Комментарии:A diphthong has to be reconstructed because of the *-j-reflex in Jpn.; cf. also Orok pidul- 'to freeze completely' (of a river) (ТМС 2, 37).
Комментарии:Cf. perhaps also Man. buǯu- 'to cook': the form may go back to PTM *büdigu-. MKor. phɨ́r < *pɨ́rh < *pɨ́rVh < *pɨ́dVh (with an early vowel reduction, which explains the reflex -r as originally intervocalic).
Комментарии:KW 58, Владимирцов 174, Poppe 21, 53, Лексика 224. Despite Лексика 223, Turk. *budra 'curls' (ЭСТЯ 2, 245) very probably represents the same root as *bɨdɨk 'moustache' and also belongs here. Jpn. *pí- should be regarded as an early contraction < *bi̯uj- < *bi̯udu.
Комментарии:Martin 246. Low tone in Jpn. is probably due to contraction. In Mong. one has to assume *büɣü-rge < *biɣe-r-gü. Cf. Manchu faχun 'rim of a wheel' ( < Kor.?).
Комментарии:Whitman 1985, 139, 245 (Kor.-Jpn.). Loss of final resonant in Kor. makes us reconstruct the medial -j-, which also explains several other phenomena: *-i-vowel in PTM (-u- would be expected after a labial); *b- in PJ (*p- would be expected before i). Korean has also lost the initial *b-, as in some other cases; cf. in this respect the interesting Old Koguryo and Silla forms: Old Koguryo *wöl, Silla *ŏl (see Miller 1979, 9). Jpn. *bì < *bi̯ujr(i)-gV.
Комментарии:EAS 147, KW 55, Poppe 56, ОСНЯ 1, 191. Despite Doerfer MT 56, TM is hardly borrowed from Mong., and (despite TMN 2, 352) the Turk. and Mong. forms are certainly related. Cf. also Turk. *bok- 'to cross (legs), bend (knees)' (VEWT 79, EDT 311), WMong. boki- (KW 49) 'to bend', reflecting a back-row variant of the same root. Part of the Mongolian (*bög-) and TM forms seem to go back to a variant *bi̯ŭ̀ki.
Комментарии:ТМС 1, 108 (TM-Mong.). An expressive root with difficult reconstruction. The two variants in Mong. and TM are probably due to interdialectal borrowing, but direct borrowing from Mong. into TM is hard to justify: the meanings are not in mutual correlation (Mong. bilka- is 'overflow', while bilki- in TM is 'to wet, moisten'; Mong. bulka- is 'to dip, rinse', while TM bulku- is 'to splash'). The Korean reflex is not quite certain (Martin KED 836 considers pulli- to be a factitive of pūd- 'swell' - which is, however, somewhat questionable).
Комментарии:SKE 184, 192, Martin 243. The Jpn. form is dubious because of irregular devoicing and somewhat aberrant semantics; cf. also an odd variation between *pằr- and *pǝ̄r- in Kor., suggesting that we may be dealing with more than one root here: one of them could have also resulted in Mong. bara- 'to end, finish'. Turkic has a peculiar disyllabic structure and Ramstedt may be right in suggesting an old compound; for the second part cf. *ek- `to sow' < `to throw' (v. sub *p`èk`a).
Комментарии:KW 71, SKE 198, ОСНЯ 2, 99-100, Лексика 183. In Turkic one would rather expect *bir-: this variant is indeed reflected in most Oghuz languages; others may have reintroduced -ü- under Mongolian influence.
Комментарии:Korean has a usual loss of narrow vowel between a stop and a fricative. The back row in PT is not quite regular (*büs- would be expected). It might be better to reconstruct *biso ( > Turk. *bɨs-, with a subsequent labial assimilation > *bus-). In TM cf. perhaps Nan. busĩ 'rodents' stores' (ТМС 1, 115).
