Comments:EAS 146, KW 100, Владимирцов 360, Poppe 110, Ozawa 154-156, Murayama 1962, 108, JOAL 35-36, TMN 1, 329-330), АПиПЯЯ 71. In TMN 2, 608 Doerfer tries to deny the comparison with TM ("es gibt kein Lautgesetz mo. ö = tu. ü"), which is not justified.
Comments:Mong. has t- due to contamination with tog-si- 'to hit, knock' q. v. ; Kor. has tìh- with irregular vowel (low tone is usual in verbs) due to contamination with tìk- 'to dip down, imprint' q. v.
Comments:Medial -j- should be postulated to account for the vowel (*ü) in PTM and for the otherwise irregular fronting *u > i in Jpn. For a possible reflex in Turkic see under *t`ep`a.
Comments:The isolated Even parallel is somewhat dubious, but the rest of the forms are well explained by the original meaning 'perceive' (whence both 'listen' and 'ask') and correspond well to each other. {In Mong. cf. also WMo don-duɣ 'соображающий', donuj- 'осмысливать'; for Evn. duja 'noise' cf. Mong. düj 'возглас для подзыва собаки').}
Comments:The isolated Manchu parallel is somewhat dubious semantically (not a 'relative by marriage', but a coeval), but the Turkic and Jpn. forms correspond well to each other.
Comments:All meanings are well explainable from the original 'make, produce a sign'; the diphthong -jǝ- in Korean is somewhat unexpected and makes us reconstruct PA *tū́jk`ú (otherwise *tū́k`u would serve just as well).
Comments:A Turk.-Mong. isogloss. The relationship to TM *dō- 'to sit down (of birds)', suggested in ТМС 1, 211, is unclear; if it exists, we may be dealing here with an archaic case of *-p`-suffixation.
