Tania Kuteva

What is the relationship between grammaticalization and long-term evolution of language?

Abstract.

The answer I propose to this question is two-fold. First, I propose that when grammaticalization took place for the first time in conjectural prehistory, its effect was complexification of language: to the lexicon there was a whole new domain, the domain of morphosyntax, added; thus it is justifiable to speak of additive complexity. I call this the Stage of Initial Complexification (Stage-of-IC). Hence, thanks to grammaticalization, the evolution of language at the Stage-of-IC can be viewed as a development along a morphosyntactic complexification trajectory.

Second, in the times after the Stage of Initial Complexification (I call it the Stage-after-IC, or Stage 2) – encompassing that portion of conjectural prehistory which follows Stage 1 as well as the whole of the time of attested history of language – the overall, long-term evolutionary trajectory of morphosyntactic development is very hard to assess, at least given the present state of knowledge. The reason for this is not the fact that at the Stage-after-IC (in that portion of it which constitutes attested history, in particular) in addition to language-internal grammaticalization (i.e. non-contact-related grammaticalization), there is also language-external grammaticalization (i.e. contact-induced grammaticalization), which, as traditionally assumed about contact-induced language change phenomena, should bring about simplification of structure. Contrary to the traditional wisdom, I will argue, both non-contact-related grammaticalization and contact-induced grammaticalization lead to complexification rather than simplification of morphosyntactic structure (cf. Kuteva and Heine forthc. on the “integrative model of grammaticalization”). What makes it hard to determine whether long-term evolution of language involves a constant increase in complexity or not is the existence of common language change processes which may counteract – or at least, may not necessarily enhance – the complexification resulting from grammaticalization, e.g. loss, cyclicity, pidginization/creolization.

I formulate four hypotheses about the effect of common processes of language change – grammaticalization included – on the long-term evolution of language at Stage 2 (i.e. the Stage-after-IC), with respect to morphosyntax: a) the gains equal the losses (i.e. balance); b) the gains are less than the losses (i.e. simplification); c) the gains are more than the losses (i.e. complexification); and d) chronological variation between (b) and (c), depending on socio-historical factors.

My tentative proposal is that Hypotheses (a) and (b) are implausible, whereas Hypotheses (c) and (d) are the ones that deserve special attention in future research on language evolution.