¹ For the former view see, e.g. Leaf, W., The Iliad², vol. I (Amsterdam, 1960)Google Scholar and Monro, D. B., Homer. Iliad, Books I–XII ⁵ (Oxford, 1906), both on 356Google Scholar; for the latter, most recently Kirk, G. S., The Iliad: a Commentary, vol. I: Books 1–4 (Cambridge, 1985), on 1.185.CrossRefGoogle Scholar

² Kirk, ibid. Kirk's argument is that the wording of Agamemnon's threat at 185, αὐτ⋯ς ἰὼν κλισίηνδε, requires us to understand αὐτ⋯ς ⋯πο⋯ρας at 356 and later as referring to personal participation on the part of Agamemnon in the physical removal of Briseis. I will argue in this paper that there is a shift in reference of αὐτ⋯ς from Agamemnon's initial threat, couched in general terms at 137ff., to the specific threat unambiguously directed toward Achilles at 184f., and that Achilles and others, ignoring the specific detail of 184f., revert to the original and essential threat of autocratic action and, with it, to the force of αὐτ⋯ς found at 137 to denote such action. Kirk is clearly not entirely happy with the contradiction he feels obliged to admit, which puts an end to the episode's ‘psychological subtleties’ about which he writes with admirable insight as reflecting ‘a close observation on Homer's part of the vagaries of human character and behaviour’.

³ e.g., LSJ ⁹, Cunliffe, , Lexikon des frühgriechischen EposGoogle Scholar, Chantraine, , DELGGoogle Scholar, all s.v. αὐτ⋯ς. For Homeric usage where the notion of physical participation is secondary or irrelevant see, e.g., Il. 1.133–4, 3.105–6, 7.96ff., 20.310ff.

⁴ That is to say, as the poet plays upon this ambiguity.

⁵ West, M. L., Hesiod: Theogony (Oxford, 1966), p. 21Google Scholar. For further examples of contemporary English usage see Garden, G. and Pesetsky, D., ‘Double-Verb Constructions, Markedness, and a Fake Co-ordination’, Papers from the Thirteenth Regional Meeting, Chicago Linguistic Society, University of Chicago, 1977Google Scholar. Clearly related to the ‘go-and-V’ construction is the idiomatic future with ‘go’; the semantics are quite different, however, as may readily be seen by comparing statements such as ‘He is going to kill her’ and ‘He'll go and kill her’.

⁶ See, e.g., The Oxford Latin Dictionary, s.v. eō, 10Google Scholar. In the imperative the verb often appears in asyndeton, as here. In Modern English ‘come’ as well as ‘go’ is used in double-verb constructions with no sense of motion, e.g. ‘Come, tell me’. (Unlike ‘go’, ‘come’ is largely confined to imperatival usage.) In Ancient Greek the imperatives ἄγε and (in post-Homeric Greek) φ⋯ρε frequently serve this semantic function.

⁷ Samarakis, A., ‘Grafeion Ideōn’, in Arnoumai (Athens, 1974), p. 28Google Scholar. I am indebted to Eirene Bouras for this reference.

⁸ cf. Garden and Pesetsky, op. cit. (n. 5). Their term ‘unexpected-event reading’ is perhaps too limited, as ‘go-and-V’ constructions appear to have a variety of uses, all of them highly charged semantically.

⁹ Stanford, W. B., The Odyssey of Homer², vol. 1 (London, 1959), on 6.282–4.Google Scholar

¹⁰ Kamerbeek, J. C., The Plays of Sophocles: Commentaries, Part III: The Antigone (Leiden, 1978)Google Scholar. Cf. Kamerbeek, on OT 431 (Commentaries, Part IV [Leiden, 1967])Google Scholar, on the use of ποῖος to imply ‘the inquirer's scorn or surprise’.

¹¹ Jebb, R. C., Sophocles: The Plays and Fragments, Part III: The Antigone (Amsterdam, 1962).Google Scholar

¹² Which of itself argues for accepting the problematical πρ⋯ξω, rejected by, among others, Denniston, J. D. and Page, D., Aeschylus: Agamemnon (Oxford, 1957)Google Scholar and Fraenkel, E., Aeschylus: Agamemnon (Oxford, 1950)Google Scholar, though printed by the latter. (Nestor's comment to Agamemnon at Il. 9.106–7, κο⋯ρην | χωομ⋯νου Ἀχιλῆος ἔβης κλισ⋯ηθεν ⋯πο⋯ρας, may also be explained by the ‘unexpected-event’ reading of ‘go-and-V’ constructions, with the additional nuance of disapproval.)

¹³ The interpretation advanced here obviates, among other things, Dawe, R. D.'s worries about the position of ἰών in its clause (CJ 84 [1988], 71–2).Google Scholar

¹⁴ Albeit not reflected in the surface syntax, as often in Greek (see above, p. 18).

¹⁵ Earlier versions of this paper were read at meetings of the Classical Association of Canada (Windsor, Ont., May 1988) and the American Philological Association (Baltimore, January 1989). I would like to thank especially Rainer Friedrich and Mark Edwards for their encouraging comments.