Sideways: five methodological studies of sociolinguistic interviews*

Abstract

Abstract Five interlocking case studies of variation in and between situations are reported. In all cases a sociolinguistic interview is contrasted with another speech event. The material is from the LANCHART panel study of variation in the Danish speech community in real time. Contrasting speech events are characterized using a genre classification and focusing in each case on the genre dispersion as a measure of how varied the speech event was. Two different phonetic variables are studied, the short (æ) and the (ɛŋ) variable. Four of the five case studies involve adults who also participated in interviews approximately 20 years later. For those informants, a comparison is made with the new recordings in order to evaluate claims of change in real time. Both auditory results and acoustic measurements are documented. The fifth case study concerns youngsters recorded in the new round of recordings (the S2), hence there is no newer recording to compare with. In all cases the older (æ) variable is sensitive to a change in situation whereas the newer (ɛŋ) variable only varies with situation for the young informants. In the final section, we discuss possible consequences for comparability and for the methodology of empirical (socio)linguistics.