Г-Л.

Social Psychology Quarterly 1995, Vol. 58, No. 4, 255-269

A TALE OF TWO THEORIES:

A CRITICAL COMPARISON OF IDENTITY THEORY WITH SOCIAL IDENTITY THEORY*

MICHAEL A. HOGG DEBORAH J. TERRY

KATHERINE M. WHITE

The University of Queensland

Identity theory and social identity theory are two remarkably similar perspectives on the dynamic mediation of the socially constructed self between individual behavior and social structure. Yet there is almost no systematic communication between these two perspectivies; they occupy parallel but separate universes. This article describes both theories, summarizes their similarities, critically discusses their differences, and outlines some research directions. Against a background of metatheoretical similarity, we find marked differences in terms of 1) level of analysis, 2) the role of intergroup behavior, 3) the relationship between roles and groups, and 4) salience of social context and identity. Differences can be traced largely to the microsociological roots of identity theory and the psychological roots of social identity theory. Identity theory may be more effective in dealing with chronic identities and with interpersonal social interaction, while social identity theory may be more useful in exploring intergroup dimensions and in specifying the sociocognitive generative details of identity dynamics-

Identity theory (e.g., Burke 1980; McCall and Simmons 1978; Stryker 1968; R.H. Turner 1978) and social identity theory (e.g., Hogg and Abrams 1988; Tajfel and Turner 1979; J.C. Turner 1982, 1985; J.C Turner et al. 1987) are two perspectives on the social basis of the self-concept and on the nature of normative behavior. These two perspectives have many similarities. Both address the social nature of self as constituted by society, and eschew perspectives that treat self as independent of and prior to society. Both regard the self as differentiated into multiple identities that reside in circumscribed practices (e.g., norms, roles), and they use similar words and a similar language—but often with quite different meanings (e.g., identity, identity salience, commitment).

Remarkably, the two theories occupy parallel but separate universes, with virtually no cross-referencing. The coexistence of such apparently similar explanatory frameworks is problematic for social science, and to our knowledge no published attempt has been made to systematically compare them. The aim of this article is to compare identity

* Correspondence should be addressed to Michael A. Hogg, Department of Psychology, University of Queensland, Brisbane, QLD 4072. Australia. E-mail: mike ©psy.uq.edu.au. FAX: +61 (7)3365-4466.

theory with social identity theory in order to highlight their similarities and differences, and to suggest some critical observations that may indicate possible directions for future research. We hope to encourage dialogue between proponents of the two theories that may pave the way for comparative studies and subsequent distinctions between and articulation of the two theories.

Identity theory is principally a microsociological theory that sets out to explain individuals' role-related behaviors, while social identity theory is a social psychological theory that sets out to explain group processes and intergroup relations. Both theories place their major theoretical emphasis on a multi-faceted and dynamic self that mediates the relationship between social structure and individual behavior. General differences can be attributed, to a significant extent, to the different disciplinary roots of the two theories—sociology for one and psychology for the other. More specific differences include the degree and type of specification of sociocognitive processes that are associated with identity-related behavior, and the relative emphasis placed on roles and on intergroup relations.

We begin by overviewing the two theories in sufficient detail, we hope, to give an initial

255

256

SOCIAL PSYCHOLOGY QUARTERLY

understanding to someone not familiar with one theory or the other. This task is difficult in itself because of the historical lack of cross-referencing between the two theories, and because of differences in theoretical emphasis among identity theorists. We go on to identify some similarities and differences between the theories, suggest strengths and weaknesses, and conclude with some thoughts on implications for theory and research.

IDENTITY THEORY

Identity theory (Stryker 1968, 1980, 1987; Stryker and Serpe 1982; also see Burke 1980; McCall and Simmons 1978; R.H. Turner 1978) explains social behavior in terms of the reciprocal relations between self and society. It is strongly associated with the symbolic interac-tionist view that society affects social behavior through its influence on self (Mead 1934; also see Blumer 1969), and was developed in part in order to translate the central tenets of symbolic interactionism into an empirically testable set of propositions (Stryker 1980, 1987; Stryker and. Serpe 1982). Identity theory, however, rejects the symbolic interactionist view of society as a "relatively undifferentiated, cooperative whole" (Stryker and Serpe 1982: 206), arguing instead, that society is "complexly differentiated but nevertheless organized" (Stryker and Serpe 1982:206). This vision of society forms the basis for the central proposition on which identity theory is predicated: that as a reflection of society, the self should, be regarded as a multifaceied and organized construct. Identity theorists refer to the multiple components of self as identities (or, more specifically, role identities). The notions of identity salience and commitment are used in turn to account for the impact of role identities on social behavior.

