A Quantitative Study of German Word Order Variation
Kris Heylen · Lirias (KU Leuven) · 2005
Word order variation in German is as an area of syntactic research in which the limitations of the types of data traditionally used in theoretical linguistics have become apparent. In a case study, we present a quantitative corpus analysis as a possible alternative to overcome these shortcomings. Traditionally, theoretical linguists have not had to worry much about the validity of the data on which they based their theories. For most linguists, obtaining relevant data seemed fairly unproblematic and easy. In the generative tradition, linguists could rely on the introspective grammaticality judgments of any single native speaker (including themselves) to uncover the principles of grammar. Researchers taking a cognitive-functional approach offered hermeneutic interpretations of “encountered” examples of language use to show how discourse properties or general cognitive abilities shape the grammar. Most research into word order variation in German was also based on these types of linguistic evidence. Recently however, there has been a growing awareness across the field of theoretical linguistics that this kind of data has limited reliability and is insufficient to deal with the complexity of linguistic phenomena. Several options are being pursued to provide grammar research with a more solid empirical basis. One of them is the use of large electronic corpora for collecting representative data samples in which theoretically interesting patterns of linguistic structure can be discovered and validated. This kind of corpus study will be applied here to word order variation in the Mittelfeld of the German clause.