¹ Two referees have acutely pointed out problems with the method. First, the work of Irene de Jong (1987) has blurred the crude distinction I wish to maintain between ‘what humans say’ and ‘what the poet says’. Second, (here I quote the other referee) ‘I would wish that the ‥ distinction between what “Homer” says and what “his heroes” say was qualified with reference to the variables of emphasis and projection … sometimes it does not matter “who is talking” (because it hardly impinges on us). At other times it does impinge and it does matter’. To the first, I think I must say that if de Jong's work invalidates my thesis, so be it. I cannot see myself that it does. To the second, I think the weight of evidence for what I am arguing is so overwhelming as to override the ‘variables of emphasis and projection’ (which do, of course, exist). In other words, the heroes' view of the gods is so consistent throughout the poem that such variables, in this case, do not add up to enough to disturb the general thesis.

I refer to the following works by name and date: Clay, J.S., The wrath of Athena (Princeton 1983)Google Scholar; de Jong, I.J.F., Narrators and focalizers: the presentation of the story in the Iliad (Amsterdam 1987)Google Scholar; Edwards, M.W., Homer, poet of the Iliad (Johns Hopkins 1987)Google Scholar; Griffin, J., Homer on life and death (Oxford 1980)Google Scholar; Heubeck, A., West, S. and Hainsworth, J.B., A commentary on Homer's Odyssey vol. I introduction and books i–viii (Clarendon 1988)Google Scholar; Hainsworth, J.B., The Iliad: a commentary vol. iii books 9–12 (Cambridge 1993)CrossRefGoogle Scholar; Janko, R., The Iliad: a commentary vol. iv books 13–16 (Cambridge 1992)Google Scholar; Jones, P.V., Homer: Odyssey 1 and 2 (Aris and Phillips 1991)Google Scholar; Jørgensen, O., ‘Das Auftreten der Götter in den Büchern ι–μ der Odyssee’, Hermes xxxix (1904) 357–82Google Scholar; Kirk, G.S., The Iliad: a commentary vol. i books 1–4 (Cambridge 1985)CrossRefGoogle Scholar; Kirk, G.S., The Iliad: a commentary vol. ii books 5–8 (Cambridge 1990)CrossRefGoogle Scholar; Morrison, J.V., Homeric misdirection (Ann Arbor 1992)Google Scholar; Mueller, M., The Iliad (London 1984)Google Scholar; Richardson, S., The Homeric narrator (Nashville 1990)Google Scholar; Rihll, T., ‘The power of the Homeric βασιγεῖς’ in Pinsent, J. and Hurt, H.V. (eds), Homer 1987 (Papers of the Third Greenbank Colloquium April 1987, Liverpool Classical Papers no.2), (Liverpool 1992) 39–50Google Scholar; Rutherford, R.B., ‘Tragic form and feeling in the Iliad’ JHS cii (1982) 145–60CrossRefGoogle Scholar; Samuel, A. E., The promise of the west (Routledge 1988)Google Scholar; Schein, S.L., The mortal hero (Berkeley 1984)Google Scholar; Silk, M.S., Homer: the Iliad (Cambridge 1987)Google Scholar; Taplin, O.P., Homeric soundings (Oxford 1992)Google Scholar; Thalmann, W.G., Conventions of form and thought in early Greek epic poetry (Baltimore 1984)Google Scholar; Willcock, M.M., The Iliad of Homer books 1–12 (Macmillan 1978).Google Scholar All otherwise unmarked references are to the Oxford text of the Iliad. I am extremely grateful to Professor Alan Sommerstein and the JHS referees for their help, as I am to M.M. Willcock (University College London) and David West (University of Newcastle upon Tyne), who submitted an early draft of this paper to a searching ἔλεγχος from which it emerged battered but considerably improved.