Although identity theory originally was formulated by Stryker (Stryker 1968, 1980, 1987; Stryker and Scrpe 1982), the term is now used more widely to refer also to related theoretical work that acknowledges links between a multifaceted notion of self and the wider social structure (Burke 1980; McCall & Simmons 1978; R.H. Turner 1978). This wider perspective, although still clearly grounded in symbolic interactionism, is not homogeneous. There are differences in emphasis and interpretation: Stryker, for instance, views identities as more stable than do some other identity theorists, and tends to

place less emphasis on the key symbolic interactionist mechanism of "taking the role of the other." Because this diversity makes it difficult to provide the brief overview required by this article, we have opted to lean toward Stryker's exposition, but with appropriate recognition of alternative emphases.

The general perspective of identity theory forms the basis for a relatively large body of microsociological literature concerned with predicting role-related behavior (e.g., Simon 1992; Thoits 1991). Accordingly, identity theorists have tended to focus on individualistic consequences of identity-related processes (Rosenberg 1981).

Role Identities

Symbolic interactionists such as Mead (1934) and Cooley (1902) considered the self to be a product of social interaction, in that people come to know who they are through their interactions with others; in this perspective, a core mechanism is that of "taking the role of the other." Because people tend to interact in groups, it is perhaps not surprising that people may have as many distinct selves as there are distinct groups whose opinions matter to them (James [1890J 1950). These two ideas come together in identity theory, which views the self not as an autonomous psychological entity but as a multifaceted social construct that emerges from, people's roles in society; variation in self-concepts is due to the different roles that people occupy. Stryker proposed that we have distinct components of self, called role identities, for each of the role positions in society that we occupy (Stryker 1968, 1980; also see Burke 1980; Stryker and Serpe 1982; Wiley 1991). For example, a person's role identities may include the fact that she is a mother, a wife, a daughter, a social worker, and a blood donor.

Role identities are self-conceptions, self-referent cognitions, or self-definitions that people apply to themselves as a consequence of the structural role positions they occupy, and through a process of labeling or self-definition as a member of a particular social category (Burke 1980; Thoits 1991). Role identities provide meaning for self, not only because they refer to concrete role specifications, but also because they distinguish roles from relevant complementary or counterroles (e.g., Lindesmtth and Strauss 1956). For example, "the role of mother takes on

A TALE OF TWO THEORIES

257

meaning in connection with the role of father, 'doctor' in connection with 'nurse,' and so on" (White and Burke 1987:312), Ultimately it is through social interaction that identities actually acquire self-meaning; they are reflexive (Burke and Reitzes 1981). Others respond to a person in terms of his or her role identities. These responses, in turn, form the basis for developing a sense of self-meaning and self-definition.

Identity is the pivotal concept linking social structure with individual action; thus the prediction of behavior requires an analysis of the relationship between self and social structure. While society provides roles that are the basis of identity and self, the self is also an "active creator of social behavior" (Stryker 1980:385). Role identities, by definition, imply action" (Callero 1985:205). From an identity theory perspective, a role is a set of expectations prescribing behavior that is considered appropriate by others (Simon 1992). Satisfactory enactment of roles not only confirms and validates a person's status as a role member (Callero 1985) but also reflects positively on self-evaluation. The perception that one is enacting a role satisfactorily should enhance feelings of self-esteem, whereas perceptions of poor role performance may engender doubts about one's self-worth, and may even produce symptoms of psychological distress (Thoits 1991; also see Hoelter 1983; Stryker and Serpe 1982). Distress may arise if feedback from others—in the form of reflected appraisals or perceptions of the self suggested by others¹ behavior—is perceived to be incon-gruent with one's identity. According to Burke (1980, 1991; Burke and Reitzes 1991), identities act as cybernetic control systems: they bring into play a dissonance-reduction mechanism whereby people modify their behavior to achieve a match with their internalized identity standards. This process in turn reduces distress.

