¹ Herodotus uses the term ‘Assyria’ to indicate what he knew as the Persian satrapy of Assyria. This included not only Assyria proper (the steppe-country of northern Iraq centred on the Tigris near modern Mosul) but also Babylonia (southern Iraq from below Baghdad to the Persian Gulf). See, for instance, 1.106, 178, 192, 193; 3.92; 4.39. In terms of Mesopotamian history, of course, Assyria and Babylonia were separate political entities, although at different times the Assyrians were masters of Babylon, or the Babylonians of Assyria. For a fuller account of their relationship see Macqueen, J. G., Babylon (London, 1964) and Assyria (forthcoming).Google Scholar

² For a list of fulfilled promises see Drexler, H., Herodot-Studien (Hildesheim, 1972), pp. 252–3.Google Scholar There is one other unful filled promise, in 7.213.3, but this concerns a minor detail (the murder of Epialtes), which could fairly easily have been overlooked.

³ Stein, H., Herodotos ⁶ i (Berlin, 1901), xlvii ff.Google Scholar

⁴ Immerwahr, H. R., Form and Thought in Herodotus (Cleveland, 1966), p. 31: ‘the Assyrian logoi, which were probably never written’.Google Scholar

⁵ Lattimore, R., Classical Philology 53 (1958), 19.CrossRefGoogle Scholar

⁶ Drexler, , op. cit., p. 253:Google Scholar ‘Ueber den Grund der Unterlassung sich den Kopf zu zerbrechen ist zwecklos’.

⁷ Erbse, H., Gnomon 41 (1969), 126,Google Scholar citing Focke, (Tübinger Beiträge 1 (1927), 14–19) as being ‘auf dem richtigen Weg’.Google Scholar

⁸ Drexler, , op. cit., p. 253.Google Scholar

⁹ Powell, J. E., A Lexicon to Herodotus (Cambridge, 1938),Google Scholar lists seven other passages where the phrase occurs. In six of these (1.15; 2.102.1; 4.16.1; 5.74.2; 6.19.3; 6.55) the following genitive indicates the object mentioned or remembered, while in the other (1.193.4) the phrase is followed by a noun-clause and no genitive is used. The phrase is used in a similar way. There are five examples (1.14.1; 2.43.3, 56.2; 4.79.2, 81.2), all of which show a genitive of the object mentioned or remembered. 6.122.1 also contains with a genitive similarly used, but the passage is probably not to be attributed to Herodotus.

¹⁰ It is noticeable for instance that no credit for building operations is given to ‘Labynetos’, especially if his arbitration after the eclipse-battle between the Medes and the Lydians in 585 B.C. (1.74.3) is taken to show that Herodotus uses the name to indicate Nebuchadnezzar (605–562 B.C.) as well as Nabonidus (556–539 B.C.). It is difficult to believe that Herodotus would have made no reference to the monarch who was responsible for most of the building-work which he saw and admired. My own feeling is that ‘Labynetos’ would have been given a good deal of space in the

¹¹ Powell, J. E., The History of Herodotus (Cambridge, 1939), chapters II and III.Google Scholar

¹² Ebert, M., Zur Frage nach der Beendigung des Herodoteischen Gescbicbtswerkes (Keil, 1911).Google Scholar

¹³ How, W. W. and Wells, J., A Commentary on Herodotus (Oxford, 1912), Vol. i, p. 380.Google Scholar

¹⁴ Wood, Henry, The Histories of Herodotus (The Hague, 1972), p. 87.Google Scholar

¹⁵ Cf. Powell, , op. cit., p. 22:Google Scholar ‘Since, therefore, the were not attached to the first capture of Babylon, there is no alternative to concluding that their original connection was with the second.’

¹⁶ e.g. Myres, J. L., Herodotus Father of History (Oxford, 1953), p. 95: ‘For Egypt, the dynastic history follows the geographical description; and there is nothing to preclude such a history of Babylonia after 1 200.’Google Scholar

¹⁷ Immerwahr, , op. cit., pp. 91–3.Google Scholar

¹⁸ Immerwahr, , op. cit., p. 93, n. 42.Google Scholar

¹⁹ A new edition of Book 2 (Lloyd, Alan B., Herodotus Book II (Leiden, 1975, 1976) is at present in course of publication, but has not yet reached 2.99.Google Scholar

²⁰ is used in the sense of ‘clan’ only here in Herodotus. Elsewhere it means ‘lineage’ (cf. 2.143.1; 3.75.1), but it is difficult to see how in this passage it can refer to anything other than groups of people linked by some sort of relationship through the male line. It is attested in this sense in inscriptions of the fifth century B.C. (cf. LSJ s.v.). While it is possible that the passage refers to groups of people who for social or religious reasons actively abstained from eating anything but fish, there is, as far as I am aware, no evidence for such groups in Babylonia, whereas other passages in Herodotus (e.g. 3.98) point towards dwellers in marshy areas who eat nothing but fish because there is virtually nothing else in such areas for them to eat.

²¹ This passage seems to me to be completely different from other passages such as 4.30 (why there are no mules in Elis) or 7.170 (information on Micythus of Rhegium). Both these latter passages are not so much afterthoughts as digressions giving information which is not, strictly speaking, necessary to the matter in hand, after which Herodotus apologizes and returns to his original subject. Not so here; the passage cannot be said to contain a piece of information which is interesting but largely irrelevant. Rather it seems to me to contain information which is relevant, but not in the position in which we find it.

²² Denniston, J. D., The Greek Particles ² (Oxford, 1954), pp. 325–7.Google Scholar

²³ 4.22–5.

²⁴ Herodotus also tells us here that the Persians call all Scythians Sacae. The truth of this is confirmed by the inscriptions of the Achaemenid kings (e.g. A). See Kent, R. G., Old Persian Grammar, Texts, Lexicon (New Haven, 1953), p. 137.Google Scholar

²⁵ Nor have they any connection with the ‘Fish-eaters’ who, according to Diodorus (3.15–22), inhabited the coast all the way from southern Persia to the entrance to the Red Sea. Fish-eaters in different parts of the world seem to have had a curious fasci nation for the Greeks. Herodotus for instance mentions them in Egypt (3.19) and in India (3.98) as well as in the present context.

²⁶ The suggestion that the Massagetae are etymologically ‘fish-eaters’ through a connection of their name with Avestan Masyδ = fish (cf. RE s.v.) carries little conviction. Connections with the Thyssagetae (4.22 and 123) and thence with the Getae, the Goths, the Finns, and the Hungarians, are extremely speculative.

²⁷ For a full discussion see Huxley, G., GRBS 6 (1965), 207–12.Google Scholar

²⁸ Eusebius, , Arm. Chron. p. 28, 28 ff.Google Scholar (ed. Karst); Malalas, p. 26 (ed. Dindorf); Kephalion, , FGrHist 93 F 1. See Rawlinson's note on 1.106.Google Scholar

²⁹ Our uncertainty about the fortunes of the text of Herodotus in antiquity is well illustrated by the alternative readings in the introductory paragraph. was the accepted reading in the fourth century (Aristotle, Rbet. 3.9.1409a) and early third century B.C. (Duris of Samos, FGrHist 76 f 64). By the early second century A.D. had replaced (Plutarch, Mor. 604), presumably from another tradition which had survived through the intervening period. It may well have been in this version that the were lost.