Deadjectival verbs: event-change path and (not always) result

MANGIALAVORI RASIA, MARÍA EUGENIA

Granada

Conferencia; 2nd International Conference on Meaning and Knowledge Representation; 2013

Universidad de Granada / Gobierno de España

We will depart from the idea that the adjectival roots in change-of-state deadjectival verbs are ‘paths of change’. Following the general assumption that PPs are the simplest representations of paths (in the spatial domain, canonically); and that they also deliver extended metaphorical (non-spatial) uses involved by our human perception of the notion of change according to which ‘change-of-state’ events can be seen as analogues of motion events (e.g., Levin&Hovav 2005); we will entertain the hypothesis that Adjectival Rhemes would represent (abstract) spatial paths which are mapped on to the dynamic (change-of-state) event (Zwarts 2003), providing a gradable (property) scale which functions as the mapping to the PATH transversed by the undergoer of the event of change. Further, the PATH structure of the adjective would be mapped onto the temporal PATH structure of the time line of the event. Since PATH can be either bounded or unbounded, we will see that the difference between closed-scale or open-scale adjectives correlates with resultativity in certain verbs and to gradable change (though by this not implying the attainment of a resulting state) in others, thus allowing the discrimination of two natural classes, which are particularly salient in Spanish. Deadjectival verbs would arise from rhematic material being incorporated/conflated from the adjective [A0] in complement position into the verbal head (Hale&Keyser 1993, Ramchand 2007, i.a.). Considering that one of the essential properties for being selected complement of a BECOME predicate (a process projection) is for the root to certain scalar structure that can be mapped to the verbal change in a systematic way (following Ramchand 2007); under the system proposed here, in their composition, the complement position of the verbal head would be filled by RHEMES (either RHEMES of process or RHEMES of result) embedded by the A0; the main difference being that RHEMES (having PATHS as subcase), do not describe participants (arguments) in the eventive structure, but actually denote a scalar property that can be measured. Measure would reflect the extent to which entities have the property in question. Consequently, only closed-scale adjectives give rise to a telic/resultative event: by homomorphism, the endpoint of event is identified with the final stage of the PATH: the attainment of the property (the rhematic material) embedded by the adjectival root (which is interpreted as rheme of result). On the other hand, while there is still a similar relation between the process and the undergoer of the change, a crucial difference will lie in whether the root is construed either a resulting/final state or as definitional of the process itself, since in open-scale cases the property denoted by the A0 is not necessarily attained by the undergoer, rather it describes the kind of change underwent. As a result, the main contrasts could be accounted for based on the kind of PATH the A0 will be able to build; and the process (change) denoted by the verb would be established via the scalar structure of that property. Further internal differences among these verbs would be part of the lexical encyclopedic properties of the root, and would not be directly encoded in the syntax.