (n. 2), 69 n. 165. [Return to text]

[35] See M. Davies, The Epic Cycle (Bristol, 1989), 48. [Return to text]

[36] Briefly noted by Zadorojnyi (n. 5), 300. [Return to text]

[37] e.g. Polyb. 1. 18. 10, 84. 1; Diod. Sic. 25. 4. 1, 24. 3. 1; Cass. Dio 49. 27. 1, 56. 12. 5; Hdn. 8. 5. 5; Procop. 6. 4. 5, 6. 1; for Latin examples, see Woodman on Vell. Pat. 2. 51. 2; Kraus on Livy 6. 33. 9. For the topos in a non-historical work, see Xen. Poroi 4. 48; and for a neat twist of the norm (besieged becoming besieger rather than vice versa) see Lucan 10. 490-1 ('obsessusque gerit - tanta est constantia mentis - / expugnantis opus' - the dynamic Caesar is the subject). [Return to text]

[38] Cf. J. V. Morrison, 'Thematic Inversion in the Iliad: The Greeks under Siege', GRBS 35 (1994), 209-27 (he does not discuss the Thucydidean parallels). [Return to text]

[39] Quotes from Mackie (n. 5), 108, 106, 113, 112, and 113 (his italics). [Return to text]

[40] I prefer to see the harbour of Syracuse as evoking Salamis rather than the Laistrygonians' harbour: see the article cited in n. 3. [Return to text]

[41] See the passages cited by P. A. Brunt, Arrian (Loeb; Cambridge, Mass., and London, 1976), i. 469-70. [Return to text]

[42] This may explain why Th. does not allude to the Charybdis story in his narrative of the main Sicilian expedition. For an explanation in terms of a shift in Th.'s attitude to mythical and geographical information, see H. D. Westlake's essay on 'Irrelevant Notes and Minor Excursuses in Thucydides', in his Essays on the Greek Historians and Greek History (Manchester, 1969), 1-38 (with p. 8 on Charybdis); I hope to consider Westlake's claims in more detail elsewhere, in a discussion of Thucydides' imaginative geography. [Return to text]

[43] (n. 5), 112. [Return to text]

[44] (n. 5), 102. [Return to text]

[45] Hornblower (n. 14), 116. The only other word in Th. with the nost- root is huponostein at 3. 89. 2, used of subsiding water (as at Hdt. 1. 191. 3, 4 (of a river), cf. 4. 62. 2 (of wood)). [Return to text]

[46] Connor (n. 15), 162 n. 9; he also sees in Thucydides' citing 'Trinakria' as the former name of Sicily (6. 2. 2) an allusion to Od. 11. 107. [Return to text]

[47] Frangoulidis (n. 5), 102. [Return to text]

[48] Quotes from article cited in n. 5, 514. [Return to text]

[49] Equally there is no tension between hope and reality at Il. 1. 59-61: Achilles begins a speech by saying that the Greeks will have to return home if they continue to be pressed by both war and plague, then suggests that they seek a cure to the plague (Allison acknowledges that Achilles is making a 'rhetorical threat'). Il. 1. 61 (ei de homou polemos te dama(i) kai loimos Achaious) is a more plausible candidate for imitation by Th.: for the destructive pairing of war and plague, cf. 2. 54. 1, 'a kind of Thucydidean counterpart to Achilles' words' (A. M. Parry, The Language of Achilles and Other Papers (Oxford, 1989), 174-5); the scholia note the parallel. [Return to text]

[50] One could apply to Th. (in a much toned-down way) the remark of R. L. Fowler, The Nature of Early Greek Lyric: Three Preliminary Studies (Toronto, Buffalo, and London, 1987), 9, on allusion in archaic poetry: 'poets throughout the archaic period continued to use traditional phrases to evoke the atmosphere of epic in a general sort of way.' [Return to text]

[51] Cf. Anderson (n. 32), 75, on this 'formulaic combination'. These passages are a reminder that there are other words than (apo)nostein to express the idea of return - some with an even more pronounced (or at least exclusive) epic feel: (apo)neomai is characterized by LSJ as an epic verb which is rare in tragedy and used only once in early prose. [Return to text]

[52] Cf. Hornblower (n. 45 above); also Allison (n. 5), 513 n. 30. Herodotus uses the verb aponosteein 29 times; he also uses the simple verb nosteein 16 times. Both verbs are often used in combination with the adverb opiso ('back'). On the other nost- compound used by Herodotus, huponosteein, see n. 45 above. [Return to text]

