조동사 축약과 to - 축약 = Auxiliary Reduction and To - Contraction

다국어 초록 (Multilingual Abstract)

Schachter(1984) and Brass(1995) tried to explain auxiliary reduction and to-contraction uniformly. Schachter, for example, made advantage of different sentence boundaries in the framework of GPSG. If a clause has a full subject, contraction is impossible because it has a sentence boundary S which blocks government from the main verb. On the other hand, if the subject is missing in a clause, it has a boundary VP, not S, then the main verb can govern the head of a subordinate clause and the contraction is allowed. Structural adjacency as well as string adjacency is required for auxiliary reduction and to-contraction. The government between the host noun or verb and the target auxiliary or to is necessary for structural adjacency. The structural adjacency condition is, however, obligatory only for to-contraction. Auxiliary reduction shows that it is not directly concerned with the structural adjacency condition. Thus, the unified theory for auxiliary reduction and to-contraction is not helpful. It is suggested in this article that to-contraction has to be treated syntactically and auxiliary reduction phonologically, as Klavans(1980) offers a theory of cliticization in which a word may be syntactically proclitic but phonologically enclitic.