An approach to syntactic reconstruction
The paper is primarily devoted to a methodological discussion. There are two different types of inquiries into diachronic syntax and, more generally, grammar: <i>stage reconstruction</i> and <i>etymological reconstruction</i>. The aim of the first type is to reconstruct and compare diachronic stages within a particular functional domain, while the second type focuses on the etymology or the origin of a particular grammatical category. It is the second type of inquiry that is the topic of this paper. I argue for a methodology based on the Historical-Comparative Method that should ensure a higher degree of reconstructional probability and exclude factors other than inheritance that might also be potentially responsible for correlations across related languages. On this approach, the construction under investigation must be individualized against its respective typological background: creating lists of morphological, lexical (input), syntactic and semantic properties – a procedure that I refer to as <i>profiling</i> (notion borrowed from Cognitive Linguistics). The general principle here is that correlations of typologically quirky properties increase the degree of probability of any reconstruction. An obvious typological quirk is the morphological profile of a construction, since the phonetic realization of morphological markers and their combinations is purely accidental and is not subject to typological universals. The morphological inventory of the construction must be reconstructible in the proto-language on the basis of the Historical-Comparative Method. The ability to reconstruct the morphological inventory also excludes language contact as a potential source for correlations. Other typologically idiosyncratic properties – if reconstructible – may also increase the degree of reconstruction probability. To illustrate how this method may be applied, I focus on the development of the independent partitive genitive from Proto-Indo-European into Baltic and Russian and, finally, into North Russian dialects. On the basis of this method I show that this category is inherited from Proto-Indo-European. I examine the syntactic profiles of this category at different stages and account for changes.