Identity theorists focus on the self-defining roles that people occupy in society, rather than on the wider range of different social attributes that can be ascribed to self. These latter attributes, which might include gender, race, ethnicity, and so forth, often function as master statuses (Stryker 1987) because in many contexts they override all other characteristics of the person. They are structurally based attributes that reflect the features of the social structure in which people's role identi-

ties are embedded, but because they do not carry specific sets of behavioral expectations (Thoits 1991) they are not separate components of self. Nevertheless, social attributes are considered to have an indirect impact on self through their effect on the role positions people can hold, the relative importance of their role identities, and the nature of their interactions with others. Identity theorists disagree, however, about how to treat social attributes.¹

Identity Salience

Identity theory links role identities to behavioral and affective outcomes, and acknowledges that some identities have more self-relevance than others. Role identities are organized hierarchically in the self-concept with regard to the probability that they will form the basis for action. Those positioned near the top of the hierarchy are more likely to be invoked in a particular situation, and hence are more self-defining than those near the bottom (McCall and Simmons 1978; Stryker 1968; Wiley 1991). Stryker (1987) argues that the notion of identity salience is distinguishable from other related microsocio-logical constructs, such as role-person merger (R.H. Turner 1978), psychological centrality (Rosenberg 1979), and identity prominence (McCall and Simmons 1978), because it is defined bchaviorally rather than psychologically. Identity salience is conceptualized (and operationalized) as the likelihood that the identity will be invoked in diverse situations. In contrast, other concepts (such as role-person merger) focus more strongly on the person's perception of the importance or significance of the identity relative to other identities (see Nuttbrock and Freudiger 1991).

The direct and explicit implication of this behavioral notion of identity salience is that identities positioned higher in the salience hierarchy are tied more closely to behavior. Thus people with the same role identities may behave differently in a given context because of differences in identity salience (e.g., Callero 1985; Thoits 1991). For example, one

Idetuity theorists disagree about how to treat social attributes, and therefore about whether they carry behavioral expectations. Foe example, Burke (e.g., 1991) treats them as identities, Thoits (1991) as influences on identities, and Stryker and Serpe (19S2) as social structural features that influence identity commitment.

258

SOCIAL PSYCHOLOGY QUARTERLY

person, may work on the weekend while another may spend time with the children, although both may have a "parent" role identity. The difference in behavior is due to differences in identity salience (cf. Serpe 1987). People also may enact role-congruent behavior even in situations that are not role-relevant: for example, people with a salient "parent" identity may, at work, engage inappropriately in behaviors related to their roles as parents (cf. Nuttbrock and Freudiger 1991). Although identity theory specifies clearly the hypothesis that salient identities engender role-congruent behavior, Stryker (1968) acknowledges that in some situations, contextual demands may be so strong that the choice of behavior will be determined solely by the nature of the situation rather than by identity salience.

As well as affecting behavior, salient identities have affective outcomes: their enactment should exert more influence than do identities lower in the hierarchy on a person's sense of self-meaning, feeling of self-worth, and level of psychological well-being (Callero 1985; Thoits 1991). This idea can be traced back to James's early view that role-congruent behaviors have self-evaluative implications which vary according to the relative importance of die different components of self. James writes:

I. who foe the time have staked my all on being a psychologist, am mortified if others know much more psychology than I. But I am contented to wallow in the grossest igaoraace of Greek. My deficiencies there give me no sense of personal humiliation at all. Had I "pretensions" to be a linguist, it would have been just the reverse [1890] 1950:309).

In addition to behavioral and affective outcomes, identity salience influences people's relationships, particularly their perceptions and evaluations, of others (Callero 1985; McCall and Simmons 1978). Although not extensively developed, one proposal is that salient identities are associated with positive evaluations of others who occupy the same role. Another, more fully explored proposal is that the number and importance of social relationships premised on a particular role identity may influence the salience of that identity. This idea is captured by the notion of commitment.

