Files
Abstract
This dissertation aims at providing a uniform analysis to account for both passive constructions (PCs) and so-called tough constructions (TCs, exemplified in (1)), focusing on the apparent possibility for the internal argument of an embedded verb to cross numerous available A-positions as well as CP boundaries (in TCs). (1) a. The article was difficult to review. b. The article was difficult to convince my editor to read. c. The article was difficult to ask my colleague to write a review about. I assume that TCs are a result of an A-movement operation, and that no (A-bar) null operator (Chomsky 1977) is at play. I show that TCs allow for long-distance movement (cf. examples 1b and 1c), and PCs do not. I argue that they are, nevertheless, the same considering their derivation: For both constructions, I posit a derivation based on suppressed case features, as well as feature inheritance allowing the internal argument of (a deeply) embedded verb to appear as the grammatical subject of the sentence. Under my approach the embedded verb enters the base configuration with suppressed case features that become available after a specific syntactic environment has been constructed around it. To account for both PCs and TCs, feature suppression is crucial. Consider example set (2) demonstrating the different argument possibilities for the verb to read in a non-TC, non-PC sentence. (2)a. She is reading. (2)b. She was reading a book. (2)c. She was reading her sister a book. The verb to read has the same meaning in all three sentences (2a) through (2c), but appears with a different number of arguments, and thus with a different number of [ACC] features in each sentence. The [CASE] availability, however, is not only dependent on the merge of internal arguments at base configuration, but rather, on the syntactic environment (cf. Chomsky 2008). Extending this approach to TCs and PCs, I provide a unified explanation of these constructions.