Att undersöka lyssnares perception av regional variation

Forskningsoutput: TidskriftsbidragArtikelVetenskapligPeer review

Sammanfattning

methodological issues regarding perception tests which measure the ability of informants to identify differences between different regional varieties of speech. Starting out from a general perspective and accounting for previous research on the subject, we then go on to discuss a test we constructed in order to examine how some standard-near urban dialects in Swedish-speaking Finland are perceived. The test consists of six voice recordings, two of which are from Helsinki, two from Turku, one from Kirkkonummi, and one from Vaasa. The article includes methodological discussions on the choice of voice samples for the test, on instructions given to the informants, on practical arrangements in the test situation, and on the choice of informant groups. In addition, we present and discuss the results obtained from the test. Finally, we draw some theoretical-methodological conclusions. The test was carried out with six different informant groups (121 informants in total) in Swedish-speaking Finland. Only one third of the voice samples were given a correct geographical assignment by the informants, and there is a high degree of individual variation in the test results. These results could not be accounted for by extra-linguistic factors. In general, approximately a half of the stated reasons for the informants’ assignment of a voice sample to certain locale were based on linguistic traits in the stimulus, but also other types of reasons were mentioned. This shows that the informants do not interpret linguistic variation only in terms of geography. As for the different urban dialects in focus, the study showed that the voice samples from Helsinki and Vaasa were the easiest to recognise, and that the informants were also able to point out specific linguistic features in them. Thus, the voice samples did contain traits that could potentially function as indexes of locality. However, these traits may also be interpreted as indexes for social meaning as they often evoked negative attitudes. On the other hand, few informants were able to place the voice sample from Kirkkonummi correctly, and the samples from Turku proved to be the most difficult to place. The samples from Turku were perceived as neutral to the extent that they either could not be assigned geographically at all, or were assigned to locales all over the Swedish-speaking area in Finland. Therefore, the Turku variety seems to be perceived as the closest candidate for a neutral standard variety of Finland-Swedish. The subject of this article is perceptual dialectology, and the aim is to discuss methodological issues regarding perception tests which measure the ability of informants to identify differences between different regional varieties of speech. Starting out from a general perspective and accounting for previous research on the subject, we then go on to discuss a test we constructed in order to examine how some standard-near urban dialects in Swedish-speaking Finland are perceived. The test consists of six voice recordings, two of which are from Helsinki, two from Turku, one from Kirkkonummi, and one from Vaasa. The article includes methodological discussions on the choice of voice samples for the test, on instructions given to the informants, on practical arrangements in the test situation, and on the choice of informant groups. In addition, we present and discuss the results obtained from the test. Finally, we draw some theoretical-methodological conclusions. The test was carried out with six different informant groups (121 informants in total) in Swedish-speaking Finland. Only one third of the voice samples were given a correct geographical assignment by the informants, and there is a high degree of individual variation in the test results. These results could not be accounted for by extra-linguistic factors. In general, approximately a half of the stated reasons for the informants’ assignment of a voice sample to certain locale were based on linguistic traits in the stimulus, but also other types of reasons were mentioned. This shows that the informants do not interpret linguistic variation only in terms of geography. As for the different urban dialects in focus, the study showed that the voice samples from Helsinki and Vaasa were the easiest to recognise, and that the informants were also able to point out specific linguistic features in them. Thus, the voice samples did contain traits that could potentially function as indexes of locality. However, these traits may also be interpreted as indexes for social meaning as they often evoked negative attitudes. On the other hand, few informants were able to place the voice sample from Kirkkonummi correctly, and the samples from Turku proved to be the most difficult to place. The samples from Turku were perceived as neutral to the extent that they either could not be assigned geographically at all, or were assigned to locales all over the Swedish-speaking area in Finland. Therefore, the Turku variety seems to be perceived as the closest candidate for a neutral standard variety of Finland-Swedish.
OriginalspråkOdefinierat/okänt
Sidor (från-till)35–77
TidskriftFolkmalsstudier
Volym53
StatusPublicerad - 2015
MoE-publikationstypA1 Tidskriftsartikel-refererad

Citera det här