In this article we define and illustrate sociophonetic variation within speech, highlighting both its pervasiveness and also the relatively minor role it has played in the development of phonetic and phonological theory. Reviewing evidence from studies of adults and children, we suggest that cognitive representations of words combine linguistic and indexical information, and that both types of information are present from the first stages of acquisition. We suggest that an exemplar-based model of phonological knowledge offers the most productive means of modeling sociophonetic variation. We discuss some of the characteristics of an exemplar-based account of sociophonetic variability and highlight some strands of investigation which would facilitate its further development.