Refine
Document Type
- Working Paper (2)
Language
- English (2)
Has Fulltext
- yes (2)
Is part of the Bibliography
- no (2)
Publisher
Psych-verbs, exhibiting a wide range of construction types, challenge the assumption that verbs sharing the same theta-grid and event structure select the same case patterns. Consequently, previous research has tried to trace back the constructional variants of these verbs to differences in deep syntactic or semantic (event- or causal) structure. However, in German a construction type may exhibit semantically distinct subclasses, and a semantic subclass may have different syntactic realizations. Thus, it is difficult to find general linking patterns of psych-verbs. On the other hand, it is characteristic of psych-verbs to have non-psych-readings as well. We argue that these readings determine case selection. In the centre of our argumentation is the principle of Lexical Economy according to which entries in the mental lexicon are as simple as possible. The case frame must be compatible with all readings of the verb. Thus, the reading with the strongest thematic distinctiveness of each argument is crucial for case selection, typically a reading with a volitional agent and a strongly affected patient. Such readings impose stronger restrictions on case selection than psych-readings. The linking theory of our approach is based on Dowty’s theory of proto-roles (1991) and its modification and implementation in Optimality Theory in Primus (2002).
Psych-verbs have been a touch-stone to linking-theories, which assume that case selection is determined by thematic roles. Though psych-verbs share the same thematic grid (experiencer and stimulus), they show different case frames. Different syntactic (structural) and semantic (event- or causal structure) approaches exist, but at least in German we will show that none of the several approaches to psych-verbs cover all differences and similarities of the several formal classes of psych-verbs in that language. In this paper we argue that the case selection of psych-verbs does not depend on their psychical reading at all. While Functional Expressivity requires that different thematic roles are expressed by different forms, Lexical Economy states that lexical entries should be minimally, i.e. verbs should only provide one case frame. Thus, the case frame of a verb must be compatible to the thematic requirements of all readings of this verb. Researchers paid little attention to the fact that polysemy is characteristic for psych-verbs. Psychverbs have (or have had) other, more specific readings, as well, and occasional psychical readings are possible for most verbs. According to the proto-role approach of Dowty (1991) and its modifications by Primus (1999b, 2002a, 2002b, 2002c), case selection is determined by the grade of agentivity or patientivity of arguments. Concrete readings have stronger agents and patients and make therefore stronger restrictions to case selection, and the psychical reading of a verb is always compatible with this reading. Thus, the case selection of psych-verbs is not affected by its psychical reading.