Inferring Identity From Language: Linguistic Intergroup Bias Informs Social Categorization

The present research examined whether a communicator’s verbal, implicit message regarding a target is used as a cue for inferring that communicator’s social identity. Previous research has found linguistic intergroup bias (LIB) in individuals’ speech: They use abstract language to describe in-group targets’ desirable behaviors and concrete language to describe their undesirable behaviors (favorable LIB), but use concrete language for out-group targets’ desirable behaviors and abstract language for their undesirable behaviors (unfavorable LIB). Consequently, one can infer the type of language a communicator is likely to use to describe in-group and out-group targets. We hypothesized and found evidence for the reverse inference. Across four studies, individuals inferred a communicator’s social identity on the basis of the communicator’s use of an LIB. Specifically, participants more strongly believed that a communicator and target shared a social identity when the communicator used the favorable, rather than the unfavorable, LIB in describing that target.