² The distinction that Homer maintains between his full, privileged understanding of events (expressed in the narrative) and human, partial understanding (expressed in what characters say) has been investigated by Jørgensen (1904), cf. Clay (1983) 1–25, Richardson (1990) 123–39, Rutherford, R.B., ‘The philosophy of the Odyssey’, JHS cvi (1986) 153 n. 43Google Scholar, Winterbottom, M., ‘Speaking of the gods’, G&R xxxvi no. 1 (1989) 33–41.Google ScholarCf. the well-known phenomenon of human and divine proper names for the same thing (see e.g. Kirk [1985] on Il. i 403–4). de Jong (1987) 214 puts the case for the sort of analysis I wish to make as follows: ‘when analysing divine interventions in the Il. one should distinguish systematically between the presentation and interpretation of NF₁ [i.e. the poet] and of the speaking characters. Differences between the two versions should not be ascribed, I think, to differences in religious belief or concepts between NF₁ and characters, but to a difference in narrative competence (the NF₁ is omniscient and knows more than the characters) or rhetorical situation’ (here de Jong gives the example of Paris wishing to excuse his defeat vis-à-vis Helen). Taplin (1992) 129 says ‘The Iliadic gods are a mixture of awesome power and quarrelsome pettiness, reflected in ethics by their mixture of roles as guarantors of justice and as amoral self-seekers’. The question I wish to clarify is ‘in whose eyes?’

³ Compare, for example, v 719–52, viii 41–77, xiii 17–31, xiv 346–51, xvi 431–61, 644–93, xvii 441–55, xviii 478–613 (and cf. Schein [1984] 51–2). See also e.g. oaths and sacrifice at n. 6, and the miracles on p.l 11. One may argue about the precise extent to which these passages demonstrate divine glamour and majesty (as a referee pointed out); but that humans never talk in these terms goes without saying.

⁴ So, e.g. ‘In the context of a society over which the Olympian gods rule, Achilles is pursuing an almost hopeless task…human success or failure can only be attributed to the whims or wills of the gods, fate, or both’ (Samuel [1988] 45). ‘[The gods] function as a higher power, and provide an explanation of otherwise inexplicable events’ (Edwards [1987] 125). ‘For the human characters in the Il., irrational evil comes from the gods’ (Edwards [1987] 128, though he goes on to point out that for the poet, these evils are not irrational ‘if one believe in gods like these’). I stress that these quotations are selected to serve my purpose: they are not supposed to characterise the whole picture of divine activity discussed in these works, which are extremely valuable and on which I shall draw in the course of this paper. I am obviously more in sympathy with e.g. Silk (1987) 30 and Mueller (1984) 125–33.

⁵ Mueller (1984) 126 is aware of the reasonableness of the interaction between men and gods: when a god intervenes, ‘the outcome is always an action that is perfectly intelligible in human terms.’

⁶ Other sacrifices and oath-ceremonies are referred to with more or less elaboration at e.g. ii 305–7, 402–31, iii 268–301, iv 44–9, viii 548, ix 357, xi 726, 771, xxiv 33, 65–70. If I were to argue against my thesis, I would concentrate on passages like these, especially where the heroes call on the gods to witness oaths and curses. It is only here that I, at any rate, get any sense of the gods' numinous majesty expressed in a human's words, e.g. ii 402–18, iii 267–301, ix 453–7, 561–72, xix 257–68. Nevertheless, the ritual context of such passages is very strongly marked. This is special language for special events (cf. Leumann, M., Homerische Wörter [Basel 1950] 22–23Google Scholar). In Homer, such language is restricted to ritual occasions.

⁷ A referee points out that Achilles' amazement may not be due to this, comparing e.g. iii 398 and iv 97 where the way the divinity looks to the human is enough to create θάμβος (cf. Richardson, N.J., Homeric hymn to Demeter [Oxford 1974] 188–90Google Scholar, though he does not deal with this passage). But at i 199, as the poet makes crystal clear, Achilles has not yet seen Athena because she approached him from behind. All he has done is felt her tugging his hair.

