As work in sociolinguistics explores new variables and communities, linguistic phenomena whose wide range of social meanings pose a challenge for any first-wave (Eckert 2012) perspective are uncovered. I focus on the Hebrew pharyngeal consonants, which have a particularly unusual combination of ideological associations. They are stigmatized and associated with speakers that are underprivileged in Israeli society: Mizrahis (Jews of Middle Eastern descent) and Palestinians; nevertheless, since they are also an older prestige form, they are also seen as ‘correct’, unlike many stigmatized variants. Using data from sociolinguistic interviews and media, I argue that the patterns of stylistic variation show that although they may initially appear irreconcilable, the disparate meanings of the pharyngeals are not mutually exclusive: speakers can and do invoke the full set of indexical links in the same interaction. These data demonstrate the significance of Eckert’s (2008) indexical field, as its ability to include potentially conflicting associations captures an intuitive part of how speakers understand social meaning – a whole that is greater than the sum of its indexical parts.

This chapter explores the cognitive links necessary to create and use indexical meaning of sociolinguistic variation, specifically, whether different sociolinguistic behaviors – speech perception, speech production, and sociolinguistic evaluation – depend on the same associative links between linguistic and social concepts. Three effects are replicated and the correlations between them examined: 1 influence of speaker gender on /s/ production; 2 influence of linguistic forms (/s/ variants) on speaker evaluation (masculinity judgments); 3 influence of social information (masculinity) on speech perception (placement of the /s/-/?/ boundary). Despite successful replication of all three effects, little evidence was found for correlations across participants: that is, participants with a particularly strong tendency to infer masculinity from hearing a speaker’s /s/ production were no more likely than others to show a large shift in their /s/-/?/ boundary in response to the perceived masculinity of the talker, or to have a particularly gender-typical /s/ production themselves. This provides preliminary evidence that the mental associations used in indexical practices may develop independently of one another.

In interactions with their peers, South African male youths use a way of talking that is different from the local varieties spoken in their communities. Labelled tsotsitaals, this practice involves inserting a slang lexicon into the syntactic base of a local language combined with other paralinguistic features. Most studies of this phenomenon have looked at Afrikaans and Bantu language-based tsotsitaals but there are also several instances of English-based tsotsitaals among male youths in English-speaking communities. Across all tsotsitaals, English-derived words make up a substantial part of the slang lexicon. Ethnographic studies among young men who use Bantu-based tsotsitaals show that English plays a key role in indexing different identities. Based on the social meanings they attribute to English use, they incorporate English in a variety of ways making up different styles of tsotsitaal. These styles reflect different social levels, identities, attitudes and aspirations. The way English is incorporated into tsotsitaals challenges common approaches that treat this performance register as an autonomous referential system.