¹ Myres, J.L., Herodotus, Father of History (Oxford, 1953)Google Scholar.

² See for instance much of Immerwahr, H.R., Form and Thought in Herodotos [A.P.A. Monographs 23] (Chapel Hill, 1966)Google Scholar, a work containing many judicious observations none the less. Wood, H., The Histories of Herodotos: an analysis of the formal structure (The Hague, 1972)Google Scholar, has little to offer, while Benardete, S., Herodotean Inquiries (The Hague, 1969)Google Scholar I find quite incomprehensible.

³ Lattimore, R., ‘The composition of the History of Herodotos’. CP 53 (1958), 9–21.Google Scholar

⁴ Comparison with the feats of oral composers of narrative poems would not be valid; see below on disparity of subject-matter.

⁵ Thornton, H. & A., Time and Style: a psycho-linguistic essay in classical literature (London, 1962), p. 37Google Scholar. Drexler, H.. Herodot-Studien (Hildesheim/New York, 1972), p. 57Google Scholar, has a similar opinion.

⁶ Howald, E., Hermes 58 (1923), 113Google Scholar; Vom Geist antiker Geschichtsschreibung (Munich, 1944), p. 11.Google Scholar

⁷ Cobet, Justus, Herodots Exkurse und die Fraqe der Einheit seines Werkes [Historia Einzelschrift 17] (Wiesbaden, 1971).Google Scholar

⁸ The total history of Egypt had no place in Herodotos' work; its beginnings were relevant, as were the latest developments concluding with its incorporation in the Persian Empire. (I hope to make the place, extent, etc. of the Egyptian logoi the subject of a further study.)

⁹ This is a point of dissimilarity between the epic and Herodotos; a great deal of his subject-matter, not merely his treatment of it, would be completely new to his audience. But see note 24.

¹⁰ Aristotle, , Poetics 23.3.Google Scholar

¹¹ Not accessible to me was the study of Turska, K., ‘De ratione qua Herodotus in compositione Historiae Homerum imitatus sit’ (in Polish), Meander 23 (1968), 283-97 and 353-69.Google Scholar

¹² Brown, T.S., ‘The Greek sense of time in history as suggested by their accounts of Egypt’, Historia 11 (1962), 257-70Google Scholar. ‘Herodotos’ plot, like that of the epics from which it sprang, was not the main thing. At times the plot seems merely an artistic device for tying together a fascinating series of digressions on the peoples of the world’(269).

¹³ See, for example, Froidefond, C., Le mirage égyptien dans la littérature grecque d'Homére à Aristote (Paris, 1971), p. 156.Google Scholar

¹⁴ Hekataios is even more poorly represented in his Genealogies than in the Periegesis. For Pherekydes frs 20, 127 give some idea of his fairly unsophisticated and unambitious style. The obvious lack of any influence on Herodotos from Gorgias and the sophistic stylists need not be a mere confirmation of the traditional (430's) date of composition. So mannered a style could not be successfully adapted to a work of such length addressed to a general public. Fornara, C., Herodotus, an interpretative essay (Oxford, 1971), p. 26Google Scholar, says: ‘The level of narrative found in Book II was typical of historical writing for Herodotus and his colleagues until Herodotus himself transcended it.’

¹⁵ See Cobet, op. cit., pp. 61-2.

¹⁶ For the opening, see White, M.E., Herodotos' starting point'. Phoenix 22 (1969), 39–48CrossRefGoogle Scholar, esp. 40-2 and 48. The apparent inconsistency between i 5.1 and i 14.4 (Alyattes) is of the sort that may be used to support the theory of an ‘unstructured’ composition. As it has been discussed ad nauseam I will merely remark that Herodotos not infrequently amends, modifies or limits an earlier statement; nor are the two statements simply contradictory.

¹⁷ Cf. Barth, H., ‘Zur Bewertung und Auswahl des Stoffes durch Herodot. Die Begriffe Klio 50 (1968). 93–110CrossRefGoogle Scholar. She shows that the historical aspect includes many examples of , which are remarkable deeds.

¹⁸ Cf. Cobet. op. cit., p. 80: ‘Herodotos in seiner Komposition vieles nebeneinanderordnet, ohne durch sofort einleuchtende Funktionen auch untergeordnet.’

¹⁹ On which see Camerer, I., Praktische Klugheit bei Herodot (Köln, 1965)Google Scholar. It is urged by Fornara (n. 24 infra) that Herodotos' main motive was ‘the magnificent picture of his plight’, i.e. a dramatic scene. Other parts of the story (e.g. the confrontation of the villains with their victim) are equally dramatic; but to me it seems that it is confusing method with intention in a McLuhanesque manner to select dramatization per se as Herodotos' motivation. But see on Atys and Adrastos, below p. 6.

