Perceptive evaluation of Levenshtein dialect distance measurements using Norwegian dialect data

The Levenshtein dialect distance method has proven to be a successful method for measuring phonetic distances between Dutch dialects. The aim of the present investigation is to validate the Levenshtein dialect distance with perceptual data from a language area other than the Dutch, namely Norway. We calculate the correlation between the Levenshtein distances and the distances between 15 Norwegian dialects as judged by Norwegian listeners. We carry out this analysis to see the degree to which the average Levenshtein distances correspond to the psychoacoustic perception of the speakers of the dialects. In 1995, Kessler introduced the use of the Levenshtein distance as a tool for measuring linguistic distances between language varieties. He applied the algorithm to the comparison of Irish dialects. The Levenshtein distance is a string edit distance measure. On the basis of linguistic distances between dialectal varieties, dialect areas can be found. More innovative is the possibility of drawing dialect maps that reflect the fact that dialect areas should be considered as continua and not as areas separated by sharp borders. Its application to the Dutch language area has produced convincing results (see Heeringa, 2004; Nerbonne & Heeringa, 1998). The results are partly similar to the map of Daan and Blok (1969), which may be considered as the most authoritative Dutch dialect map up till now. Still, it is desirable to validate the method further. In this article we validate the Levenshtein distance. We will investigate to what extent dialect distances found with Levenshtein distance correlate with distances as perceived by the dialect speakers themselves. We will try to find an answer to the following question: May Levenshtein distance-based dialect distances be considered as a good approximation of the perceptual distances? To answer this question, we will use a set of 15 Norwegian varieties. Results for Dutch may be impressive, but the Dutch dialect area is a flat, regularly populated landscape. In contrast with this, the Norwegian dialect area is less The present article reports on part of a study supported by NWO, the Netherlands Organization for Scientific Research. We are grateful for the permission from Kristian Skarbø and Jørn Almberg to use their material and for the help of Jørn Almberg during the whole investigation. We thank Saakje van Dellen for her obliging help with the data entry and Peter Kleiweg for letting us use the programs that he developed for the visualization of the maps and dendrograms in this article. Finally, we would like to thank John Nerbonne for valuable comments and for correcting our English. Language Variation and Change, 16 (2004), 189–207. Printed in the U.S.A. © 2004 Cambridge University Press 0954-3945004 $9.50 DOI: 10.10170S0954394504163023