Degrees of semantic control in measuring aggregated lexical distances

The goal of the current study is to show how aggregated lexical variation can be studied by means of corpus-based techniques, which differ in their amount of semantic control. In the current variationist field, one finds many studies of phonological or morphological variation on the basis of corpora1. Remarkably at first sight, though, studies of lexical variation in corpora are rare, especially in comparison with dialectology, where the study of lexical variation is part of the main research goal. In contrast to the other corpus-based variationist studies, however, the dialectological account of lexical variation is very much restricted to elicited data, as stored in well-known dialect atlases. Therefore, the current study sets out to show how this void of corpus-based studies of lexical variation can be filled, while taking into account possible issues with lexical semantic complexity. In the introduction to the paper, we would like to point out two things. First, we will explain why there is a plethora of studies on phonological and morphological variation and a scarcity of studies on lexical variation. Second, we will shed a different light on what can be understood under lexical variation from a corpus-linguistic point of view.