Quantifiers, "Again" and the Complexity of Verb Phrases

Many linguists have used the modifier again as a diagnostic for syntactic complexity, whether in terms of lexical decomposition (McCawley 1976) or hidden functional categories (von Stechow 1996). Evidence for this complexity relied on establishing again as a propositional modifier. Sentence internal application of this modifier substantiated the existence of embedded propositions. In this paper I follow this tradition of identifying embedded propositions, although I concentrate on the structural characteristics of verb phrases (VPs) rather than decomposition or hidden functional categories. I propose that VPs can be divided into two separate categories, those that contain more than one proposition and those that contain only one. Furthermore, this division is based on an aspectual di stinction. VPs with non-stative, transitive verbs are propositionally complex whereas VPs with intransitive or stative, transitive verbs are propositionally simple. To explain this partition, I propose that verbs combine with their arguments differently depending on their aspectual category. Non-stative, transitive verbs are functions that only take their objects as arguments (cf. Kratzer 1996). The subject is related to the verb through an agentive predicate. In contrast, other verbs take all their syntactic arguments as functional arguments. I present thi s thesis in section 3 and then argue for it in section 4. The argument is structured as follows. First I establish that again's syntactic complement strictly determines the content of the presupposition introduced by again. Then, I demonstrate that VPs with non-stative, transitive verbs allow again to introduce a presupposition that does not involve the subject (a subjectless presupposition). This suggests that the verb and object form a proposition independent of the subject. In contrast, VPs with intransitive verbs or non-stative, transitive verbs do not allow subjectless presuppositions. Hence, the verb and object do not form a proposition. As discussed in section 4, the different functional characteristics of verbs explain these facts . In section 5, I explore some consequences of the theses in section 3. As discussed below, the complexity of non-stative, transitive VPs predicts that Quantifier Phrases (QPs) should be interpretable within the VP. The systematic interaction between again and QPs al lows one to evaluate this claim. Surprisingly, evidence shows that QPs cannot be so interpreted.