⁸ If this is what γλαυκῶπις means: cf. e.g. Kirk (1985) on i 200.

⁹ A referee draws my attention to Zeus's ‘compulsion’ of Achilles at xxiv 116, expressed in the same way. This is how gods and humans frequently interact in the Il.: Rihll (1992) 46 argues strongly that power is negotiable in the Il.: ‘neither Zeus nor Agamemnon have an unchallenged right to command’ and need to adopt different tactics (from bluster to persuasion) to get their way.

¹⁰ A referee adds v 407, where Dione tells Aphrodite how foolish Diomedes is to fight the gods—that man does not live long—and vi 128–41, where Diomedes informs Glaucus that he will not fight with him if he is a god.

¹¹ Even this analogy has its weaknesses. I can find, for example, only seven places where humans fear the gods (iii 418, v 827, 863, ix 244, xx 380, xxi 248, xxiv 170) and four where they fear Zeus's thunderbolt (vii 479, viii 77, 138, xvii 594–96). I discount xiii 624, xxiv 358, 689. As for humans fearing humans, I gave up counting when I reached fifty examples. Again, the heroes rarely acknowledge the gods even when their prayers are answered. They sometimes rejoice, like Glaucus at xvi 530–1 or Achilles at xxii 224 (though note that at xxii 393 Achilles claims the victory was all his doing), but more often than not they carry on without any acknowledgement at all, e.g. Ajax at xvii 645–55. A notable exception is x 570–1.

¹² In the Epic of Gilgamesh, Enkidu makes the same sort of comment to the goddess Ishtar after he has slapped her in the face with the shoulder of the Bull of Heaven: ‘If only I could get at you as that does, I would do the same to you myself’ (VII v in Myths from Mesopotamia, tr. by Dalley, S. [Oxford 1989]Google Scholar).

¹³ A referee rightly points out that the Myrmidons are afraid of the armour (xix 15). Here it is only Achilles who looks at it with pleasure. The referee adds xviii 205–6, but this is different. As far as Achilles is concerned, nothing miraculous is happening. Again, however dramatic the Trojan response to his appearance and shout, they do not acknowledge it as a miracle either.

¹⁴ i 37 (Chryses to Apollo to punish the Greeks), i 407 (Achilles via Thetis to Zeus for glory), i 451 (Chryses to Apollo to end the plague), ii 412 (Agamemnon to Zeus to destroy Troy and Hector), iii 320 (the armies to Zeus over the outcome of the duel), iii 351 (Menelaus to Zeus to have revenge on Hector), iv 119 (Pandarus to Apollo to kill Menelaus), v 115 (Diomedes to Athena to kill Pandarus), vi 305 (Theano to Athena to kill Diomedes), vi 240 (Hector tells the women to pray to the immortals), vi 476 (Hector to Zeus concerning his son), vii 179 (Greek troops to Zeus, about the winner of the draw to fight Hector), vii 202 (Greek troops to Zeus that Ajax win), viii 242 (Agamemnon to Zeus that the Greeks be not destroyed), viii 346–7 (Greeks to all the gods under Hector's onslaught), viii 526 (Hector to Zeus and the other gods that he will rout the Greeks), ix 170 (Nestor to Zeus for his mercy), xi 183 (the embassy to Achilles, to Poseidon), xi 454 (Phoenix's father to the furies), xi 568 (Meleager's mother to Hades and Persephone), x 278 (Odysseus to Athena for glory), x 284 (Diomedes to Athena for protection), x 462 (Odysseus to Athena for guidance to the Thracian camp), xi 735 (Nestor and his men before battle, to Zeus and Athena), xv 372 (Nestor to Zeus that the Greeks be not destroyed), xvi 233 (Achilles to Zeus for Patroclus' safety), xvi 514 (Glaucus to Apollo to heal his wound), xvii 45–6 (Menelaus to Zeus before attacking Euphorbus), xvii 498 (Automedon to Zeus for courage (?)), xvii 645 (Ajax to Zeus to shed light on the battlefield), xxiii 194 (Achilles to the winds to set Patroclus' pyre alight), xxiii 770 (Odysseus to Athena to give him speed), xxiii 871 (Meriones to Apollo to hit the target), xxiv 308 (Priam to Zeus to grant him an omen for a safe journey to Achilles).

