Reconstructions

Note

Also Maranao baroaŋ ‘primitive, ape, bear (animal)’. This cognate set is geographically more restricted than most cognate sets that are attributed to PWMP. Reflexes of *baRuaŋ, *biRuaŋ, however, are widespread in Borneo and are clearly native, as they have sometimes undergone extensive sound changes which have not affected loans. This is particularly noticeable in languages that have strengthened non-phonemic transitional glides, as with Long Glat (Modang) wahaguŋ (where /g/ reflects the intervocalic glide that was phonetically present between the vowels *-ua-, which then metathesized), Kiput (where /f/ reflects the similar glide), Bintulu (where *R disappeared, and from intermediate *buaŋ the same transitional glide was strengthened to /b/ and the preceding *u centralized) and Berawan (Long Terawan) (where intervocalic *R became /k/, the first syllable was lost, and the strengthening of the phonetic glide and centralization of *u paralleled the development in Kiput). Indeed, the cognation of Bintulu bebaŋ and particularly of Berawan (Long Terawan) kebiŋ with e.g. Malay beruaŋ is far from obvious, but is borne out by careful study of the sound correspondences.

Dempwolff's (1934-38) *ba(r)uaŋ can be disambiguated as containing *R. Of the four languages that he compared (Ngaju Dayak, Malay, Toba Batak, and Javanese), the first two have merged all vocalic oppositions in prepenultimate position, Toba Batak baruaŋ differs in meaning and may not be cognate, and Javanese beruaŋ, like Maranao baroaŋ and Buginese baruaŋ, appears to be a Malay loanword. His unambiguous reconstruction of first-syllable *a was thus apparently an illusion. Both Banjarese and Minangkabau preserve the contrast of *a and *i in prepenultimate position (Adelaar 1992), and these witnesses agree in indicating *i. The reflexes in Murut (Paluan), Kadazan Dusun, and Lun Dayeh, on the other hand, indicate *a.

Finally, the first appearance of this word must date from at least the entry of the ancestral population of present-day Austronesian speakers into Borneo (hence to approximately the period 2,000-1,500 B.C.). It may, however, be considerably older. The bear, generally designated by a reflex of *Cumay, is also found in Taiwan, but is absent in the intervening Philippines. In Southeast Asia it is found only on Borneo, the Malay Peninsula, and Sumatra. Since *Cumay and *baRuaŋ, *biRuaŋ are not cognate, no certain inference can be made about a PAn word for ‘bear’. The distribution of word and referent in this case provides an instructive contrast with reflexes of *qaRem ‘pangolin’ (q.v.), where the animal has the same discontinuous distribution (Taiwan and Borneo), but the native names are cognate, thus requiring a PAn reconstruction.