Linguistic variation has consistently been found to have social meaning in its association with the status and stance of speakers in the context of interaction. This indexical function of variation can contribute to the advancement of ongoing linguistic change. Style shifting in individual sociolinguistic interviews is an indirect indication of social meaning, but the clearest demonstrations are found in studies of speakers in a range of social contexts. We explore and clarify the nature of social meaning in variation, and its relation to linguistic change. Phonological variables are most readily adapted to convey social meaning by their frequency, flexibility and freedom from referential functions. After providing several kinds of evidence of social meaning in phonological variation, we argue that meaning accrues specifically to concrete soundsto phonetic elements and not to the phonological structures in which those sounds participate. Mergers, near mergers, splits, chain shifts and parallel shifts are not generally objects of social perception, conscious or unconscious, and are motivated by more abstract principles of change.
FOREWORDThe authors of this article approach the study of social meaning from a joint background in the approach to linguistic variation as 'orderly heterogeneity,' reflecting the influence of Uriel Weinreich at Columbia University in the 1960s. We both spent our youth immersed in the sociolinguistic currents of New Jersey's Bergen County, where we learned early that the exponents of social class were realized as distance from local phonology, and where we experienced similar day-to-day interaction with socially distinctive and classrelated character types. In the years that followed, Labov has studied the larger structure of variation with a focus on linguistic change, as evidenced by studies of random samples, neighborhoods, and small group interaction. Eckert's focus, based on long-term ethnographic studies, has been on meaning as evidenced in the use of a wide range of variables in the stylistic construction of socially located personae. The variables we consider in this paper are those that coincide in our workvariables that have been found to show both clear macro-social patterns and stylistic sensitivity to interpersonal context, particularly sound changes in progress. And while the authors continue to debate issues of awareness and agency in this process, our differences are independent of the observations that follow.