¹⁵ The same holds for prayers offered to Zeus to witness events or seal oaths (iii 276, 298, vii 76, 411, xix 259); ‘statement’ prayers, where a god is invoked, though not asked directly for help (e.g. iii 365, xii 164); and wishes (ii 371, iv 288, vii 132, x 329, xii 275, xvi 97, xvii 561, xviii 8, xxiii 650). See also n. 6 and Bremer (n. 16) on how comparatively ungrateful the heroes seem to be for the gods' help.

¹⁶ See Bremer, J.N., Greek religion (Oxford 1994) 39.Google Scholar

¹⁷ In n. 4, I disagreed with Edwards (1987) who suggested that the gods acted irrationally in men's eyes (though cf. Edwards (1987) 136, where he rightly says ‘the poet needs to satisfy his audience's desire to find an order and rationality in human experience’). Achilles' speech here seems to me to support my disagreement. The rationality of the gods' intervention in human life, expressed in terms of (e.g.) quid pro quo, just deserts, or however it might otherwise be expressed, is simply not raised. Life, says Achilles, is not irrational. It is simply lived under divine control. In human eyes, then, the gods' acts may seem capricious or unpredictable—but that is not the same as irrational. Interestingly, the only time that the issue of human deserts is raised is in relation to τιμή and there, of course, we are talking about human deserts in human eyes—a very different, and deeply contested, issue (as Taplin [1992] 50–1 rightly emphasises).

¹⁸ Janet Watson points out to me by letter that only major Greek heroes (Achilles, Odysseus and Diomedes) converse with undisguised gods. Lesser heroes, she goes on, like the Aiantes, ‘may be aware that a god has addressed them in the likeness of a mortal but do not know which one’ (and she cites xiii 68–72). This observation seems to me at one with the general argument of this paper.

¹⁹ The exceptions are xix 87 where Μοῖρα is associated with Zeus and the Erinyes, and xxvi 49 where Μοῖραι are said to give men an enduring heart. In these places it is clearly personified as a god. My analysis is rather different from that of Schein (1984) 62–63.

²⁰ A referee astutely points out that all these references are put in the mouths of the characters.

²¹ See further S. West (1988) on Od. iv 379–81, who shows (with examples) that ‘Homer's gods are omniscient in a rather limited sense’. Greek tragedy also manipulates fate inconsistently for, I would argue, a similar literary effect: cf. e.g. the oracles in Sophocles' Trachiniae and Philoctetes (see Davies, M., Sophocles' Trachiniae [Oxford 1991] 268–9Google Scholar). Homer deals with Patroclus' fate more consistently. At viii 477 Zeus announces it is ordained (θέσϕατον) for him to die, and at xvii 268–73 movingly helps to protect him: he had not hated him while he was alive, comments Homer, impressing on us the needlessness of Patroclus' death. At xviii 9–11 Achilles tells us that he knew all along from his mother that ‘the best of the Myrmidons’ would die at Troy, which he now sees meant Patroclus. At xix 328–33, Achilles says he had hoped he alone would die at Troy and Patroclus would return.

²² Hainsworth (1993) on xii 237–43 points out that epic takes a rational view of omens, regarding them as confirmation or discouragement of decisions already taken, rather than allowing them to determine the action.

²³ On Homeric ‘misdirection,’ see Morrison (1992), cf. de Jong (1987) 68–81. Taplin (1992) 198 describes the changing revelations as ‘the Homeric technique of increasing precision’.