[53] See 8. 115 on the losses of the Persian contingent that returned with Xerxes after the battle of Salamis (cf. Aesch. Pers. 482-514); 8. 119 for a story (which he rejects) 'about the nostos of Xerxes' - that he was hit by a great storm when he was sailing back to Asia (this is his only use of the noun nostos); and 9. 89 for the losses suffered by the army left behind under Mardonios during its return to Asia after the battle of Plataia (note apenostese at 90. 1). [Return to text]

[54] For the 'none/few out of many returned' motif, cf. also Xen. Anab. 3. 5. 16, quoted above; also e.g. Caes. BG 7. 88. 4; Livy 23. 24. 10. [Return to text]

[55] This analysis is drawn from the excellent discussion of J. Marincola, 'Thucydides 1. 22. 2', CPh 84 (1984), 216-23; the use at Hdt. 2. 120. 5, he notes (221 n. 18), offers a glimpse of the approach of Hecataeus. [Return to text]

[56] (n. 55), 221 n. 17. [Return to text]

[57] A passage which Hornblower, Comm. on Thuc., ii. 128-9, thinks Th. had in mind at 1. 138. 6, where he calls Pausanias and Themistokles 'the most brilliant of the Greeks of their time'. [Return to text]

[58] Cf. Th.'s description of the slaughter at the River Assinaros as the 'greatest in the war' (pleistos . . . phonos, 7. 85. 4): Hornblower, Comm. on Thuc., ii. 145, lists references to claims that that is an echo of Hdt. 7. 170; J. W. Blakesley, Commentary on Herodotus (London, 1854), noted on Hdt. 7. 170 that 'it may reasonably be concluded that this passage was written before the annihilation of the Athenian expedition sent against Syracuse'. [Return to text]

[59] I am not convinced by the claim of Frangoulidis (n. 5), 101, that Th.'s reference to oral tradition (akoe(i)) specifically embraces Demodokos' songs of nostoi at Od. 8. 489-90. [Return to text]

[60] Cf. B. Shimron, 'Protos ton hemeis idmen', Eranos 71 (1973), 45-51, on the epistemological aspect of the phrase in Herodotus. It is also found in other geographic and ethnographic texts, e.g. Ps.-Scyl. 112; Paus. 10. 22. 2. [Return to text]

[61] Note ad loc. He added: 'Or is the meaning, "this action, in which Greeks alone were concerned," &c, as if it were Hellenikon on?'. But the participation of non-Greeks is important in the narrative (e.g. 7. 57. 11, 58. 3, from the catalogue of forces). [Return to text]

[62] HCT ad loc. (in his 1966 edn. of Book 7 he echoed Arnold's objection, as did Marchant ad loc.). Note that Hellenikon is retained by many translators, e.g. Crawley, Jowett, Warner, and Lattimore. [Return to text]

[63] 'Anhang' to 3^rd edn. of Classen/Steup, 282-3. [Return to text]

[64] Steup also proposed deleting te in the phrase tois te kratesasi lamprotaton and taking the phrase together with dokein . . . ismen: that is, Th. would be saying not that the Sicilian expedition was 'the greatest . . . of Hellenic events we have heard of', but that it was 'of Hellenic events we have heard of the most splendid for those who won . . .'. But Steup's view that Th. could not have claimed that the Sicilian expedition was the greatest Hellenic event ever seems overly austere. [Return to text]

[65] So too with other superlative clauses with kata ton polemon (2. 25. 3, 94. 1, 3. 113. 6, 4. 40. 1, 3. 113. 6) or en to(i) polemo(i) (3. 98. 4, 7. 44. 1, 85. 4). For kata as 'in relation to, concerning', see LSJ s.v., B. IV. 2 (for other Thucydidean examples with polemon, see 2. 62. 1, 100. 2, 6. 72. 2); kata can mean 'during or in the course of a period' (B. VII. 1), as Steup's interpretation requires (cf. e.g. Hdt. 7. 137. 1 'long afterwards during the war of the Peloponnesians and the Athenians'; Arist. Pol. 1307a4-5); the closest parallel in Th. would perhaps be 4. 48. 5 hosa ge kata ton polemon tonde, where Lattimore has 'the civil war . . . ended here, at least as far as this war is concerned', Crawley 'at least as far as the period of this war is concerned': it is evidently a question of where the emphasis lies in a phrase of broad meaning rather than a choice between two absolute meanings. [Return to text]

