Rethinking Sino-Tibetan phylogeny from the perspective of North East Indian languages

Abstract

Sino-Tibetan has more speakers than any other language phylum, and covers a major proportion of the land area of East Asia. Despite some two centuries of study and publication, the subclassification of Sino-Tibetan remains highly controversial, as does its external affiliation (van Driem 2008a; Blench 2008a,b; Handel 2008). Originating as “Indo-Chinese” in the middle of the nineteenth century, it originally carried racial connotations (van Driem 2002). The first recognition of the phylum probably dates to Julius von Klaproth (1823) who recognised three parallel branches: Chinese, Burmese and Tibetan. Von Klaproth explicitly excluded Austroasiatic and Daic, unlike many later classifications, which sequentially included all the regional phyla. Although such views still sometimes surface (primarily in Chinese publications), they have been fairly conclusively rejected by most scholars. Considering the importance of Sino-Tibetan and its history of scholarship, there is a striking lack of consensus as to its internal classification. Historically speaking, there have been two opposing camps: those who consider Sinitic (i.e. the several varieties of Chinese) as representing a primary branch of the family (Wolfenden 1927; Benedict 1972, 1976; Bodman 1980; Weidert 1987; Bradley 1997b, 2002; Matisoff 2003, 2008; Thurgood and LaPolla 2003; Handel 2008) and those who situate it within the remaining languages, consequently applying the name Tibeto-Burman to the whole phylum (Shafer 1955, 1966/67; van Driem 2002). In recent years, successful reconstructions of low-level groups have begun to appear (e.g. Sun 1993; Mortensen 2003; VanBik 2007; Wood 2008; Button 2009), raising hope that higher-level reconstructions may eventually be able to be placed on a stronger footing or at least, that their validity will be able to be more rigorously tested. Nonetheless, many putatively Sino-Tibetan languages remain very sparsely documented, with accessible comparative lexical material of any significant scale being largely confined to Chinese and (to a lesser extent) Indian sources. The largest-scale comparative database of Sino-Tibetan languages compiled to date, the Sino-Tibetan Etymological Dictionary and Thesaurus (STEDT) project, was finally made generally available in “beta” form in October 2010, with some additional functionality added in 2012 (though still in “beta”). Consequently, historical linguists can now see the evidence for Proto-Tibeto-Burman forms given in Matisoff (2003), which did not explicitly present the data on which most reconstructions are based.1 Some potential problems with the linguistic data employed in Matisoff (2003) have been pointed out in various reviews (e.g. Sagart 2006; response by Matisoff 2007; further reply Sagart 2008; Hill 2009). One point that we will underscore here, however, is that there are also problematic disconnects with the archaeological evidence (Blench 2008b). For example, while ‘iron’ is reconstructed at the Proto-Tibeto-