Fuzzy dialect areas and prototype theory. Discovering latent patterns in geolinguistic variation

Abstract

In this article, a threefold link is established between the concept of dialect areas as scientific constructs, prototype theory as a descriptive model and factor analysis as an operationalisation of the former two. While the idea of using prototype theory to model emic, folk concepts of dialect areas is not new, it is here for the first time used to establish a scholarly, etic model of dialect areas, which will make it easier to compare emically and etically defined dialect areas in the future. Dialect areas can be conceived of as being crisp or fuzzy, but in most cases, they are best conceptualised as being fuzzy. Following work by Gaetano Berruto, fuzzy dialect areas are defined on the basis of sets of similarly distributed variants. In a second, more practical step, an operationalisation of this model is presented that uses factor analysis to extract spatial patterns from geolinguistic data that satisfy the model’s definition of dialect area. This methodology is illustrated by applying it to dialect data from Bavarian Swabia (Southern Germany). The geolinguistic structures revealed demonstrate the utility of factor analysis as a tool both for a detailed, in-depth differentiation of fuzzy dialect areas and for the detection of hitherto unknown, even very weak spatial patterns.