[66] Havelock (n. 8), 69 n. 122. [Return to text]

[67] Cf. Connor (n. 15), 206 n. 54, who well defends the presence of Hellenikon, arguing that it evokes 2. 64. 3 (where there is a repetition of Hellenon . . . Hellenes). For a stress on Greekness in other significant contexts, cf. the description of the battle of Sybota as 'the greatest sea-battle ever fought by Greeks against Greeks' (Hellesi pros Hellenas, 1. 50. 2); and 2. 71. 2, 6. 76. 4, 8. 46. 3 in freedom contexts. On Th.'s use of Hellen- language, see further Rood (n. 14), 240 n. 53, on 1. 109-10 (the Athenian expedition to Egypt); also 247 n. 85 for the suggestion that the stress on Greekness at 7. 87. 5 reinforces the parallel between the Athenian disasters in Egypt and Sicily; and 228, 247, on the phrase 'Greek war' at 1. 112. 2. Cf. also D. R. Shipley, A Commentary on Plutarch's Life of Agesilaos: Response to Sources in the Presentation of Character (Oxford, 1997), 204, on the phrase 'Greek war' (polemos Hellenikos) Plut. Ages. 15. 2: 'the epithet does not identify, but condemns.' [Return to text]

[68] H. R. Immerwahr, 'Ergon: History as a Monument in Herodotus and Thucydides', AJPh 81 (1960), 261-90; cf. Hornblower (n. 14), 30-2. [Return to text]

[69] Ibid. 266, 268. [Return to text]

[70] Ibid. 276; cf. 283 n. 69. [Return to text]

[71] There is no close Thucydidean parallel, for instance, to the present tense in a phrase like Hdt. 9. 75: esti de kai heteron Sophanei lampron ergon exergasmenon ('there is another splendid deed, too, performed by Sophanes'): that Herodotus' account of the deed is an otherwise unmotivated prolepsis is itself significant; Macan ad loc. commented that 'the lampron ergon, though wrought, and wrought out, in the past, is conceived as existing in the present: it is for ever'. [Return to text]

[72] Cf. Connor (n. 15), 206 n. 54. There is another revealing use of ergon as 'achievement' in the Sicilian narrative at 7. 21. 2, where Gylippos says that he hopes 'to accomplish a deed worth the risk' (I have strengthened Lattimore's 'something'). [Return to text]

[73] Note that the other section of Th. where the phrases dokein emoi and hon . . . ismen tend to occur, the Archaeology, can itself be labelled Herodotean on other grounds (e.g. its manner of reasoning about the past). [Return to text]

[74] See N. Marinatos Kopff and H. R. Rawlings, 'Panolethria and Divine Punishment', Parola del Passato (1978), 331-7, at 331 n. 1, quoting W. M. Calder. [Return to text]

[75] H. Strasburger, 'Thukydides und die politische Selbstdarstellung der Athener', Hermes 86 (1958), 17-40, at 39 n. 3 (tentatively); Marinatos-Kopff and Rawlings (n. 74). Cf. also Connor (n. 15), 208 n. 57: the passage seems 'to raise the question of theodicy but to leave it quite open. It is not a statement of Thucydides' theology, but a way to lead an enlightened and sophisticated audience to confront the awesome possibility that there may be a divine dimension to human history.' [Return to text]

[76] Comm. on Thuc., ii. 124 n. 3. The noun is not common later, but note Polyaenus 1. 43. 2, where it is used of the Athenian slaughter at Syracuse - an apparent echo of Th. 7. 87. 6. [Return to text]

[77] Aesch. Th. 71, 933, Ag. 535, Ch. 934, Eum. 552; Soph. Aj. 839, El. 1009, Ph. 322; Eur. Andr. 1225, El. 86; there is also an active use ('all-destructive'), Aesch. Pers. 562, Suppl. 414. It is also attested in a proverbial saying at Ar. Lys. 1039 ('that old saying is well and truly said - "neither with the deadly pests [sc. women] nor without the deadly pests"', trans. Sommerstein, who notes ad loc. that the saying is not elsewhere attested in this form). [Return to text]

