¹ The Herodotean examples (though not the Homeric) are said by Kühner-Gerthdot;, (Griechische Grammatik, i, § 396 (2)Google Scholar) to refer to past time from a present standpoint-‘may have said’, ‘are likely to have been’, etc.- and in some cases (1. 2; 2. 98; 5. 59) actually to refer to present time. Hdt. I. a is regarded by Goodwin as a ‘future realisation’ (Moods and Tenses, §238); sim. Thuc. 1. 9. The latter, however, is taken by Forbes to mean ‘they would not be called “many” by the poet‘ (see also Kiihner-Gerth, loc. cit.; this again makes the optative ‘present’ rather than either ‘future’ or ‘past’); and both Goodwin and Forbes deny that these have any connexion with the Homeric and other Herodotean ‘past optatives’. The translations ‘they might possibly be Cretans’ (Turner), ‘these may have been Cretans’ (Monro), etc., are too weak; the optative does not in itself imply doubt: ‘they would be Cretans’, ‘they (certainly) would not be many’, etc..

² Taken by Goodwin to refer to future, with meaning ‘may it prove that they have driven’ (M.T. § 93); this itself, however, is equivalent to a past wish-or, since there is admittedly nothing unfulfilled about it, a past hope: ‘may they have driven’ or ‘I hope they drove/have driven‘, and Goodwin himself, in speaking of ‘laxity of usage’ and comparing this with clear cases of the Homeric ‘past optative’, comes near to dispensing with the future interpretation altogether.

¹ For the very commonly held futuristic interpretation of the optative see Goodwin, , M.T., pp. 385 ff. and §3234f., 243, 401Google Scholar; Monro, , Homeric Grammar, §315Google Scholar; Thompson, , Greek Syntax, p. 133Google Scholar. Hahn: ‘the optative … got transferred to past unreal conditions in place of the past indicative’, Subjunctive and Optative(American Philological Association Monographs), § 129; sim. Goodwin, , M.T., p. 388Google Scholar (4). But the evidence is quite consistent with the optative being the original construction for all tenses and the indicative itself a later substitute.

² Goodwin and perhaps Monro occasionally seem to imply this (M.T., § 442; H.G., § 299), but they do not make the principle either original or operative throughout; on the contrary, it conflicts with their basic theory. Sim. Schwyzer, , Griechische Grammatik, ii. 320, 324, 328Google Scholar; see also Chantraine, , Grammaire homirique, ii, §§ 323–5.Google Scholar

¹ Such examples as , Gilbert; cod. B], Xen. Comm. 3. 5. 8, if correct readings, are probably ‘future’ rather than either ‘present’ or ‘past’, and represent a change of viewpoint. Eur. Bacch. 1343 has only been found puzzling because it has been thought that refers necessarily to the future.

² Some of the ‘past optatives’ already quoted may be of this kind. Eur. Supp. 764: perhaps general (if not ‘present’). Sim. Od. 18. 79: (‘would that you had neither existence nor birth’), though a past reference is fitting (‘would that you did not exist now, nor had ever existed’-or ‘come into existence’, inceptive aorist-‘before’). Moorhouse, in C.R. lxii. 61 admits that the mood may be timeless and rightly denies a past meaning; but he ignores the possibilityof a past application. It comes to very much the same thing here if we say that ‘to have birth’ (now or in general) involves, and to that extent includes, to have been born at some time in the past and that Antinous may well be allowed to be thinking of the latter rather than of the former: ‘would that you did not exist now’ (compatible with ‘would that you had perished long ago’); ‘nay rather, would that you had never been born in the first place’. Similar considerations apply t in II. 13. 826, and to in Hdt. 7. 11 where the perfect may be taken as having a present or timeless force. Eur. Hipp. 407: , “formelhaft', Wecklein (and cf. K.-G. i, §395 (3)); possibly without a specific temporal reference, therefore, though if there is one, it must be past. Hel. 1215 may be likewise taken as general, simply, or as having a mixed reference: ‘I hope he perished or may yet perish.’Google Scholar