University of Leuven, Belgium
Belgian Dutch and Netherlandic Dutch are very closely related languages from a diachronic as well as a synchronic point of view, which implicates that these language varieties are to a great extent mutually comprehensible. Nevertheless, as is illustrated by the subtitling of several TV programmes, certain differences exist between the Dutch language varieties in both countries. Additionally, the same holds for the different regiolects within (especially) Flanders and the Netherlands (Van Bezooijen, R. & R. van den Berg 1999).
Recently, much descriptive work has been done on these existing linguistic distances, as well as on the role of some extra-linguistic factors which can predict the success of mutual communication (Gooskens 2006). However, for the Dutch situation previous research has exclusively examined this topic by means of subjective introspection by informants; no research exists that carries out statistically supported measurements to investigate intelligibility. Therefore, this study is innovative in the field: it measures every (extra)linguistic factor by a specifically developed method.
In this project, the linguistic variance of the Low Countries is well represented. Although it does not focus on dialects (as was always the case in previous studies), it includes a Netherlandic and Belgian realisation of standard language as well as four more peripheral, colloquial varieties from both countries. All of these varieties are evaluated by secondary school students through a computer-controlled spoken lexical decision task. The informants heard ‘real’ as well as ‘pseudo’-words in various Dutch varieties, after which they had to decide as quickly as possible whether the presented words were existing Dutch words or not. The faster their reaction, the better the intelligibility.
We selected the words via a Stable Lexical Marker analysis on three corpora (blogs – footballfora – CGN). This analysis highlighted the typical lexical items of each (sub)corpus. We selected binational words (which only differ phonetically between the different varieties), national words (which are typically Flemish or Dutch) and regional words (which are lexically specific for a certain region).
Furthermore, this experiment will shed light on the precise correlation and impact of the parameters linguistic distance, language attitudes and familiarity on mutual intelligibility. Phonetic distance will be measured by the Levenshtein distance (Kessler 1995), which is an algorithm that expresses phonetic similarity in one number. Lexical distance will be measured by profile-based uniformity (Geeraerts, Grondelaers & Speelman 1999), which compares the words that speakers of different language varieties use for a wide range of concepts. Finally, language attitudes will be measured by attitudinal scales and familiarity by questionnaires.
- Geeraerts, D., S. Grondelaers & D. Speelman 1999. Convergentie en divergentie in de Nederlandse woordenschat. Amsterdam: Meertens Instituut.
- Gooskens, C. 2006. “Linguistic and Extra-Linguistic predictors of inter-Scandinavian intelligibility”. Linguistics in the Netherlands, 23(1): 101-113.
- Kessler, B. 1995. “Computational dialectology in Irish Gaelic”. Proceedings of the European ACL, 60-67. Dublin: ACL.
- Van Bezooijen, R. & R. van den Berg 1999. “Taalvariëteiten in Nederland en Vlaanderen: hoe staat het met hun verstaanbaarheid?”. Taal en Tongval, 51(1): 15-33.
Session: Paper session
Friday, April 4, 2008, 13:45-15:15