Summary: | Khanty is a Uralic language spoken in Western Siberia. Surgut Khanty with 2000 speakers is one of the variants of Eastern Khanty. Most speakers are bilinguals and Surgut Khanty is nowadays an endangered language. The main aim of this dissertation is to find the functions of rich morphosyntactic devices in Surgut Khanty. I limited the analysis on comparisons variation between alignments. The main theoretical frame is founded on discourse-based functionalism. Mainly the model of the method derives from Preferred Argument Structure (Du Bois 1987) and in this framework, morphology (noun phrase types of referents as lexical NPs, pronouns or zero anaphora), semantics (animacy and person) and pragmatics (information status as new or given information, referentiality as referent tracking, topicality) will be studied. Referent tracking means here coding the discourse referentiality of noun phrases in discourse. I mapped the distribution of each category and configured them with a pragmatic frame. The data used in the dissertation is originally spoken narrative text from after 1980 s, including the data I collected from interviews with native Surgut Khanty speakers. The data consists of 295 minutes 20 seconds of audio recorded personal narratives. The dissertation contains an abundance of grammatical sketch which also contains previously undescribed grammatical features. I have analysed five alignments. The remarkable findings are the followings. 1) The dative shift alternation can also trigger subject conjugation regardless of the typological tendency and previous study on Khanty. 2) In object variations, I have compared nominative/accusative object and oblique object which has not been mentioned in previous studies of Khanty. The analyses demonstrate that the oblique object can be regarded an object since it is a semantically obligatory argument, but it also functions as an oblique in referent tracking in discourse. 3) The object conjugation does not pertain to the first and the second person pronouns as object in ...
|