David Perlmutter

David Perlmutter of UCSD will speak at the UCSD Linguistics Department Colloquium on November 28, 2011, at 2:00 pm in AP&M 4301.

"Subject Position" and Clause Types in the Germanic Languages

The ten modern Germanic languages other than English all have two basic clause types: V2 clauses, where the finite verb is said to be required to be in 2nd position, and V1 clauses, where it is said to be required to be in 1st position. V2 clauses exhibit a Subject Alternation: the subject sometimes precedes and sometimes follows the verb. A central goal of this paper is to explain the Subject Alternation: why it exists and why it occurs in the environment it does. A second goal is to explain the  ways V2 and V1 clauses differ and the ways they are alike. Third, we construct a grammar fragment that accounts for the linear position of left-peripheral (LP) constituents (subjects, WH-constituents, topics, the finite verb, and subordinate clauses).

Using data from German, we begin by showing that in so-called “V2” clauses the verb is not always in 2nd position as commonly assumed. We propose the Split V2 Hypothesis, under which two distinct constraints are responsible for the verb’s position. We show that, using Optimality Theory to rank these two constraints with alignment constraints that position other LP constituents accounts for the linear order of all these constituents and the verb. We then show that these ranked constraints, posited to account for sentences where the subject is postverbal, automatically predict the Subject Alternation and the environment in which the subject is preverbal.

Next we show that the Split V2 Hypothesis makes it possible to predict both the contrasts and the commonalities between V2 and V1 clauses by positing just one difference between them.

Having shown that the set of so-called “V2” clauses includes some V3 clauses, and that the set of so-called “V1” clauses includes some V2 and V3 clauses, we propose a novel typology of clause types in Germanic based on the constraints that derive them rather than the V2-V1 surface typology. We end by showing that this new typology makes a correct and novel prediction about what constitutes “narrative style” in Yiddish.

The theoretical framework of this paper is Optimality Theory unencumbered by old assumptions. Throughout, we ask how much information is needed to account for the linear order of LP constituents, to solve the Subject Alternation problem, and to discover the grammatical basis of the contrast between so-called  “V2” and “V1” clauses in the Germanic languages. We conclude that all these problems can be solved without recourse to the kinds of devices that predominate in the literature: a highly articulated constituent structure, surface-false underlying and intermediate structures, movement of constituents from one node to another, and the like.