Mittwoch, 15.00 Uhr
I will argue that, contrary to the common assumption, the complements vs. adjuncts dichotomy based on such notions as obligatoriness, predictability, filling a semantic role, and iterability, plays little, if any, role in the syntax. To this end, I will critically examine various tests assumed to provide evidence for configurational treatment of the dichotomy and argue that they actually don't.
Two rather different phenomena I'll go into in some detail are case assignment and the ‘do so’ substitution test. On the former point, I will recall facts from Finnish, Korean and Russian, as well as adduce some new facts from Polish, showing that case assignment is — contrary to the dominant assumptions — largely indifferent to the complement-adjunct distinction and that apparent case assignment differences can be reduced to semantic features, e.g., referentiality in Polish. On the latter point, I'll argue against the popular view that ‘do so’ is a V'-proform (Lakoff and Ross 1966/1976) or, more generally, that it is a surface anaphor (Hankamer and Sag 1976). Instead, I'll show that ‘do so’ shares many characteristics with deep (pragmatically controlled) anaphora and thus cannot be used for determining VP constituency.
I'll conclude by showing how these results are compatible with non-configurational treatments of the complement-adjuncts dichotomy as, e.g., in (a line of inquiry in) Head-driven Phrase Structure Grammar or in Jackendoff's Conceptual Semantics.
zum Programm der AG 8
zur alphabetischen Übersicht der Abstracts
zur zeitlichen Übersicht