Key points are not available for this paper at this time.
Abstract This study examines the retention of a non-native dialect feature by British Asians in London. We examine the use of one Punjabi feature ( t -retroflexion) and one British feature ( t -glottaling) across three groups: first-generation non-native immigrants and two age groups of second-generation British Asians. Cognitively oriented models predict that non-native features will either be innately blocked (Chambers, 2002) or reallocated by native generations. A socially oriented model allows for more gradual change. Contrary to the cognitive view, the older second generation neither blocks nor clearly reallocates use of t -retroflexion; they closely mirror the first generation's non-native use. However, they simultaneously control nativelike t -glottaling, reflecting a robust bidialectal ability. It is the younger second generation who exhibit focused reallocation in the form and function of t -retroflexion. This 20-year lag corresponds to major changes in demographics and race relations in the community over 5 decades. The study shows that acquisition of the local dialect and retention of exogenous features should be seen as independently constrained rather than as mutually exclusive.
Sharma et al. (Sat,) studied this question.