Laudator Temporis Acti

Homer, Iliad 1.17-21 (Chryses, priest of Apollo, speaking; tr. Peter Green):

Atreus's sons, and you other well-greaved Achaeans,
may the gods who have their homes on Olympos grant you
to sack Priam's city, and win a safe homecoming!
But release my dear daughter, accept the ransom I offer,
show respect for Zeus's son, Apollo, the deadly archer.

Ἀτρεΐδαι τε καὶ ἄλλοι ἐϋκνήμιδες Ἀχαιοί,
ὑμῖν μὲν θεοὶ δοῖεν Ὀλύμπια δώματ᾿ ἔχοντες
ἐκπέρσαι Πριάμοιο πόλιν, εὖ δ᾿ οἴκαδ᾿ ἱκέσθαι·
παῖδα δ᾿ ἐμοὶ λύσαιτε φίλην, τὰ δ᾿ ἄποινα δέχεσθαι,        20
ἁζόμενοι Διὸς υἱὸν ἑκηβόλον Ἀπόλλωνα.

G.S. Kirk on line 18:

Out of 88 Iliadic and 94 Odyssean occurrences of θεοί in the nominative this is the only one in which it is scanned as a monosyllable by synizesis (though cf. θεοῖσιν at Od. 14.251). It is certain that the poets were capable of expressing this kind of simple thought without the use of such a rare pronunciation (it became commoner later) and the abandonment of such an extensive and strict formular system. Bentley's ὔμμι θεοὶ μὲν δοῖεν is the best emendation proposed so far, the postponement of μέν being unusual but not impossible.

Simon Pulleyn on line 18:

θεοί Scanned as one syllable, the sounds ε and οι being run together in pronunciation; this is known as synizesis or synecphonesis. Out of a total of 182 occurrences of the nominative θεοί in Homer, this is the only one scanned as a monosyllable. In order to avoid this apparent anomaly, Bentley conjectured ὔμμι θεοὶ μὲν δοῖεν. Normally, one expects that the words which come immediately before μέν and δέ will be corresponding elements in a binary contrast; thus we would expect another nominative contrasted with θεοί doing something to or for the Greeks (ὔμμι). However, Homeric usage is not absolute on this point and more or less violent postponements of μέν are found (e.g. 13.13). Bentley's conjecture is thus not unattractive.

Martin L. West's Teubner edition of the Iliad (1998) records no variants or conjectures for line 18 in the critical apparatus.

Arthur Platt, "Notes on the Text of the Iliad," Journal of Philology 18 (1890) 126-133 (at 126-127, on 1.18):

It is commonly objected to this line that the synizesis of θεοί in the oldest part of Homer is impossible: Bentley accordingly conjectures ὔμμι θεοὶ μὲν δοῖεν. But this reading puts the μὲν quite in the wrong place; if μὲν is to qualify the whole clause it ought to be the second word in it, and it does not make any sense to make it qualify θεοί. It the line is to be altered at all it would be better to read: ὑμῖν μὲν τοι δοῖεν, for it would not be un-Homeric to use Ὀλύμπια δώματ᾿ ἔχοντες by itself for the gods, and this reading is such as might give rise to corruption and our text. I believe that it is a safe rule in Homeric criticism that the order of words is not to be meddled with except under pressing necessity, that the freer uncontracted forms of Homeric words have been everywhere cut down without remorse but that the words themselves have been preserved with most extraordinary fidelity. You can hardly read half a dozen lines of Homer without coming across forms of words which ought to be restored to an older form, but you may read all Homer through and scarcely come across a dozen lines where any plausible emendation has been made which alters the order of the words or the words themselves¹.

Now there is another place in Homer where the short form of θεοῖσι occurs, ξ.251. It is true that the Odyssey has many more short and contracted forms of words than the oldest part of the Iliad. But if θεοῖσι is contracted once and once only in the former, why not also once in the latter? And ξ.251 is a line which defies emendation; the only way to get rid of it would be the rough method of downright expulsion. And why should there be any difficulty about it? Will the opponents of all contraction and synizesis in Homer consider this simple fact? Homer uses three forms of the same infinitive, τεuxέμεναι, τεύχεμεν and τεύχειν. He uses three forms of the same genitive, Φοίβοιο, Φοίβοο and Φοίβου. In the oldest parts therefore he uses the uncontracted forms and two stages of contraction, uses them perpetually and without the least hesitation, and no one will dare endeavour to remove them or to explain them away. And it would be easy for the merest beginner in Homeric science to add more cases of the same kind. Perhaps it would be as well to talk less about contracted forms and short forms of the dative plural and so forth, until some one shall have "corrected" the "un-Homeric contractions" of ἄγειν and τοῦ in A.338, 340. And if θύρεον in ι.340 can be scanned as an iambus, why not θεός as a monosyllable?

¹ Cf. however on the other side Meurad [sic, read Menrad] de Contractionis Usu Homerico, p. 168. But most of his instances are puerile, and simply shew the weakness of his case.

Other conjectures include ὑμῖν μὲν δοῖεν πότ' (Nauck), ἀλλ' ὔμμι μὲν δοῖεν (Fick), δοῖεν μὲν θεοὶ ὔμμιν (Brandreth).