Atelier Doculang Linguistique de terrain et documentation
Stavros Skopeteas (Université Göttingen)
Understanding variation is a particular challenge with spoken languages, not only because of the lack of established written standards, but also because they are frequently used in the context of increasing multilingualism which accelerates processes of diachronic change.
This presentation deals with Caucasian Urum, an Anatolian dialect of Turkish, spoken in the Small Caucasus (Georgia). The syntactic structure of the variety spoken by the old generation is very similar to the Anatolian varieties of Turkish, although the population did not have substantial exchange with speakers of Turkish in the last two centuries. However, young speakers who migrated to the urban centres and are socialized in a multilingual environment deviate to a large extent from the language of their ancestors. Russian has a noticeable impact on their use of Urum, both at the lexical level (frequency of borrowings) as well as at the structural level (in several areas of grammar, e.g., the use of complementizers, the directionality of syntactic projections, the optionality of number morphemes, etc.).
The methodological challenge is to create field methods that capture this variation and its consequences for our understanding of the ongoing change. In this presentation, I will concentrate on the change from OV to VO that is reported for several Turkic varieties spoken in Germanic or Slavic/Baltic context, e.g., Macedonian Turkish, Karaim, Gagauz or the Turkish spoken by emigrants in the Netherlands.
An investigation of a rich spoken corpus of Caucasian Urum reveals that the frequency of VO linearizations increases in the speech production of young Urum speakers, which indicates that these speakers use VO sequences in a wider array of contexts. Starting with this observation from corpus, I address the question whether these frequencies correlate with a general change in the directionality of V-projections. An acceptability study was carried out in the field with a sample of young speakers, a sample of old speakers, and two control groups (Turkish, Russian) in order to answer this question. The result shows that young speakers (a) have different linearization preferences for verb clusters and (b) create new patterns of adjunction (adverb order). These facts are only partially explained by the influence of Russian. Young speakers create a new type of variety that may have been triggered by the exposition to Russian, but it is not an exact copy of the contact language.