Language and Categorization

This chapter is focussed on linguistic bias in intergroup relations. It is based on the linguistic intergroup bias model, according to which people use different words for describing people and their behaviour on the basis of group membership. In particular, they attribute positive behaviours of ingroup members and negative behaviours of outgroup members to stable enduring characteristics, while attributing negative behaviours of ingroup members and positive behaviours of outgroup members to transitory characteristics dependent on situation or context. This kind of linguistic bias may occur not only in informal communication but also in the mass media, where it can reinforce positive or negative social stereotypes without viewers or readers necessarily being aware how this process is taking place. The chapter concludes that recognizing and limiting the use of such biased language is an important component in producing quality journalism.

The Effect of Perspective-Taking on Linguistic Intergroup Bias

In this experiment, we examined the effect of perspective-taking—actively contemplating others’ psychological experiences—on linguistic intergroup bias. We asked some participants to adopt the perspective of a character (an Italian or a Maghrebian), while others did not receive similar instructions, and complete a short dialogue comprised of a series of vignettes, resulting in a 2 (perspective-taking: presence vs. control) × 2 (group: ingroup vs. outgroup) between-participants design. We analyzed the texts produced on the basis of the linguistic category model. As expected, participants were more likely to describe the outgroup member using less abstract terms when we asked them to take the perspective of a Maghrebian. Since the level of abstraction of the terms used to describe a person’s behavior is considered an index of stereotype use, one might describe Maghrebians less stereotypically when he or she can see the world from their perspective. The results extend previous findings on the role of perspective-taking as it relates to reducing intergroup stereotypes.

Intergroup Threat and the Linguistic Intergroup Bias: A Stress Biomarker Study

This study investigates the physiological consequences of derogation. In the face of an ingroup threat, an opportunity to derogate the outgroup is associated with increases in salivary cortisol, a stress biomarker. These findings support the intergroup anxiety model, which suggests that following an anxiety-inducing threatening experience, outgroup evaluations can amplify emotional reactions and stress. In this study, we investigated whether threatened participants who derogated would show higher cortisol levels. Canadian undergraduates ( N = 169) read either a threatening or favourable passage about Canadians. They then engaged in either a linguistic intergroup bias (LIB) or control task. Salivary samples were collected prethreat, postthreat, and posttask. The results suggest that ingroup threat was associated with increases in cortisol levels. Threatened participants who displayed LIB showed higher cortisol levels than those who were threatened and did not show LIB. These findings illustrate the importance of incorporating a neuroscientific approach.

Social Identity and the Linguistic Intergroup Bias: Exploring the Role of Ethnic Identification in the Context of Intergroup Relations Between Hong Kong and Mainland China

Two survey experiments examined how linguistic intergroup bias (LIB) varies according to ethnic identification within a homogenous ethnic group (Hong Kong citizens). Study 1 showed that Hong Kong citizens who identified as “Hongkonger” used more abstract expressions to describe prosocial behaviors of the in-group (Hong Kong citizen) and antisocial behaviors of the out-group (Mainland Chinese); those who identified as “Chinese” exhibited less LIB. Study 2 found similar results for a context based on location of behaviour (in Hong Kong vs. in China) rather than the nationality of the protagonist. The combined evidence suggests that LIB can have an important intragroup as well as intergroup dimension.

Evidence of Linguistic Intergroup Bias in U.S. Print News Coverage of Immigration

This study content analyzed all print newspaper articles addressing U.S. immigration from Mexico appearing in Arizona, California, New Mexico, and Texas during a 1-year period for presence of linguistic intergroup bias. Across all four states, negative statements outnumbered positive statements; this negativity bias was more pronounced for out-group than in-group statements in all states except California. Consistent with the linguistic intergroup bias, positive in-group and negative out-group statements were encoded using more abstract language than negative in-group and positive out-group statements.