A corpus based analysis of English, Swedish, Polish, and Russian prepositions

Abstract

In this study, the use of most frequent spatial prepositions in English, Polish, Swedish, and Russian is analyzed. The prepositions and their contexts are extracted from corpora by means of concordance tools. The collostructional strength between the prepositions and the most frequent nouns in the PPs (Gries et al. 2005) is then computed in order to get a more detailed picture of the contexts in which a given preposition is most likely to appear. The results of the investigation are then analysed within the framework of cognitive semantics, especially Croft and Cruse's (2004) taxonomy of construal operations, and Talmy’s (2005) classification of spatial images. Background and aim Prepositions define relations between objects, or, rather, conceptualizations of objects. In order to define their meaning it is necessary not only to describe the relation the preposition expresses but also the objects involved. Still, a description based on geometrical notions (the dimensionality of the objects) does not cover all aspects of the semantics of prepositions. Croft and Cruse (2005) propose a model that enriches the geometrical descriptions with construals as focus, scale of attention, perspective and viewpoint. The main difficulty in cross-linguistic description of preposition semantics is due to the fact that cross-language differences in prepositional systems increase as we move from physical senses of prepositions into the metaphoric extensions of their meaning. The meaning chains (Brugman 1981, Gawronska 1993, Taylor 1988) have different shapes in different languages. Another difficulty lies in different degrees of lexicalization and in formulating criterions for regarding prepositions as parts of lexicalized multiword entries. These problems are of central importance for language technology application, especially Machine Translation (MT). In the present study, we investigate whether the collostructural analysis, proposed by Gries at al. (2005), may be helpful in identification of nouns and noun classes that tend to co-occur with certain prepositions in 4 different languages, and whether the values of the collostructural strength may contribute to a better understanding of similarities and differences among prepositional systems. Another question we are concerned with is whether and how the results may improve the treatment of prepositions in MT.