²⁴ The efforts made by the gods constantly to thwart the will of Zeus (cf. viii 5–12) and divert the course of action so clearly predicted in places such as viii 473–7, xv 72–7 and xvii 596–614, and Zeus's own desire to change fate (e.g. xvi 431–61—admittedly fruitless, cf. xxii 167–85) add to this effect (in the readers' view) of the negotiability of existence. If the gods can play like this with the will of Zeus, and Zeus himself seems in theory able to change fate (cf. xvi 443=xxii 181), what price inevitable fate? How helpless are humans in its grasp? For the fluctuation of events in Homer, see Morrison (1992) 95.

²⁵ And, I would argue, accords with human experience. Many people feel that the decisions they take are entirely their own; but many of the same people at the same time look back over their lives and have the sense that God was guiding them. We are no nearer than Homer to solving the problem of divine omnipotence, free will and responsibility for action. In fact, Homer's solution (that both men and gods are 100% responsible) is remarkably appealing. Cf. Schein (1984) 58, Thalmann (1984) 85–6. Gaskin, R., ‘Do Homeric heroes make real decisions?’, CQ xl (1990) 1–15CrossRefGoogle Scholar (especially 6–7) is an excellent analysis of that particular problem, demonstrating conclusively that they do. This has important implications for the arguments about heroic freedom and independence in this paper.

In this respect, it is worth saying how useful a multiplicity of gods is to the poet (see further Edwards [1987] 121–42). This is the means of creating conflict in Olympus, which can be used to make sense of the swinging fortunes of men on earth (a device as old as Gilgamesh). The gods can contest among themselves the issue of their favourites (e.g. i 493–567, xiii 345–60, xv 89–238, xvi 444–9, xvi 354–67, xxiv 23–76 and the battle among the gods in xxi), and can deceive one another as they go about their business (cf. e.g. Apollo, learning late of Athena's schemes at x 515, and Poseidon's interventions and the deception of Zeus in xiii–xiv): see how dejected they are when they cannot intervene (xii 179–180). Men, in other words, have a chance. As they often say, the gods' favours constantly shift. Life would be intolerable if they did not.

²⁶ Schadewaldt, W. in ‘Die Entscheidung des Achilles’ (Von Homers Welt und Werk [Leipzig 1965])Google Scholar argues that in Achilles Homer created the first image of human freedom in the West. Cf. Rihll (1992) 50 ‘[Achilles] seeks his own freedom; freedom of action and freedom to live’, and Gaskin (n. 25) 15. For the Il.'s human dimension, cf. de Jong (1987) 228: ‘I submit that the Il. mainly presents a human vision of the events around Troy’.

²⁷ Janko (1992) 4 points out that Homer's handling has the effect of ‘leaving an undefined area between free will and natural forces…Homer's characters are seen to suffer for their choices, which is clearly tragic, and yet the whole outcome seems to be beyond their individual control or even preordained, which is tragic in another way’. Exactly. Cf. Rutherford (1982), a richly rewarding article on tragic elements in the Il. Griffin, J., ‘The epic cycle and the uniqueness of Homer’, JHS xcvii (1977) 39–53CrossRefGoogle Scholar and Davies, M., The epic cycle (Bristol 1989)Google Scholar between them draw out the contrasts between Homer and the Cyclic poets.

²⁸ In the light of this analysis, it is perhaps necessary to reassess some of the bolder generalisations about men and gods. Thalmann (1984), for example, talks of man being ‘ultimately insignificant’ (90), as does Schein (1984) 62. That is not the impression I get from the Il., let alone from the Od., and is certainly not the way the heroes view matters. Likewise, it is common to talk of the gods' combined triviality and grandeur (see e.g. Schein [1984] 52–3, Taplin n. 2 above). Since the heroes themselves never talk in these terms, the generalisation, I think, needs some refining.