forever and there is no satiety; where food and drink do not diminish when
consumed; where to wish for something is to possess it; where a hundred
years are as one day.’^34
Celestial riverHomeric rivers sometimes have the puzzling epithet δι(ε)ιπετ . Adjectives
in -πετ (es-stem) normally mean ‘falling’ in some defined way, but this
does not go well with the apparent locative δι, which should mean ‘in the
sky’.^35 Alternatively it may be interpreted as ‘flying’, like -πτη (a-stem) in
0ψιπτη, Gκυπτη. It must have been so understood quite early, because
we also findδιιπετ applied to birds (Hymn. Aphr. 4). ‘Flying in the sky’ is a
clear and coherent concept. But what sense does it make in reference to rivers?
It is evidently an old, stereotyped formula that may reflect an obsolete
piece of cosmology. Heinrich Lüders collected Vedic evidence for the idea of
celestial rivers, in particular a celestial Indus. One passage that he cited in
this connection was RV 2. 28. 4, where it is said of the rivers váyo ná paptu ̄
raghuya ̄ ́ párijman, ‘swift as birds they fly their circuit’.^36 Whether this particu-
lar verse refers to rivers in heaven is disputable. But certainly the river Saras-
vatı ̄ is summoned to the sacrifice ‘from high heaven, from the mountain’
(5. 43. 11 divó br
̇
hatáh
̇, párvata ̄t); she fills not only the terrestrial regions but
the broad atmospheric spaces (6. 61. 11). In post-Vedic mythology the Ganges
is conceived as having its source in heaven (MBh. 3. 107. 21–108. 18; 5. 109. 6;
- 71, 89). It flows in the sky among the gods (MBh. 1. 158. 18; Rm. 1. 36.
7; it fell from the sky, ibid. 41. 22–42. 16). In the Avesta the holy river Arədv ı ̄ is
said to have come down to earth from the stars at Ahura Mazda ̄’s command
(Yt. 5. 85, 88).
Two Mycenaean gold rings, usually dated to Late Helladic II, are embel-
lished with cult scenes. In the sky is the sun (in one case also the moon), and
this celestial field is demarcated by a wavy band which looks more like a river
than anything else.^37 It has been suggested that it represents the Milky Way,
- 71, 89). It flows in the sky among the gods (MBh. 1. 158. 18; Rm. 1. 36.
(^34) Dillon (1948), 101, cf. 104 f.; Vendryès (1948), 308.
(^35) The long second vowel in the compound is the regular outcome of metrical lengthening.
The critic Zenodorus spelled the word διειπετ-, which is also written in the papyrus of
Euripides’Hypsipyle fr. 752h. 31 (cf. fr. 815); but this would imply an old dative, as in
δι(ε)φιλο, which would be inexplicable. On the problems of the word cf. Schmitt (1967),
44–6, 221–36; E. Risch, Gnomon 41 (1969), 325 f.; J. T. Hooker, IF 84 (1979), 117–19.
(^36) H. Lüders, Va r u n
̇
a (Göttingen 1951–9), 138–44, 146–51. Lüders was exercised by the
interpretation of δι(ε)ιπετ (677–9), but it was his posthumous editor L. Alsdorf who, in
the annotation on p. 143, adduced RV 2. 28. 4 as the key to it.
(^37) M. P. Nilsson, The Minoan-Mycenaean Religion (2nd edn., Lund 1950), 179 fig. 83; 347
fig. 158.
350 9. Cosmos and Canon