Cognitive Consequences of Linguistic Change in Progress – Bill Labov (University of Pennsylvania)

When:
October 15, 1996 all-day
1996-10-15T00:00:00-04:00
1996-10-16T00:00:00-04:00

Abstract
Sociolinguistic research over the past two decades finds linguistic change in progress in the vowel systems of all the large cities of North America, leading to an increasing diversification of English dialects through mergers and chain shifts. The chain shifts form three distinct patterns of unidirectional change in opposing directions: the Northern Cities Shift, the Southern Shift, and the Canadian Shift. The Phonological Atlas of North America, based on an ongoing telephone survey of urbanized areas, shows that these chain shifts are highly differentiated along geographic lines that follow the original settlement patterns.Since these chain shifts rotate vowels into radically different phonetic positions, the question arises as to what extent they interfere with cross-dialectal comprehension. Several series of experiments were carried out by exposing listeners from Philadelphia, Chicago and Birmingham to the vowel systems of their own and other cities, with varying degrees of context. The results show considerable impairment in the capacity of phonological oppositions to discriminate words. Though local listeners show some advantage in comprehension over outsiders, this reduction in intelligibility applies within as well as across dialects. Exposure to a wider range of dialects characteristic of college-educated subjects reduces ability to recognize local dialects, though it enhances the capacity to use phrase and sentence context to decipher utterances.These results sharpen the problem of understanding the causes of linguistic change, since increased of cross-dialectal contact through mass media appears to co-exist with increasing local diversity. The results also bear crucially on the problem of automatic speech recognition, since solutions to the problem of dialect diversity in speech recognition cannot be modeled on human performance.

Center for Language and Speech Processing