[78] Cf. Dover, HCT iv. 465 (in his earlier commentary on Book 7 (Oxford, 1966), he argues ad loc. that to legomenon 'apologizes for the unusual word panolethria(i)' (so too Dunbar on Ar. Birds 1239), but this seems less satisfactory). At 7. 68. 1, the only other occurrence of the phrase in Th., it marks as proverbial the joy of defeating one's enemies: note that Th.'s specific verbal formulation there is not elsewhere attested. Cf. e.g. Diod. 20. 30. 1, where it alludes to a proverb attested at Th. 3. 30. 4 and elsewhere (see Gomme ad loc.). Marinatos-Kopff and Rawlings (n. 74), 332 n. 3, argue that the phrase can denote a specific reference, referring to Headlam's note on Herodas 2. 45; but in the passages they cite specific references are marked as such (e.g. Ael. VH 2. 12: touto de pou to tou Euripidou). The exception (Ael. VH 1. 30: soteres esthloi kagathoi parastatai, touto de to legomenon) is an iambic line uttered by a character in an anecdote; Snell printed it as TGF Adesp. 14, but its provenance is uncertain (cf. Pl. Symp. 197e1-2 for the similar phrase parastates kai soter aristos). Rawlings and Marinatos-Kopff also cite Pind. Nem. 3. 52-3: legomenon de touto proteron epos echo, said by the scholiast to be an allusion to Hesiod fr. 40 M-W; this is again different because explicit. [Return to text]

[79] 6. 3. 20: 'the Athenians held the view that now there was a good chance that the Thebans would be, as the saying goes, tithed' (to legomenon de dekateuthenai); 6. 5. 35: 'if you and we would agree together, there is a good chance that, as the old saying goes, the Thebans would be tithed' (to palai legomenon dekateuthenai). [Return to text]

[80] H. W. Parke, 'Consecration to Apollo: dekateuein to(i) en Delphois theo(i)', Hermathena 72 (1948), 82-114, at 84. Cf. also Walbank on Polyb. 9. 39. 5. [Return to text]

[81] Lattimore (n. 9), 407. His arguments against the attribution of a Herodotean theodicy to Th. do not seem so strong: this would 'involve Thucydides offering Nikias a lesson in matters theological' (a reference to 7. 77: but arguably Th. does offer a lesson - or at least a comment - on the claims of Nikias' rhetoric, whatever view one takes); 'undermine the elaborate comparison with the Pylos campaign' (but this need not be all-embracing; Athenian reaction to Pylos would fit well); 'and cloud the (admittedly puzzling) authorial statement that the Athenians did not support the expedition adequately (2. 65)' (but the divine could be working through the human). [Return to text]

[82] Plutarch, De Stoicorum Repugnantiis 1049c1, interestingly groups the Trojan, Persian, and Peloponnesian Wars as wars marked by panolethria. Pertinent for Th. is the frequency of other expressions for total destruction or harm in Aeschylus' Persae (see e.g., besides panolethroisin at 562, discussed in n. 77 above, lines 282, 353, 670, 716, 729, 732 (panoles); I owe this observation to Tom Harrison): these reinforce the parallelism in Th. between the Persian invasion of Greece and the Athenian invasion of Sicily. [Return to text]

[83] P. Vidal-Naquet, 'The Place and Status of Foreigners in Athenian Tragedy', in C. B. R. Pelling (ed.), Greek Tragedy and the Historian (Oxford, 1997), 109-19, at 113. [Return to text]

[84] See Rood (n. 14), 193-8, with references to earlier scholars. [Return to text]

[85] The Appian passage is well discussed by N. Purcell, 'On the Sacking of Carthage and Corinth', in D. Innes, H. M. Hine, and C. B. R. Pelling (eds.), Ethics and Rhetoric: Classical Essays for Donald Russell on his Seventy-Fifth Birthday (Oxford, 1995), 133-48, who sees Scipio's tears as 'a clear bid to put Rome on the map of historical culture' (p. 138). See also J. Hornblower, Hieronymus of Cardia (Oxford, 1981), 104-6. [Return to text]

[86] K. J. Dover, The Evolution of Greek Prose Style (Oxford, 1997), 169; I assume that Dover's 'perfect' and 'sombre' distinguish this trimeter from the freer use of the iambic trimeter in other genres (e.g. comedy). The trimeter has also been noted by W. R. M. Lamb, Clio Enthroned: A Study of Prose-Form in Thucydides (Cambridge, 1914), 266 (without comment), and by Hornblower (n. 2), 69 (Dover wrongly refers to 115). For another iambic trimeter in Th., see M. Haslam, 'Pericles Poeta', CPh 85 (1990), 33, on 2. 61. 2. [Return to text]

