It is shown that postulating a Focus Phrase above vP enables us to explain such diverse phenomena as the Malayalam question word's position contiguous to V, the`remnant' in English pseudogapping, the clause-final floated' focus marker in English, and the position of the`cleft focus' in English and Malayalam clefts. Assuming a Kaynean view of the underlying structure of SOV languages, we argue that the`canonical' positions to which the verb's internal arguments are moved in these languages are above this Focus Phrase. Postulating an iterable Topic Phrase above the Focus Phrase (and above thè canonical' positions in SOV languages) enables us to account for the definiteness/specificity constraints on clause-internal scrambling in Malayalam, German and Dutch, and on object shift in Scandinavian. Finally, it is shown that all the functions attributed to an`outer' Spec position of vP are better fulfilled by the Topic/Focus positions above vP that we postulated.

The Malayalam conjunctive suffix -um and disjunctive suffix -oo, when suffixed to (a phrase containing) a ''question word,'' yield (respectively) a universal quantifier and an existential quantifier. A ''question word'' (I assume) signifies a variable (Nishigauchi 1990); and a conjunction/disjunction operator applied to a variable interprets it as an ''infinite conjunction/disjunction'' (the meaning of a universal/existential quantifier). The operator ''applies to'' a question word by ''association with focus'' (Rooth 1985). Malayalam has the disjunctive -oo at the end of a question. Universally (I claim), questions contain a disjunction operator generated as the head of ForceP (of the ''more finely articulated C'' of Rizzi 1997). From this position it applies to question words by association with focus, yielding question interpretations that (I show) capture the semanticists' intuition that wh-phrases are existential quantifiers. Association with focus yields a satisfactory account of wh-insitu, and I show that it must apply even to wh-in-C.

In the Dravidian languages reduplication of numerals and pronouns gives rise to distributive meanings. For example, in Telugu, as in Kannada, Tamil, and Malayalam, the other major Dravidian languages, reduplication of numerals gives rise to distributive readings. We find that the reduplicated numeral construction always gives rise to distributive readings. There are no collective readings of a reduplicated numeral construction. Additionally we observe that it gives rise to two further distributive interpretations that are not present in the non‐reduplicated construction. The main focus of this chapter is to explain the cause of the obligatory distributivity that is associated with the reduplicated numeral construction and to account for the various additional distributive readings that are possible in such constructions. This work is embedded in a larger project which examines the cross‐linguistic morpho‐syntax and semantics of reduplication, not restricted to numerals only, and its relation to plurality.

In this paper we point out that in a wide variety of languages, reflexive anaphors seem sensitive to Principle B when they are morphologically simple. While this is now acknowledged by many linguists, we show that (further), when reflexive anaphors in these languages are morphologically complex, they still contain a pronominal element which obeys Principle B. We also provide evidence that many complex reflexive forms which are currently taken to be local (or ‘strict’) anaphors, are on closer examination seen to be only non‐local: i.e. they can take both local and long‐distance antecedents. We suggest a syntactic process of “reflexivization’ which enables anti‐local (pronominal) elements to take an antecedent in the minimal clause. We lastly show that truly local anaphors – like reciprocals and distributives – also contain a pronominal element which obeys Principle B.

No abstract