Attaining a Native Accent

In their paper Can Late Learners Attain a Native Accent in a Foreign Language: The Age Factor In Second Language Acquisition, Bongaerts, Planken, and Schils begin with a discussion of the ability of non-native language learners to acquire a native accent. They cite many examples of previous studies which show a lack of ability for non-native learners to acquire a native accent past a certain age. The discussion based on the research of Lenneberg seems to suggest that there exists a critical age for native accent attainment.  However, these studies are not uniform in agreement on the precise critical age or its relationship to other areas of language acquisition—such as grammar, syntax, etc. In each case, the studies did show conclusively that near native accent attainment was most likely to be achieved the earlier the age of arrival (AA) into the country of immersion.

The basis of the experiment by Bongaerts, Planken, and Schils was to test the theory of a critical age. To test this, the authors composed three groups of speakers. The first group consisting of native British English speakers who were determined to have no regional accent. The second group consisted of native Dutch speakers whose second language was British English and had studied English extensively in graduate studies, often lecturing in the language. The third group was comprised of native Dutch speakers whose accent was determined to be definitely non-native. Each group was given four tasks and their pronunciation was recorded. The recordings were judged by four native British English speakers from York.

The results showed that the judges made almost no distinction between Groups 1 and 2, and that, in fact, four of the speakers from Group 2 (native Danes) outscored all speakers in Group 1 (native Britons). The members of Group 3 performed as predicted and were judged to be definitely non-native.  A possible explanation for these results is that the judges were all from York and were able to detect slight regional influences in the members of Group 1.  Also, Group 2 was trained in a form of English pronunciation used by British journalists and broadcasters and considered to be very geographically neutral. Another possibility for the nature of the results is that the fact that Dutch and English are both Germanic languages. Perhaps a study that tested the relationship between the speakers of English and another language, such as French, would show greater differences between actual native speakers and non-native learners because of differences in rhythm and stress.

However the study does suggest that it is not impossible, as Scovel would suggest, for non-native learners to acquire a native accent. The authors do not discredit the advantages of second language acquisition in early childhood as opposed to post-pubescent learning, but they do suggest the critical age be instead regarded as a sensitive age.