[87] On such 'concluding evaluations', see I. J. F. de Jong, Narrative in Drama: The Art of the Euripidean Messenger-Speech (Leiden, 1991), 74-6, 191-2; cf. e.g. Eur. HF 1014-15('I know of no man more wretched'), Phoen. 1478-9 ('for this city, some struggles have turned out most happily; others most unhappily'); Soph. OT 1284-5 ('all ills that can be named, none of them is absent'; cf. Th.'s double negative ouden hoti ouk apoleto), Tr. 811-12. For other closural features in 7. 87. 5-6, see D. P. Fowler, 'First Thoughts on Closure: Problems and Prospects', MD 22 (1989), 75-122, at 91-2. [Return to text]

[88] (n. 86), 169 n. 19. Cf. the iambic trimeter at Xen. Anab. 5. 2. 24: 'while they were still fighting and still doubtful what to do next, some god showed them a way of saving themselves (theon tis autois mechanen soterias / didosi)'; here too the tragic language (theon tis is common in tragedy, and found in an opening position at Aesch. Eum. 70 and Eur. Alc. 298; soteria is also common in tragedy, esp. at the end of a line; for the phrase mechanen soterias, cf. Aesch. Sept. 209 and Eur. Phoen. 890) makes one less inclined to see the metrical effect as accidental. [Return to text]

[89] See Rood (n. 14), 198 n. 72, for the 'unworthy of misfortune' motif in tragedy. [Return to text]

[90] Cf. D. P. Fowler, 'Second Thoughts on Closure', in D. H. Roberts, F. M. Dunn, and D. Fowler (eds.), Classical Closure: Reading the End in Greek and Latin Literature (Princeton, 1997), 3-22, at 15-16. [Return to text]

[91] See e.g. C. W. Macleod, Collected Essays (Oxford, 1983), 146; M. Lefkowitz, 'Patterns of Fiction in Ancient Biography', The American Scholar 52 (1983), 205-18, at 218. [Return to text]

[92] There are two or three passages in the Themistocles where Peloponnesian War motifs have perhaps been retrojected to the time of the Persian Wars: at 7. 1, Themistokles' proposal to meet the Persian fleet as far away from Greece as possible (a proposal not attested in any other source) resembles the strategy Thucydides attributes to Hermokrates in his account of the Syracusan response to news of the impending Athenian invasion of Sicily (6. 34); at 10. 8-9, Plutarch's account of the mixed emotions at those who watch the Athenians abandoning Athens by ship before Salamis and of the suffering of those left behind echoes two scenes in Thucydides (6. 30-1, the departure of the fleet to Sicily, and 7. 75, the departure of the Athenian army from Syracuse; admittedly the motifs had become a stock part of emotive historiography); and at 14. 3, Themistokles' tactic of waiting for the wind recalls Thucydides' description of Phormion's tactics at Naupaktos (2. 84). [Return to text]

[93] The Suda reports that Simonides composed an elegy on the battle of Artemisium and a melic poem on the battle of Salamis. Since there is evidence for a melic poem on Artemisium, Bergk proposed that the Suda had confused the two poems. But the recently published Simonides papyrus confirms that he did write an elegy on Artemisium. West retained an elegy on Salamis in IE². But he has since written that he 'probably ought to have discarded the heading he en Salamini naumachia ['the sea battle at Salamis'] and the testimonia referring to it; the other fragments (6-9) may equally have come from the Artemisium poem, or some other' ('Simonides Redivivus', ZPE 98 (1993), 1-14, at 2-3). [Return to text]

[94] So too I. Scott-Kilvert (Penguin trans.); L. Piccirilli, Le Vite di Temistocle e di Camillo² (Milan, 1996); J. L. Marr, Plutarch: Life of Themistocles (Warminster, 1998). [Return to text]

[95] J. F. Lazenby, The Defence of Greece 490-479 BC (Warminster, 1993), 196. [Return to text]

[96] R. J. Lenardon, The Saga of Themistokles (London, 1978), 102; A. J. Podlecki, The Life of Themistocles: A Critical Survey of the Literary and Archaeological Evidence (Montreal, 1975), 49-50; E. Culasso Gastaldi, 'Temistocle, Eschilo, Simonide e il culto della vittoria', in E. Corsini (ed.), La Polis e il suo teatro (Padua, 1986), 31-47, at 43. [Return to text]

