Overcounting numeral systems and their relevance to sub-grouping in the Tibeto-Burman languages of Nagaland

This paper presents an appraisal of the most influential genetic classifications that have been proposed for the Tibeto-Burman languages of Nagaland, north-east India, and weighs up the evidence for the validity of a 'Naga' branch within Tibeto-Burman. The survey concludes that while phonological and lexical correspondences might be generally useful for establishing the affiliation of these languages to the Tibeto-Burman family, such criteria shed limited light on the problem of establishing lower level sub-groupings. Apart from the Konyak languages, which have been convincingly grouped with Jinghpaw and the Bodo-Garo languages on the basis of lexical innovations, sub-grouping within the languages of Nagaland remains inconclusive for want of robust diagnostic criteria. The paper presents new evidence for an intermediate grouping of languages of the central and southern regions of Nagaland (i.e. the Ao and Angami-Pochuri clusters), the historicalcomparative basis for this being typologically rare overcounting numeral systems documented in the languages of these two regions in the late 19 th and early 20 th centuries. The complexity of overcounting numeral systems lessens the likelihood of their being borrowed, and such patterns have never been reported in contiguous Konyak, Karbi, Zeme, Kuki or Tangkhul languages. The paper concludes that the erstwhile presence of overcounting patterns in Tibeto-Burman languages of central and southern Nagaland must therefore represent an inherited innovation indicative of a close genetic relationship.Published versio