01
Origins
Kra-Dai languages are tonal, mostly monosyllabic and isolating, and centuries of contact with Chinese have given them large layers of Chinese vocabulary, including many numerals: Thai sǎam 'three' and sìi 'four' are Chinese loans. For a long time they were therefore classified with Chinese in Sino-Tibetan. In 1942 Paul Benedict argued instead that their core vocabulary links them with Austronesian ('Austro-Tai'), and Weera Ostapirat later showed regular correspondences in that core.
The homeland is usually placed in southern China, in the hills and coast of Guangdong and Guangxi, around 2000 BCE. The four main branches, Kra, Hlai, Kam-Sui and Tai, divided there before the Tai languages spread southwest into mainland Southeast Asia much later.
02
Rebuilding the Longer Word
Modern Kra-Dai words are short, but their ancestors were not. Comparison shows that many roots once had two syllables, or a weak first syllable before a stressed one, and that most branches later dropped the first part. Buyang, a Kra language, still says ma⁰ ta for 'eye', where Thai has taa; Laha has a similar form. Traces survive inside Tai too. Pittayawat Pittayaporn reconstructs Proto-Tai 'eye' as *p.ta and 'water' as *C̬.nam, each with a reduced first consonant, and clusters like these are the likely reason why Wuming Zhuang, as recorded by Zhang Junru and colleagues, has ɣa 'eye' and ɣam 'water' where Thai has taa and náam.
This matters for two reasons. It explains correspondences that would otherwise look irregular, such as Thai t against Zhuang ɣ, or Thai h in hǎw 'louse' against Kam and Buyang t. And it makes the comparison with Austronesian possible at all: Austronesian *mata 'eye' and *maCay 'die' match the Kra-Dai words only once the first syllable is restored. Ostapirat's reconstructions of Proto-Kra-Dai therefore have two syllables, *maTaː and *maTaːj, but many details, written with capital letters, are still unresolved, and different scholars reconstruct the same roots differently.
Where its languages are spoken
- Thailand
- China
- Vietnam
- Laos
- Myanmar
Turning points
- SplitContested
Four branches in southern China
Kra, Hlai, Kam-Sui and Tai diverge in southern China and on Hainan. The date and the homeland, on the mainland coast or, on Sagart's view, Taiwan, are disputed.
Sources (2)
- Contact
Chinese expansion south
As Chinese states and settlers push into the south, Kra-Dai languages absorb successive layers of Chinese vocabulary, including numerals, which later misled scholars into classifying them as Sino-Tibetan.
Sources (1)
Contested classifications
- Minority position
Austro-Tai
Evidence for
Benedict (1942) and Ostapirat (2005) point to regular correspondences in basic vocabulary, where Kra-Dai often matches the final syllable of Austronesian words (ta 'eye' : *mata; taay 'die' : *m-atay). Sagart argues Kra-Dai branched off inside Austronesian, from Taiwan. The Austronesian tree records the same proposal.
Evidence against
The shared words are a limited set, and early borrowing between neighbours on the south China coast could explain them. The structural differences are large, and even supporters disagree on whether Kra-Dai is a sister or a daughter of Austronesian.
Sources (2)
- Largely rejected
Tai as Sino-Tibetan
Evidence for
Tai and Chinese share tone systems, monosyllabic words and a large vocabulary, including numerals. The older Western and Chinese classifications put them together.
Evidence against
The shared vocabulary is borrowed in datable layers, and the tones developed in parallel and later, as they did in Vietnamese. The core vocabulary points elsewhere. Almost all specialists now treat Kra-Dai as a separate family.
Sources (1)