Introduction to the Chapters by Werker and Jusczyk: Rachel Clifton
When considering this loss, we should remember (and Werker reminds us) that not every non-native contrast gets 'reorganised' in the same way. The work of Best and colleagues with Zulu clicks is a theoretically interesting example (Best, McRoberts, & Sithole, 1988). These authors suggested that this discrimination is not lost because it is not assimilable to English phonology, and is therefore spared because it is not processed as speech. In a critical review of the cross-language literature, Burnham (1986) proposed that some contrasts, which he labelled 'fragile', are lost in infancy if they are not experienced at that time, whereas other contrasts, labelled 'robust', are not lost until early childhood (4-8 years). Furthermore, the latter can easily be retrained at older ages, whereas the former cannot. Burnham suggested that robust contrasts may have more distinct psychoacoustic cues, as well as participation in more allophonic variation in the person's native language. Whether or not one agrees with Burnham, his distinction stresses that care must be taken in selecting the particular non-native contrasts in cross-language research, because general conclusions cannot be drawn on the basis of one or two contrasts.