[97] A. M. Bowie, 'Tragic Filters for History: Euripides' Supplices and Sophocles' Philoctetes', in C. B. R. Pelling (ed.), Greek Tragedy and the Historian (Oxford, 1997), 39-62, at 43, emphasizing the contrast with the lack of propaganda in tragedy. [Return to text]

[98] The first edition of Iambi et Elegi Graeci (1972) and the shorter version Delectus ex Iambis et Elegis Graecis (1980) respectively. [Return to text]

[99] Frost ad loc. finds this contrast 'peculiar': but note that in De Herodoti malignitate Plutarch is citing Simonides as proof in an openly polemical work. [Return to text]

[100] e.g. Flacelière (Budé edn., n. on p. 224); also Schaefer and Sintenis (noted in Schneidewin's 1835 edn.). [Return to text]

[101] A. J. Podlecki, 'Simonides: 480', Historia 17 (1968), 257-75, at 267 (assuming that Simonides' poem was an elegy). [Return to text]

[102] Schneidewin (n. 100), 9; C. M. Bowra, Greek Lyric Poetry² (Oxford, 1961), 341 n. 1 ('probably a quotation'); J. H. Molyneux, Simonides: A Historical Study (Wauconda, IL, 1992), 188-9; Podlecki (n. 101), 267, regards this as a definite citation. The issue is unfortunately not discussed in the discussion of Simonides' poems on the Persian Wars by W. Kierdorf, Erlebnis und Darstellung der Perserkriege (Göttingen, 1966), 16-29. [Return to text]

[103] Note that Aeschylus is only quoted once in the Themistokles, at 14. 1, on the number of ships in the Persian force (seemingly because the messenger stresses his personal knowledge). O. Poltera, Le langage de Simonide: Étude sur la tradition poétique et son renouvellement (Bern, 1997), 93, however, argues that Plutarch's 'holding out until evening' is derived from Aeschylus, and so that 'le réminiscence simonidéenne suit nécessairement l' inquit'. But see below on how Aeschylus and Plutarch/Simonides differ in their accounts of what lasted until evening; and note that elsewhere (323 n. 47) Poltera takes 'holding out until evening' as Simonidean (he cites it as a parallel to support West's interpretation of deron at fr. 17. 5 West as referring to the length of the battle of Plataia). [Return to text]

[104] Epic: see C. B. R. Pelling, 'Aeschylus' Persae and History', in id. (ed.), Greek Tragedy and the Historian (Oxford, 1997), 1-19, at 6 n. 18 (note also his analysis (2-6) of light/dark imagery in Herodotus and Aeschylus: clearly we cannot tell how whether this imagery was important in Simonides' poem). Historiography: see Broadhead's note on Aesch. Pers. 426-8; Kraus's note on Livy 6. 9. 11. [Return to text]

[105] Ziegler, Plutarchos, RE 21 (1951), 916, notes the frequency of Plutarch's quotations of Simonides (22 verbatim citations); for a list, see W. C. Helmbold and E. N. O'Neil, Plutarch's Quotations (Baltimore, 1959). I see no reason to follow Wilamowitz, Sappho und Simonides: Untersuchungen über griechische Lyriker (Berlin, 1913), 144 n. 2, who claims that Plutarch's citation was 'nicht aus eigener Lekture, sondern aus Historikern'. Simonides is cited elsewhere in the Lives at Thes. 10. 2 and Tim. 37. 1 (phrases cited kata Simoniden) and as historical evidence in case of variants (Thes. 17. 5, on the colour of Theseus' sail and the name of his helmsman; Lyc. 1. 4, in a dispute over Lykourgos' genealogy). [Return to text]

[106] Probably a dedicatory epigram, in the view of editors of Simonides (e.g. Bergk, Campbell), though B. Perrin, Plutarch's Themistocles and Aristides (New York and London, 1902), 176, suggested that it was 'perhaps in the great Salamis-hymn'. [Return to text]

[107] A. J. Podlecki, 'Simonides and Themistocles: Supplementary Notes', Historia 18 (1969), 251. [Return to text]

[108] W. R. Connor, 'Lycomedes against Themistocles? A Note on Intra-genos Rivalry', Historia 21 (1972), 569-74, at 572 n. 7; cf. Perrin (n. 106), 216, arguing that Plutarch was citing from memory. [Return to text]

