Comments:The root evidently denoted some k. of plant used for making ropes or yarn. Note however that the Turkic reflex is poorly attested and not quite reliable; in PTM a reconstruction *bartu-kin is also possible (and perhaps preferable). The PA form should thus perhaps be changed to *bằrtò - a Tung.-Jpn. isogloss.
Comments:АПиПЯЯ 77, Дыбо 11, Лексика 83-84. PJ has an irregular devoicing here (or should one think about a secondary assimilative voicing *baǯi < *paǯi in PTM? In that case PA *pàǯa should be reconstructed).
Comments:EAS 57, SKE 179, Martin 228, ОСНЯ 1, 172, АПиПЯЯ 68. One of the few common Altaic monosyllabic roots. Mong. *baɣu- 'to bind' is probably < Turk. (Щербак 1997, 103). Doerfer's (TMN 2, 254) criticism is unacceptable ("unklar, da kor. Nominalstamm, tü. Verbalstamm").
Comments:PT *bokursɨ probably < *bakursɨ (with a vowel narrowing in the first syllable of a trisyllabic word); PTM *bāgba < *bāk-ba through assimilation. Note a remarkable similarity of the Turkic and Japanese derivation. Cf. other similar roots: *p`i̯ùgV, *pŏ̀k`è, *p`ū̀ge, *p`àgò.
Comments:EAS 57, SKE 191, Poppe 21. "Verbal" low tone in Korean. Doerfer (TMN 1, 207) attempts to dismantle Ramstedt's comparison (Tung.-Kor.-Mong.) by preferring Ramstedt's own earlier (KW 35) derivation of Mong. baraɣun 'right; West' < bara- 'to end' - which is evidently much weaker semantically.
Comments:Gombocz 1905, KW 66, Poppe 53 (Turk.-Mong.), Martin 243-244, АПиПЯЯ 16, 68, 72, 280, Дыбо 12. Verbal low tone in Kor. Doerfer's (TMN 1, 235; 4, 275) criticism of the Turk.-Mong. match ("lautgesetzlich nicht vergleichbar;...passen im Vokalismus nicht...") is incorrect, because *-e- is well reconstructable for Proto-Mongolian. PTM *burgu is a contraction < *bedu-rgu (a similar phonetic development cf. in PTM *xürgü 'tail' < *k`i̯udo-rgV). In Kor. one would expect *pùrɨ̀-; the attested pɨ̀rɨ̀- is an obvious result of vocalic assimilation.