most numerous class of these in the Indo-European traditions consists of
compound adjectives and nouns.
Nominal composition was a characteristic feature of Indo-European, and
the language provided its users with a set of templates according to which
new compounds could be formed whenever required, just as in modern
English we can freely improvise new compounds on the patterns of ‘dish-
washer’, ‘mind-bending’, ‘kitchen-cum-diner’, ‘pro-hunting’, or ‘ex-convict’.
In the ancient languages many compounds came into being to serve com-
monplace needs. Early examples can be reconstructed from Sanskrit
durmanas- = Avestan dusˇmanah- = Greek δυσμεν , all meaning ‘ill-
spirited’ in one sense or another, or from Vedic sudína-= Greek ε1δα,
ε1διεινο ́ , ‘good sky, fine day’, with its antonym surviving in Old Church
Slavonic du ̆zˇdı ̆, Russian дождь‘rain’. There is nothing intrinsically poetic
about these.
Poets, on the other hand, create and recycle compounds for decorative
rather than pragmatic purposes. In Vedic, Greek, and Germanic verse we
constantly encounter compounds which cannot have been current in normal
speech and which serve to enrich the utterance with a condensation of associ-
ated ideas. In a limited number of cases parallel formations in different
languages (with the same roots but sometimes a different stem) may point to
common inheritance. The Homeric α, νδροφο ́ νο‘man-slaying’, an epithet
of heroes and of the war-god, is composed of the same elements as Vedic
nr
̇
hán-, which is applied to Rudra and to the Maruts’ lightning-weapon. They
appear in inverse order in the Avestan hapax jə ̄nəra-.^14 In Hesiod’sGκυπτη
Aρηξ‘swift-flying hawk’ the epithet corresponds to that in RV 4. 26. 4 s ́yenáh
... a ̄s ́upátva ̄‘swift-flying eagle (or falcon)’.^15 Homeric beds are $Pστρωτα ̇,
which corresponds to Avestan hustarəta- (of the couches of the righteous, Yt.
- 9). Birds of prey are Gμηστα‘raw-eaters’ in Homer, a ̄ma ̄ ́d- in the Vedas
(RV 10. 87. 7, AV 11. 10. 8), both from o ̄mó-+ ed.^16 When Empedocles
calls the gods δολιχαωνε (B 21. 12, 23. 8; P. Strasb. a(ii) 2), he may have
known the word from older poetry, or put it together himself; in any case it
is matched by Vedic dı ̄r g h a ̄ ́yu- (of Indra, RV 4. 15. 9) and -a ̄ ́yus
̇
-, Avestan
darəga ̄yu- (Y. 28. 6, 41. 4).^17
There are several examples of a curious type of compound in which a
modifying prefix is attached to a proper name. In the Atharvaveda (13. 4. 2,
- we findMahendrá, that is, maha-Indra, ‘Great Indra’. In the Iliad we
(^14) Schmitt (1967), 123–7; H. Schmeja in Mayrhofer et al. (1974), 385–8; Durante (1976), 97;
Campanile (1977), 118 f.; (1990b), 60.
(^15) Hes. Op. 212; cf. Schmitt (1967), 236 f.; Campanile (1990b), 155; Gamkrelidze–Ivanov
(1995), 455.
(^16) Gamkrelidze–Ivanov (1995), 603. (^17) Schmitt (1967), 161 f.; Schlerath (1968), ii. 164.
80 2. Phrase and Figure