[109] In extant poetry down to the fifth century, the word is only used at Soph. OT 192, in an unusual sense (Jebb translates it 'amid cries'). Poltera (n. 103), 93-4, is tempted, nonetheless, to take it as Simonidean; he even speculates that Thucydides' use of the adjective in a naval context at 6. 31. 6 (describing the Athenian fleet sent to Sicily; see above on the use of lamproteti in this passage) could be an echo of Simonides. [Return to text]

[110] Note that Plutarch uses periboetos with nike at Nik. 16. 8, with kalos at Pomp. 40. 5 and De mul. virt. 256a9, and of the Greek victory at Plataia at de Herodoti malignitate 871e5. Cf. also 871b3, where a famous story about a prayer made by the women of Corinth (ten kalen ekeinen kai daimonion euchen) is illustrated by a Simonidean epigram. [Return to text]

[111] e.g. in epinician: cf. Pind. Pyth. 8. 97 ('The dream of a shadow is man, no more. But when the brightness comes, and God gives it, there is a shining of light on men (lampron pheggos), and their life is sweet'), Nem. 8. 34 (spiteful deception 'violates the beautiful and the brilliant'). See in general L. Kurke, The Traffic in Praise: Pindar and the Poetics of Social Economy (Ithaca and London, 1991), index, s. v. 'light'. [Return to text]

[112] For the phrase lampron ergon (or cognate expressions), cf. Hdt. 1. 174, 3. 72, 6. 15, 9. 75; Xen. Hell. 4. 1. 21, 5. 2. 28 (with C. J. Tuplin, The Failings of Empire: A Reading of Xenophon Hellenica 2. 3. 11 - 7. 5. 27 (Stuttgart, 1993), 97 with n. 32). Lampros is used of the Persian Wars at Th. 3. 59. 2, and in general encomiastic contexts at e.g. Isoc. 9. 39; lamprotes is gained through war (Isoc. 10. 17, 6. 104) and preserved through peace (Th. 4. 62. 2). [Return to text]

[113] Simonides' poem on Plataea has been much discussed: see esp. D. Boedeker, 'Simonides on Plataea: Narrative Elegy, Mythodic History', ZPE 107 (1995), 217-29. Cf. also M. A. Flower, 'Simonides, Ephorus, and Herodotus on the Battle of Thermopylae', CQ 48 (1998), 365-79, on the possible use by Ephorus of Simonides' poem on Thermopylai as a way of 'correcting' Herodotus. [Return to text]

[114] See his paper 'Simonides and Horace', in D. Boedeker and D. Sider (eds.), Praise and Desire: Contexts for the New Simonides (forthcoming). [Return to text]

[115] This link is also made by Connor (n. 15), 206 n. 54. [Return to text]

[116] The attribution of the epigram to Eurymedon by Diod. 11. 62. 3 has recently been defended by P. J. Stylianou, 'The Untenability of Peace with Persia in the 460s BC', Archbishop Makarios III Foundation (1989), 339-71, at 353-8; H. T. Wade-Gery, 'Classical Epigrams and Epitaphs: A Study of the Kimonian Era', JHS 53 (1933), 83-4, followed Domaszewski's view that the eight-line epigram was in fact two four-line epigrams, one for Eurymedon, the other for the Cyprus victory. [Return to text]

[117] Discussions of the land/sea dichotomy (see esp. A. Momigliano, 'Terra Marique', JRS 32 (1942), 53-64) tend to study not the configuration of individual battles, but the motif of imperial control of land and sea, which is mainly Hellenistic and Roman (though foreshadowed at e.g. Th. 2. 62. 2; Xen. Anab. 6. 6. 13; Isoc. 5. 47, 9. 54). For the dichotomy in Th., see further Hornblower, Comm. on Thuc., i. 9. [Return to text]

[118] Cf. E. Kemmer, Die polare Expressionsweise in der griechischen Literatur (Würzburg, 1903), 166-7; Nisbet and Hubbard on Horace, Carm. 1. 6. 3; G. E. R. Lloyd, Polarity and Analogy: Two Types of Argumentation in Early Greek Thought (Cambridge, 1966), 90-4, who notes (91) the Homeric use of 'foot-soldiers and horsemen' at Il. 2. 809-10 (ek d' essuto laos / pezoi th' hippeis te), cf. 8. 58-9. The separation of land-force and ships (often stressed by men . . . de) is standard, e.g. in enumerations of the strength of armies or of different engagements, e.g. Hdt. 5. 108. 2, 112. 1; Th. 4. 1. 4; Ctes. FGH 688 F 14. [Return to text]

