3 - The Data, the Community and a Data-Led Identification of MWEs

Summary

Chapter 3 presents a new procedure for extracting formulaic language from corpus materials and further methodological aspects. In the first section, the speech community whose language is investigated is introduced and it is shown that the community of users of the Swiss variety of standard written German is a speech community sharing not only this common language variety but sharing in a culture that, at one level, is distinguished both from other linguistic communities within the country and from German-speaking communities elsewhere. The specifics of the linguistic data that form the basis of subsequent investigations are also presented. The second section concerns the identification of instances of formulaic language in corpus data. A robust operationalisation of the MWE-concept is presented, and the final section of the chapter documents a new procedure for extracting operationalisation-compliant sequences from corpus materials. It is shown that this procedure delivers accurate as well as comprehensive results and is therefore well suited for the purposes of the study.