Click here to submit your abstract to the 2024 conference now! Submissions close on 21 February, 23:59 GMT.

On Tocharian exceptionality to the centum-satem isogloss

The centum/satem division of the Proto-Indo-European family is a near perfect isogloss. However, Tocharian and Hittite are two centum languages located within the satem language area. Two theories are normally used to justify this anomaly: either there was a late split after Tocharian and Hittite, or a growing dialect wave did not reach either of them due to geographical distance. In this paper, I argue that there may have been pressures inhibiting change in Tocharian, which could explain that anomaly in a novel way.
Firstly, I take a holistic view of language complexity, where each level only changes in relation to the language as a whole, and where the overall complexity of a language may change over time. (Sampson 2009) In this case, more radical syntactic, morphological, and lexical changes may have inhibited a change in the phonemic inventory of the language. I support this by showing examples from a similar language area in Kupwar, India, among others. Secondly, I examine the historical evidence for other languages in the region in order to show that horizontal influence from non-IE languages may have also influenced the proposed merge into [k]. Thirdly I posit that the merge of the PIE dorsal consonants may have been due to a decline in the general use, domain and range of Tocharian, which would be in line with recent research correlating small phonemic inventories with small language communities. (Hay and Bauer 2007)
The arguments above can be used to show that Tocharian may not have been susceptible to a weak dialect wave, or to a possible early branching of IE. This paper therefore sheds new light on the history and development of Tocharian, and in turn on the nature of the ‘textbook’ isogloss example.