This chapter argues for incorporation of concepts and methods from the domain of Implicit Social Cognition (ISC) into the field of language attitudes research. As support, this chapter reports on a quantitative study that employed both an audio Implicit Association Test and traditional self-report questionnaires to measure participants' implicit and explicit attitudes toward foreign and U.S. accented speech stimuli. The IAT revealed a pro-U.S. accent bias, while the explicit measure found a pro-foreign accent bias. These results support the argument that the distinction between implicit and explicit attitudes as separable attitude constructs resulting from distinct mental processes is an important one for language attitudes research and that both attitude constructs should be evaluated when studying language attitudes.
TopBackground
Over the past half century or more, traditional language attitudes studies have produced a rich body of literature and a variety of findings, establishing quantitatively—and not inconsequentially—that reactions to foreign accented speech are complex. Specifically, traditional language attitudes research have confirmed not only the existence of a general negative affect toward nonstandard accents (e.g., Gluszek & Dovidio, 2010; Lambert, 1967; Rubin & Smith, 1990), but also a variety of reactions to specific, identified accents, attributing this variation to a number of factors, including reaction type (e.g., affective and evaluative reactions), speaker trait (e.g., solidarity [kindness, friendliness, etc.] and status [competence, intelligence, etc.]) (Cargile & Giles, 1997; Ryan, 1982), the aggressiveness of the message (Cargile & Giles, 1997), stereotypes associated with the speaker’s nationality (Frumkin, 2007; Kristiansen, 2001), and the degree of nonstandardness of the speaker’s accent (Brennan & Brennan, 1981; Bresnahan, Ohashi, Nebashi, Liu, & Shearman, 2002). Interestingly, the idea of a general negative affect was never reconciled with the sometimes positive reactions revealed in these studies of reactions to specific accents.
The general bias against nonstandard language varieties has led to the conclusion that nonstandard accents are generally dispreferred (e.g., Gluszek & Dovidio, 2010; Lambert, 1967; Rubin & Smith, 1990). This finding seems to explain the significant negative social, political, judicial, and economic consequences potentially suffered by individuals who speak with a nonnative accent (Kinzler, Shutts, DeJesus & Spelke, 2009; Lippi-Green, 1997; Matsuda, 1991). For the underlying cause of these negative reactions, researchers have relied historically on concepts from Social Identity Theory (SIT; Tajfel & Turner, 1986/2004) and its notions that social identity is derived from group membership, that social interaction is essentially intergroup interaction, and that the negative reactions results from relative unfavorable comparison of outgroup members (Turner & Giles, 1981). These studies demonstrated that language attributes signal group membership status (Bresnahan & Kim, 1993; Bresnahan et al., 2002; Giles, Hewstone, Ryan & Johnson, 1987; Reid & Giles, 2005; Ryan, 1983; White & Li, 1991) and are therefore at the very core of intergroup behavior. The power of group membership was also studied in narrower focus in many traditional language attitudes studies that considered reactions to specific, identified foreign nationalities, including Mexican (Ryan, Carranza & Moffie, 1977); Malaysian (Gill, 1994); Japanese (Cargile & Giles, 1998); Chinese (Cargile, 1997); Mexican, Lebanese and German (Frumkin, 2007); and Italian, Norwegian, and Eastern European (Mulac, Hanley & Prigge, 1974).