Distribution of Valency Complements in Czech Complex Predicates: Between Verb and Noun

Distribution of Valency Complements in Czech Complex Predicates: Between Verb and Noun Václava Kettnerová and Eduard Bejček Charles University in Prague, Faculty of Mathematics and Physics Institute of Formal and Applied Linguistics Malostranské náměstí 25, 118 00, Prague 1, Czech Republic {kettnerova,bejcek}@ufal.mff.cuni.cz Abstract In this paper, we focus on Czech complex predicates formed by a light verb and a predicative noun expressed as the direct object. Although Czech – as an inflectional language encoding syntactic relations via morphological cases – provides an excellent opportunity to study the distribution of valency complements in the syntactic structure with complex predicates, this distribution has not been described so far. On the basis of a manual analysis of the richly annotated data from the Prague Dependency Treebank, we thus formulate principles governing this distribution. In an automatic experiment, we verify these principles on well-formed syntactic structures from the Prague Dependency Treebank and the Prague Czech-English Dependency Treebank with very satisfactory results: the distribution of 97% of valency complements in the surface structure is governed by the proposed principles. These results corroborate that the surface structure formation of complex predicates is a regular process. Keywords: Czech complex predicates, distribution of valency complements, formal rules 1. Introduction Multiword expressions (MWEs) have proven to be a serious challenge for NLP (Sag et al., 2002). From various types of MWEs, those that involve verbs are of great significance as the verb is the syntactic center of the sentence. In this paper, we focus on one type of Czech verbal MWEs – on complex predicates with light verbs (CPs). CPs consist of two syntactic elements – the light verb and a predicative noun, adjective, adverb or verb – which function together as a single predicative unit. With respect to the wide range of issues, we limit our study to a central type of Czech CPs – to those CPs in which the predicative noun is expressed as the direct object (e.g., vést jednání ‘to hold talks’, mít potíže ‘to have difficulties’, udělat chybu ‘to make a mis- take’, dostat příkaz ‘to get an order’). We aim to formulate principles governing the surface structure of these CPs and to verify these principles on corpus data. CPs exhibit a discrepancy between syntax and semantics. Their syntactic center is the light verb, which requires the predicative noun as one of its valency complements. Their semantic core is, however, the predicative noun, which se- lects the light verb (Algeo, 1995). Both the light verb and the predicative noun typically contribute their valency com- plements to the syntactic structure of CPs. See examples (1) and (2) with the verb dostat ‘to get’. Unlike the syntactic structure with the predicative verb dostat ‘to get’ in (1), the structure with the CP dostat příkaz in (2) is constructed of both valency complements of the light verb dostat ‘to get’ and complements of the predicative noun příkaz ‘order’, which adds the valency complement střílet ‘to shoot’ to the resulting structure. (1) Vojáci dostali od velitele zbraně. soldiers …

Content truncated to 500 words