On Korean adjectival verb clause which cannot be explained by the verb’s own properties.

Abstract

This paper shows that, in Korean, there are adjectival verb clauses with the structure and the meaning which is not explainable in terms of the semantics of the adjectival verb, and argues that they could be explainable if and only if each of them is seen as a construction, that is, a symbol which is defined as a directly associated pair of a form and a meaning, in this case, the form is a chunk of several forms. This article demonstrates it with several constructions: distributional constructions ([NP{place}-ey NP{theme}-i Adj.V{being distributed}]) with a color adjectival verb at the slot Adj.V{being distributed}; relational constructions ([NP{theme}-i NP{relatee}-hako Adj.V{being related}]) with an evaluation adjectival verb at the slot Adj.V{being related}; and directional action constructions ([NP{agent}-ka NP{goal /recipient}-hanthey V{directional action}]) with a attitude adjectival verb at the slot V{directional action}. Such quirky combinations of the constructions and the adjectival verbs mentioned above emerges when the event projected by the construction at question and that by the adjectival verb at question are close. That is to say, metonymy which enables mentioning a part to bring out the whole plays an important role.