Explaining Danish-Swedish asymmetric word intelligibility : An error analysis

Previous research has shown that the Danish-Swedish mutual intelligibility is asymmetric at the text level. Danes perform better in tests developed to investigate intelligibility of Swedish at the global level of whole texts than Swedes participating in Danish tests. This asymmetry has usually been attributed to non-linguistic factors such as a more positive attitude towards Swedes and more experience with Swedish among Danes than vice versa. Also strong evidence has been found for general linguistic explanations of the asymmetry such as speech rate which has been measured to be higher in Danish than in Swedish. Also Danes seem to benefit from the fact that Swedish is similar to written Danish and Swedish when they listen to spoken Swedish. This benefit is smaller for Swedes because spoken Danish has developed away from its written Swedish and Danish form. In the present investigation we investigated Danish-Swedish mutual intelligibility at the word level. We also found an asymmetry at this level and therefore conclude that at least part of the explanation for the asymmetric Danish-Swedish intelligibility has to do with linguistic characteristics that are present in single words. To gain insight into the linguistic factors that cause this asymmetry we made a detailed analysis of the kind of errors that the listeners made when listening to cognate word pairs with asymmetric intelligibility. We focus on sound correspondences causing asymmetric problems.