On the nature of prejudice

Canadian Psychology, Jan 1994 by Zanna, Mark P

Abstract

It was proposed that, in addition to stereotypic beliefs (beliefs that typical members of the outgroup possess certain characteristics or traits), prejudice (or negative attitudes toward outgroups) is based on symbolic beliefs (beliefs that typical members violate cherished traditions, customs, and values) as well as on emotions and past experiences that are associated with the outgroup. In several studies (the first of which is presented in the present paper) we have found that (1) although related, stereotypes, symbolic beliefs, emotions, and past experiences are far from redundant, and (2) each factor is, indeed, an important predictor of prejudicial attitudes. Thus, there would appear to be more to prejudicial attitudes than stereotypic beliefs. We have also discovered that the relatively more negative attitudes held by individuals high in authoritarianism are predicted best by symbolic beliefs. The implications of these results for understanding the development and reduction of prejudice are discussed.

Traditionally, social psychologists have believed there is a strong relation between prejudice and stereotypes. Some have proposed that prejudicial attitudes are based on, or are the result of, stereotypical beliefs. Others have suggested that stereotypical beliefs justify, or rationalize, prejudicial attitudes. If I (along with everybody else in my reference group) believe Afro - Canadians to be lazy and stupid, it would make sense for me to evaluate Afro - Canadians unfavourably. Conversely, if, for some other reason, I were to evaluate Afro - Canadians unfavourably, I might then be motivated to believe that Afro - Canadians are lazy and stupid.

This conceptualization, especially the notion that prejudice is based on stereotypes, is consistent with the way attitudes, in general, have been conceptualized in recent years by social psychologists. Perhaps the most prominent theory of attitudes has been that proposed by Martin Fishbein and Icek Ajzen (Ajzen & Fishbein, 1980). According to Fishbein and Ajzen, attitudes (which they define as our feelings or affective responses to an attitude object) are based upon evaluative beliefs. In this view, our attitudes toward a group are based upon the evaluative implications of the characteristics or traits we attribute to the group, which seems to me to be the same as saying that the amount of our prejudice is based onthe extent to which we endorse the stereotype of the group.

Recently, several attitude theorists, myself included, have suggested a slightly modified view of the attitude concept. For example, I (Zanna & Rempel, 1988) have proposed that affective and evaluative responses should no longer be considered synonymous (cf. Abelson, Kinder, Peters, & Fishe, 1982) and that attitudes should be viewed as evaluations based on, or developed from, three general classes of information: (1) affective information or the emotions associated with the attitude object, (2) cognitive information or the beliefs associated with the attitude object, and (3) information concerning past behaviours or behavioural intentions. [I've further suggested, again along with others (e.g., Eagly & Chaiken, 1993), that evaluations or attitudes influence three modes of response, including affective, cognitive and behavioural responses.]

When evaluations are based primarily on (utilitarian or instrumental) beliefs about the attitude object, this view can be reduced to something like the formulation proposed by Fishbein and Ajzen. When evaluations are based primarily on affects produced by or associated with the attitude object, this view can resemble the formulation proposed by Bob Zajonc (Zajonc, 1980). Finally, when evaluations are based on inferences from past behaviour, this view can be similar to Daryl Bem's theory of self - perception (Bem, 1972).

In any event, if attitudes are not always entirely based upon the sort of utilitarian or instrumental beliefs suggested by the Fishbein and Ajzen model, then prejudicial attitudes are perhaps not always based entirely upon stereotypical beliefs. In fact, several years ago Bob Gardner and his colleagues (at the University of Western Ontario) demonstrated that English Canadian prejudice toward French Canadians was virtually unrelated to their consensual stereotype of French Canadians (Gardner, 1994; Gardner, Wonnacott, & Taylor, 1968). More recently, Chuck Stangor and his colleagues (at the University of Maryland) demonstrated that whites' prejudice toward Afro - Americans is based on the affects or emotions that they associated with Afro - Americans, in addition to their stereotypical beliefs (Stangor, Sullivan, & Ford, 1991).

And, in recent years David Sears, John McConahay and their colleagues, have suggested that prejudice toward Afro - Americans (in addition, perhaps, to being based to a greater extent upon affect) is basedupon (in my terms) symbolic beliefs such as "Blacks are getting more than they deserve," "the streets are unsafe today," and so on, rather (or to a greater extent) than on traditional stereotypic beliefs (e.g., McConahay & Hough, 1976; Sears, 1988).