A B S T R A C TThe 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.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.

The three mainland Scandinavian languages (Danish, Swedish and Norwegian) are so closely related that the speakers mostly communicate in their own languages (semicommunication). Even though the three West Germanic languages Dutch, Frisian and Afrikaans are also closely related, semicommunication is not usual between these languages. In the present investigation, results from intelligibility tests measuring the mutual intelligibility of Danish, Norwegian and Swedish were compared with results of similar tests of mutual intelligibility between speakers of Dutch, Frisian and Afrikaans. The results show that there are large differences in the level of intelligibility depending on test group and test language. Correlations between the intelligibility scores and linguistic distance scores showed that intelligibility can to a large extent be predicted by phonetic distances, while intelligibility is less predictable on the basis of lexical distances.

By means of a large-scale web-based investigation, we established the degree of mutual intelligibility of 16 closely related spoken languages within the Germanic, Slavic and Romance language families in Europe. We first present the results of a selection of 1833 listeners representing the mutual intelligibility between young, educated Europeans from the same 16 countries where the test languages are spoken. Next, we present the data from a sub-group of listeners who had not learned the test language and had had minimal exposure to it. This allows us to investigate how well the listeners understand the test language on the basis of structural similarities between their own language and the test languages. Finally, we compare the results of the two data sets to the traditional genealogic characterisation of the three language groups. We expect the intelligibility results from the second group of listeners who had had minimal exposure to the test language to be a better reflection of the genealogical characterisation than the results from the larger group who had sometimes been exposed to the test language or had learned it at school. ARTICLE HISTORY

We examine various string distance measures for suitability in modeling dialect distance, especially its perception. We find measures superior which do not normalize for word length, but which are are sensitive to order. We likewise find evidence for the superiority of measures which incorporate a sensitivity to phonological context, realized in the form of n-gramsalthough we cannot identify which form of context (bigram, trigram, etc.) is best. However, we find no clear benefit in using gradual as opposed to binary segmental difference when calculating sequence distances.

The aim of the present investigation 1 was to get an impression of geographic influences on dialectal variation in a country. In previous investigations, the correlations between linguistic distances and geographic distances using dialect data from the Netherlands and Norway were calculated (Gooskens and Heeringa 2004, Nerbonne et al. 1996). The results showed a high correlation in the case of the Dutch data while the correlation was considerably lower in the case of the Norwegian data. This seems to reflect the fact that for Norway, in particular, the direct distance between two settlements does not reflect the difficulty of travel and therefore social contact, which is expected to play a role in keeping linguistic distance within limits. Holland is a country with a flat, regularly populated landscape with few natural obstacles such as mountains and rivers. This is in great contrast with Norway with its high mountains and many J/ords which especially in the past made it quite difficult to travel between places. These differences in geographical situations are clearly reflected in the correlations between the linguistic and geographical distances between the dialects of the two countries.The present investigation aims to find more successful ways of predicting linguistic distances by means of geographic distances in Norway. To this end, old and new traveling data are used, providing information about travel times by road, train, and boat between the places where the different dialects are spoken. The results show that a large part of the linguistic variation can be accounted for by geography in Norway, just as in the Netherlands. However, in the case of a geographically more complicated This article is based on a paper read at the Second International Conference on Language Variation in Europe, Uppsala, 12-14 June 2003. The author wishes to thank the following people for their help at different stages in the investigation: Dag Bj0rnland (BI Norwegian School of Management) for providing the data on old travel times, Femke Jongerius (University of Groningen) for data entering, Wilbert Heeringa (University of Groningen) for help with the Levenshtein distances and statistics and for making the maps, Peter Kleiweg (University of Groningen) for his software for creating the maps and John Nerbonne (University of Groningen) for comments on an earlier version of this paper.Brought to you by | Nanyang Technological University Authenticated Download Date | 6/7/15 4:58 PM Travel time as a predictor of linguistic distance 39 country like Norway, travel times reflect the influence of geography on linguistic variation better than straight-line distances, particularly when the historical aspect is also taken into consideration since the former circumstances of travel are reflected in the modern dialects. Travel times can vary independently from straight-line distances, and this variation was more pronounced in the past, when the present dialect situation crystallized.