Austro-tai

She thinks the evidence for Austro-Tai is more convincing than for Dene-Yeniseian.

Simultaneous membership of Tai-Kadai in Li Fang-kuei's Sino-Tibetan and Benedict's Austro-Tai had posed a problem to Zhengzhang, as he had found both of these mutually exclusive proposals to be well-grounded: Sagart's 1990 suggestion that Chinese and Austronesian are genetically related, soon supported by Xing Gongwan, provided a solution to the conundrum, leading Zhengzhang to the present proposal.

The paper contained some newly-proposed Austro-Tai etymons.

One striking illustration will suffice here: the writer had not even included IN °kutu, Thai (T) °hraw "head louse"; in his list of possible correspondences, but the KS forms (Sui tu, Mak tau, Then tiu) led him to reconstruct T °[t]hraw, and he was delighted to uncover OB kat "lice"; he was then able to interpret the already recognized Li cognate sau~su as a development from a °tl- initial cluster, exactly paralleling IN °təlu "3", Li śu~su, and on this basis was able to reconstruct Austro-Thai (AT) °kut(a)lu.

Sagart rejected the possibility that Proto-Kra-Dai and Proto-Austronesian are sister languages, therefore rejecting the Austro-Tai hypothesis in its original formulation.