Homeric ΔΙΙΠΕΤΕΟΣΠΟΤΑΜΟΙΟand the Celestial Nile

Homeric διιπετής, which occurs only in the verse–end formula διιπετέος (Il. 16.174, 17.263, 21.268, 326; Od. 4.477, 581, 7.284; cf. Hes. fr. 320 Merkelbach–West), is usually interpreted as "fallen from Zeus, i.e., from heaven, . . . fed or swollen by rain" (LSJ),¹ for high–thundering, cloud–gatherer Zeus is the sky who rains and snows (Il. 12.25; Od. 9.111, 14.457, Alc. Z 14.1 Lobel–Page 5 338.1 Voigt; Ar. Nub. 368, 373; Babrius 65.1; Marcus Aurelius 5.7), and rivers can be "swollen by rain and melted snow" (). Yet it only takes a moment for the desperate nature of this interpretation to become clear: it has at least four flaws.²

First, there is διι–. We may explain the quantity of the second iota (long in διιπετής, short in Διί) either as metrical lengthening, or as a survival of the Mycenaean , with or without the emendation to διειπετής proposed by one Zenodorus (schol. E²Od. 4.477, var. l. Zenodotus, HQ) and supported by some moderns,³ but the form itself is unexpected. Even supporters of the usual interpretation admit that διι is a very odd way of saying "from Zeus," which should be expressed by a genitive of source:⁴ indeed, when Euripides wants to characterize a statue as having "fallen from the sky," he uses διοπετής (IT 977; cf. fr. 971 Nauck² ). They account for διι- in one of three ways: either as chosen on the analogy of διίφιλος where the dative is appropriate,⁵ or as a compendious expression meaning "by (the command of) Zeus,"⁶ as an Arcado– Cypriot use of the dative as an ablative.⁷ This last approach relies in turn on the hypothesis (itself controversial) of an Achaean stratum of the Homeric language, in which are alleged to persist forms specific to the Mycenaean dialect.⁸

Second, the final element, πετής, is elsewhere in Homer derived from πέτομαι (Il. 12.201, etc.;Il. 8.42, etc.; cf. Hes. Theog. 267). Unlike διιπετής, these words are accented paroxytone (but note παλιμπετές Il. 16.395, Od. 5.27, which means "falling or flying backward"; Pind. Pyth. 3.105; and Aesch. Supp. 782), but this shows only that during the Alexandrian period, when the written accents were devised, scholars associated them with a different root from διιπετής. Leaving our word aside, for examples of -πετής derived from πίπτω we must look to Aeschylus, Pindar, and subsequent authors.⁹ (Words with πετής from πετάννυμι are attested in later Greek.)¹⁰

Third, the earliest post–Homeric uses of διιπετής (Eur. Bacch. 1267, Rhes. 43, Hypsipyle fr. 1. sp. IV, 1.31 Bond; Hippoc. Morb. Mul. 1.24) demand the translation "translucent" vel sim. (Etym. Magn. s.v.), and Zenodotus (apud schol. Il. 17.263) glosses the word in its Homeric use as διαυγής,¹¹ yet the semantic development from "fallen from Zeus, etc." (on the usual interpretation) to "translucent" can only be seen as "encore inexpliqué."¹²

Fourth, one of the three named rivers to which Homer applies διιπετής is the Nile (Od. 4.477, 581; the others are Spercheius, Il. 16.174, and Scamander, Il. 21.268, 326), yet if ever there were a river that is not "fed or swollen by rain" it is the Nile: rainfall is scarce in Egypt (Hdt. 3.10.3), and, knowing this, Diogenes hypothesized that the Nile was fed by subterranean springs (64 A18 Diels–Kranz; cf. Sen. Q.Nat. 4.2.28–30, 6.8.3–5; schol. Ap. Rhod. 4.269 = p. 277 Wendel). The application of the epithet to the Nile would have, therefore, to be classed among Homer's (admittedly numerous) formulary illogicalities.¹³

Not all scholars accept the usual view. In 1958 M. Treu pointed to a then newly discovered papyrus fragment of Alcman containing the adjective διαιπετής in the phrase (III 66–67 PMG, PMGF = fr. 26.66–67 Calame), which was taken to describe a star "falling through . . . , less prob. . . . flying through" heaven (LSJ Suppl. s.v.),¹⁴ to hypothesize that the Homeric instances of διιπετής are all miscorrections of an original διαιπετής, "flying through." This theory has won few adherents,¹⁵ and R. Renehan (1972b) has with more...