Transitivity and Change of State Verbs

Saundra Wright

Department of English, CSU-Chico

Abstract

One of the trademarks of externally caused change of state verbs (e.g., break, freeze) is their ability to participate in the causative/inchoative alternation:

            (1)       a.         The window broke.
                     b.         John broke the window

This behavior has been contrasted with that of internally caused change of state verbs (e.g., bloom, corrode).  As Levin & Rappaport Hovav (1995) argue, these verbs occur in the inchoative variant, but fail to occur in related causative forms:

            (2)       a.          The roses bloomed.

                     b.                         *The gardener bloomed the roses.

However, following work by McKoon & Macfarland (2000) and Wright (2001), I argue that this claim is not entirely accurate: some internally caused change of state verbs are found in transitive causative constructions:

            (3)       a.         The acid corroded the metal.

                    b.         Early summer heat wilted the petunias.

Nevertheless, my survey data indicates that there is a wide range of acceptability judgments associated with transitive uses of internally caused change of state verbs; while some constructions are judged as being acceptable (3); others are uniformly judged as being unacceptable (2b).

            In my analysis, I argue that semantic and pragmatic properties—as opposed to syntactic properties—are responsible for the range of acceptability ratings found across transitive uses of change of state verbs.  I claim that a variety of factors combine to determine the argument expression options associated with a particular verb.  These factors include controllability (the degree to which an event can be externally manipulated)¹, causer-type (whether it involves a human-driven or nonhuman-driven event)², and subject-modification (whether the causer is in a modified or nonmodified form)³.  Moreover, I argue that it is possible to predict how frequently a verb occurs transitively, as well how acceptable it sounds in a transitive construction, by a statistical model that combines and weights these different factors.  In my rating tasks, subjects judged the acceptability of change of state verbs in a variety of transitive sentences.  I demonstrate that these actual acceptability ratings can be predicted by a formula that combines the different semantic and pragmatic factors discussed above.  Consequently, this analysis accounts for the syntactic similarities found across all change of state verbs, while accounting for the range of acceptability ratings found in their transitive uses.   

¹The wind eroded the riverbank. / ??The heat bloomed the roses.

²Acid corroded the metal. / ??The scientist corroded the metal.

³Intense sunlight wilted the roses. / ??Sunlight wilted the roses.

References

Levin, B., and M. Rappaport Hovav (1995).  Unaccusativity at the Syntax-Lexical

Semantics Interface.  Cambridge MA: MIT Press.

McKoon, G. and T. Macfarland (2000).  Externally and Internally Caused Change of

State Verbs.  Language, 76, 833-858.

Wright, S. (2001).  Internally Caused and Externally Caused Change of State Verbs. 

PhD Dissertation, Northwestern University, Evanston, IL.