Suomalais-Ugrilaisen Seuran Aikakauskirja 82 [antikvár]

Knut Bergsland COMPARATIVE ASPECTS OF ALEUT SYNTAX This paper is a continuation of the paper published two years ago in this journal (1986: 1-75 = 63-137): "Comparative Eskimo-Aleut phonology and lexicon", and aims at shedding more light on the relationship of Aleut and the Eskimo lan-guages.without pretending to be a full treatment of the subject.* As mentioned in the previous paper, a crucial point of difference between Aleut and Eskimo is the construction of transitive verbs. Eskimo has a well-known erga-tive system: on the one hand a two-place verbal predicate in number agreement with a nominal subject (agent) in the relative case and a nominal object in the absolutive case, or a person suffix for the one or both; on the other hand a one-place verbal predicate in number agreement with a nominal subject in the absolutive case and an optional oblique object in the instrumental case, also called modalis, in Yup'ik an ablative-modalis. Between the two constructions there is mostly a difference of definiteness, e.g. Central Alaskan Yup'ik (CAY)i Angute-m ner-a-a neqa. man-RELSg eat-moic-sgSsg fish-ABSsg (zero suffix) 'The man eats the fish.' Angun ner'-uq neq-mek. man-ABSsg eat-iNDic-sg (zero) fish-MODSg 'The man eats a fish (or some food)'. Without the specified objects: Angutem neraa. 'The man eats it.' vs. Angun ner'uq. 'The man eats, is eating'. In Aleut there is no instrumental or modalis case, and a specified object, in the absolutive case, entails a one-place predicate in number agreement with a nominal subject in the absolutive case, whereas a two-place predicate, with a nominal subject in the relative case, is anaphoric: Tayagu-x qa-x qa-ku-x. man-sg fish-sg eat-PRES-sg 'The man eats the/a fish.' Tayagu-m qa-kuu (-ku-a). man-REL eat-PREs-sg3[sg] 'The man eats (or ate) it.'