up.^83 One may say that bipolarity (not trifunctionality) is the fundamental
structuring principle of Indo-European thought.
For example, the concept of ‘all intelligent beings’ is expressed by ‘gods and
men’ or ‘immortals and mortals’: RV 1. 35. 2 amr ́
̇
tam mártiyam
̇ca; 6. 15. 8
deva ̄ ́sas ́ ca mártiya ̄sas ́ ca; Y. 29. 4 = 48. 1 dae ̄va ̄is ˇc a ̄ masˇ
̇
ya ̄is ˇc a ̄;Il. 2. 1 θεο τε
κα? α, νρε; 20. 64 θνητο4σι κα? α, θανα ́ τοισι; Lokasenna 45. 3, 55. 6 goð
o ̨ll ok gumar; Gylf. 21 guðanna ok manna.^84
Where the focus is not on the gods but on the animal world, the phrase is
‘creatures two-footed and four-footed’: RV 4. 51. 5 (and often) dvipa ̄ ́c
cátus
̇
pa ̄t; Y. 9. 18 bizan
̇grana ̨m ... caθβarəzan
̇grana ̨m, cf. 19. 8, Yt. 1. 10; Vd. 15.
19 bipaitisˇtanaca caθβarəpaitisˇtanaca; on the Iguvine tablets, VIb. 10 dupursus
peturpursus; Cicero, De domo 48 ministro omnium non bipedum solum sed
etiam quadrupedum impurissimo.^85 The same classification is implicit in the
Hittite tale of the cow that gives birth to a human child after being impreg-
nated by the Sun-god. She remonstrates: ‘Now I ask you please: [a calf] should
have four legs. Why have I borne this two-legged thing?’^86 It also underlies
Aeschylus’ kenning-like expressions δπου Zφι and δπου λαινα,
‘two-legged snake’, ‘two-legged lioness’ (Supp. 895, Ag. 1258): in each case the
beast is a metaphor, and the epithet identifies its reference as a human one.^87
It may be noted that in some of these examples there is no ‘and’ joining the
words for four-footed and two-footed. Further instances of such asyndeton
will appear in the following paragraphs. It is no doubt an ancient traditional
feature of these pairings.^88
The estate-owner’s livestock is summed up as ‘herds and men’ (i.e. slaves):
RV 5. 20. 4 góbhih
̇
... vı ̄ráih
̇; Y. 31. 15 pasə ̄ usˇ v ı ̄ra ̄at
̃ca ̄, 45. 9 pasu ̄sˇ v ı ̄rə ̄ n
̇g,
cf. Yt. 10. 112; Tab. Iguv. VIa. 32 (and often) ueiro pequo; Ovid, Met. 1. 286
pecudesque uirosque.^89 In most of these the original words péu- and
wı ̄ ́ro have survived, and in that order; the Umbrian ueiro pequo, though it has
the inverse order, seems to be a dual dvandva, a remarkable archaism of a
type known from the Rigveda, by which two terms forming a pair are both
(^83) Gonda (1959), 337–47; Campanile (1977), 98–104; Watkins (1995), 44–7, 250 (‘merisms’).
(^84) Many other passages could be quoted, and some will be in the next chapter.
(^85) Further material in Schmitt (1967), 210–13; cf. R. Lazzeroni, SSL 15 (1975), 1–19; Watkins
(1994), 650; Gamkrelidze–Ivanov (1995), 394 f., 398.
(^86) CTH 363 iii 20ff., trs. Hoffner (1998), 86.
(^87) Campanile (1990b), 29, refers to Y. 9. 18 and Vd. 18. 38, where he says that the four-legged
wolf is distinguished from the two-legged wolf; but no two-legged wolf appears in those texts.
(^88) Cf. P. Chantraine, Revue de philologie 27 (1953), 16–20; B. K. Braswell, A Commentary on
the Fourth Pythian Ode of Pindar (Berlin–New York 1988), 300.
(^89) More in J. Wackernagel, ZVS 43 (1910), 295–8 (=Kl. Schr. 280–3; Schmitt (1968), 30–3);
cf. Schmitt (1967), 16 f., 213–16; É. Benveniste in Cardona et al. (1970), 308 f.; Watkins (1994),
649 f.
100 2. Phrase and Figure