²⁰ The respectable authorities for such a view include: Jacoby, F., ‘Herodotus’, RE Suppl. ii (1913). 361-2Google Scholar; Pohlenz, M., Herodot der erste Geschichtschreiber des Abendlandes (Leipzig, 1937; Stuttgart, 1961), pp. 208-11Google Scholar; Powell, J.E., The History of Herodotos (Cambridge, 1939), pp. 31-4.Google Scholar

²¹ Dihle, A., ‘Herodot und die Sophistik’, Philologus 106 (1962), 207-20.CrossRefGoogle Scholar

²² Segal, C.P., ‘Croesus on the Pyre: Herodotus and Bacchylides’, Wiener Studien 5 (1971), 40.Google Scholar Cf. Fornara, op. cit., p. 36 n.14: ‘vehicles for philosophy’.

²³ The career and achievements of Polykrates are very prominent in the History. But as Cobet points out (op. cit., p. 178), Polykrates' career is not told as a career with a message or lesson, but in single scenes each with its own import — a single aspect of Polykrates' Lebenslauf. Hence a non-chronological arrangement, so that we learn, for example, from the Persian Mitrobates just before Oroites traps the Samian ruler the interesting fact that the tyrant originally seized power with a mere handful of hoplites. On Polykrates and others, see Waters, K.H., Herodotos on Tyrants and Despots [Historia Einzelschrift 15] (Wiesbaden, 1971).Google Scholar

²⁴ Fornara, C., Herodotus: an interpretative essay (Oxford, (1971), p. 81Google Scholar, says: ‘The knowledge and expectations of his audience are an essential pre-supposition for his narrative’, but this appears to apply rather to events later than those described in the History.

²⁵ Contra, Fornara, op. cit., pp. 29-30.

²⁶ This use of the flash-back is typical and capable of interpretation in two ways; my view is that it represents considered application of structural principles, while others might take it as showing that at this point the historian ‘has just remembered’ the origin of Demokedes and that the non-chronological order proves that the various items were set down in the order in which they were recalled. Leonidas is first heard of at v 41; his full genealogy is given before Thermopylae at vii 204. His wife Gorgo has been mentioned at v 48 and 51 and will appear again at the end of Book vii; in vii 205 we only hear that Leonidas had mairied Kleomenes' daughter.

²⁷ Cf. Starr, C.G., Political Intelligence in Ancient Greece [Mnemosyne, suppl. 33] (Leyden, 1974), pp. 20 fGoogle Scholar. with n. 2.

²⁸ The inevitably somewhat subjective views as to what constitutes a digression and what is genuinely part of the historical account are summarized by Cobet, op. cit., pp. 55-6.

²⁹ Breitenbach, H.R., ‘Herodotus pater historiae’, SZG 16 (1966), 465.Google Scholar

³⁰ The order of introduction of the Greek states may be influenced by reasons other than their importance. Ionian cities apart, Corinth gets first mention, under her tyrant Periander; this is purely incidental to Ionian affairs (compare also the discussion of the Arion episode above). Athens and Sparta receive joint introduction twice over, first for Kroisos and later for Aristagoras; evidently their equal importance in the Persian Wars is foreshadowed here. (The explicit ‘editorial’ statement in vii 139 that next to the gods the Athenians were responsible for the defeat of Xerxes may be intended as a corrective to the impression of Spartan dominance given both by the reiterated fact of Spartan command in all the battles and the prominence of the highly Spartan-coloured account of Thermopylae in the same book.)

³¹ Beck, I., Die Ringkomposition bei Herodot und ihre Bedeutung für die Beweistechnik [Spudasmata 35] (Hildesheim/New York, 1971)Google Scholar does not deal with widely separated mentions of the same matter, but rather with repetitions of words or phrases in particular sections — patterns of composition on a limited scale, which are primarily stylistic but also utilitarian in intention.

³² See de Romilly, J., Histoire et raison chez Thucydide (Paris, 1956).Google Scholar

³³ It is pleasing to note the gradual emergence of a more commonsense attitude to Greek literature; critics now mostly treat Aeschylus and Sophokles as playwrights not as theologians. A similar courtesy should be extended to the Father of History. However the ‘essentially tragic view’ of Herodotos is still current, see C.P. Segal, op. cit. (n. 22 above).

³⁴ In addition to such episodes as Arion (above p. 5) one might consider the diversification of the account of the Plataea campaign by the inclusion of legendary history in the speeches of Tegeates and Athenians, by the speech of Mardonios, and the lengthy, rhetorical message of Pausanias to the Athenians. This whole matter will repay further study; and see next note.

³⁵ ‘The usual structure of a battle-account in Herodotos is battle, pursuit, mention of dead’: H.R. Immerwahr's view quoted with approval by Harrison, E.B., ‘The South Porch of the Nike Temple and the Marathon Painting’, AJA 76 (1972), 363.CrossRefGoogle Scholar An examination of Herodotos' major battles will show that there is no such ‘usual structure’ — though it might appear almost inevitable. A brief glance at ‘patterns’ of battles in Roman historians, beginning with speeches by the opposing generals and frequently consisting of a mass of cliches, will serve to emphasize the variety of Herodotos' accounts.