[119] Some scholars do refer the epigram just to the battle of Salamis: in that case, pezoi will refer to the fighting on the island of Psyttaleia that is stressed in Aeschylus' Persae. N. G. L. Hammond, 'The Campaign and Battle of Marathon', JHS 88 (1968), 13-57, at 27 n. 64, objects that 'the emphatic position of pezoi can hardly be reconciled with the small-scale attack on Psyttalia'; Page, FGE 221 n. 1, cites the Persae against this; and the forcefulness of the polar expression probably renders Hammond's qualms about the position of pezoi unnecessary. O. Hansen, 'On the So-called Athenian Epigrams on the Persian Wars', Hermes 127 (1999), 120-1, argues that the epigram may refer to a battle earlier than the Persian Wars (e.g. the battles against Boiotia and Chalkis of Hdt. 5. 77); pace Hansen, the epigram's stress on saving Greece from slavery makes this extremely implausible. [Return to text]

[120] See H. R. Immerwahr, Form and Thought in Herodotus (Chapel Hill, NC, 1966), index, s.v. 'land and sea'. [Return to text]

[121] Presumably - with Gomme, following the scholia and Plutarch - Artemision and Salamis, Thermopylai and Plataia: that is, the battles involved in ta Medika taken as the Persian invasion of Greece; this would exclude Mykale where there was more of a confusion between land and sea battles: a landing was followed by victory on land (cf. Gomme HCT i. 151: 'not much of a sea-battle'). Perhaps Th. defined the Persian Wars as he does because was thinking of Hdt. 7. 120, where the size of Xerxes' expedition is assessed against that of previous expeditions: as an attack launched by a Greek force against the Asian shore, Mykale is not relevant to this; and if Mykale were included, then (as Gomme asks) why not also the other battles in the continuing war against Persia? [Return to text]

[122] See Pelling (n. 104), 6-9, for a good discussion of the reasons for the different use in Persae and Herodotus of the land/sea dichotomy. [Return to text]

[123] (n. 116), 84 n. 54. [Return to text]

[124] HCT i. 289 n. 1. [Return to text]

[125] Cf. F. W. Walbank, 'Alcaeus of Messene, Philip V, and Rome', CQ 36 (1942), 134-45, at 134-7, for linking between epigrams (AP 6. 171, 9. 518) through claims of domination over land and sea. [Return to text]

[126] See P. Amandry, 'Sur les "épigrammes de Marathon"', in Theoria: Festschrift fur W.-H. Schuchhardt (Baden-Baden, 1960), 1-8, at 6-8, followed by Hammond (n. 119), 27-8. Against this notion, see the excellent remarks of Pelling (n. 104), 9-12. [Return to text]

[127] Note, e.g., the explicit allusion in the Eion herms to Homer's portrayal of Menestheus, the leader of the Athenian contingent against Troy: 'Once from this city Menestheus went . . . to the holy plain of Troy; Homer once said that as marshal in battle he was outstanding among the stout-corsleted Danaans. So it is not unseemly that Athenians be called marshals in war and manliness. They too were of steadfast heart who once at Eion on the waters of the Strymon subjected the sons of the Medes to fiery hunger and chilling Ares . . .' (Page, FGE 40; trans. D. A. Campbell). See in general N. Loraux, The Invention of Athens: The Funeral Oration in the Classical City, trans. A. Sheridan (Cambridge, Mass., 1986), 132-71. [Return to text]

[128] It should be noted that we do have some epitaphs for Athenian casualties in battles against fellow-Greeks (e.g. Tod 59; ML 48) in which we hear of the excellence (arete) to the dead, of how they brought fame (eukleia) to a native land which now misses (pothei) them; but that none of them seems to have the precise elements that link Th. 7. 87 with the Persian Wars epigrams. If indeed these epitaphs did echo Persian Wars motifs, other Greeks could have read them as a shocking perversion of the spirit of the Persian Wars. [Return to text]

[129] See n. 3; Tom Harrison points to some more links in a forthcoming paper. I am grateful to Simon Hornblower and Christopher Pelling for their remarks on an earlier version of this paper. At the HISTOS end, John Moles and Tony Woodman edited the paper (lightly). [Return to text]

Back to HISTOS: Contents of the current volume

Back to HISTOS Main Menu

To Durham Classics Department Home Page