Commitment

Identity theory proposes that the salience of a particular identity will be determined by the person's commitment to that role. Commitment, defined as the "degree to which the individual's relationships to particular others are dependent on being a given kind of person" (Stryker and Stratham 1985:345), reflects the extent to which important significant others are judged to want the person to occupy a particular role position. Commitment to a particular role identity is high if people perceive that many of their important social relationships are predicated on occupancy of that role. The consequence of vacating such a role is loss of a social network that is psychologically important, for example, for the self-concept and for self-esteem (Hoelter 1983).

Stryker (1980) identified two types of com-mitmeat: 1) interactional commitment, reflecting the number of roles associated with a particular identity (the extensivity of commitment), and 2) affective commitment, referring to the importance of the relationships associated with the identity—in other words, the level of affect associated with the potential loss of these social relationships (the intensivity of commitment). The more strongly committed a person is to an identity—in terms of both interactional and affective commitment—the higher the level of identity salience will be. In terms of network relationships, the more fully a person's important social relationships are based on occupancy of a particular identity, in comparison with other identities, the more salient that identity will be. Similarly, the larger the number of persons included in such a set of social relationships, the more salient the identity (Stryker and Serpe 1982).

By acknowledging the impact of social networks on people's self-concepts, identity theory links the wider social structure (in terms of role positions) and the person's more intimate social networks (through levels of commitment to different role positions) to the self-concept, and also connects social structure to the development and maintenance of social relationships (Serpe 1987).

Summary

In summary, identity theory postulates that self reflects the wider social structure insofar as self is a collection of identities derived

A TALE OF TWO THEORIES

259

from the role positions occupied by the person. Society in the form of role positions provides a person with a sense of self-meaning and influences social behavior through these role-related components of self. Hence the impact of society on behavior is mediated by self-referent role identities. In an extension to this basic proposition, identity theory distinguishes among identities in terms of their hierarchical position in a person's structure of identities—a distinction that is used to account for variation in behavioral choice and that has implications for affective outcomes. The relative salience of different identities, in turn, is based on the number and strength of important social relationships that depend on occupancy of specific roles. Central characteristics of identity theory are that 1) it represents a social psychological model of self in that social factors are seen to define self; 2) the social nature of self is conceived as derived from the role positions that people occupy in the social world; 3) in an enduring sense, these role identities are proposed to vary in regard to their salience; and 4) although identity theorists acknowledge that reciprocal links exist between self and society, they have been most interested in individualistic outcomes of identity-related processes. The impact of role identities on relations with others has not been an important focus of the theory, and their influence on the broader social structure has not been spelled out clearly.

SOCIAL IDENTITY THEORY

Social identity theory is intended to be a social psychological theory of intergroup relations, group processes, and the social self. It has its origins in early work in Britain by Henri Tajfel on social factors in perception (e.g., Tajfel 1959, 1969a) and on cognitive and social belief aspects of racism, prejudice, and discrimination (e.g., Tajfel 1963, 1969b, 1970), but was developed and fully formulated in collaboration with John Turner and others in the mid- to late 1970s at the University of Bristol (e.g., Tajfel, 1974, 1978, 1982; Tajfel and Turner 1979; J.C. Turner 1982). During the 1980s significant theoretical and empirical advances were made as an increasing number of researchers, mainly in Europe but also in North America and Australia, came under its umbrella. Such popularity has quite naturally spawned

healthy controversy (cf. Abrams and Hogg 1990), but also has produced a number of books that document strong and continuing development (e.g., Hogg 1992; Hogg and Abrams 1988; Oakes, Haslam, and Turner 1994; Tajfel 1984; J.C. Turner 1991; Turner and Giles 1981; Turner et al. 1987). During the early to mid-1980s John Turner initiated an important theoretical development of social identity theory to produce self-categorization theory (J.C. Turner 1985; Turner et al. 1987). Although distinct from social identity theory in some respects, it is related closely enough to be considered as part of the same theoretical and metatheoreti-cal enterprise as social identity theory (cf. Hogg forthcoming; Hogg and McGarty 1990).

The development of social identity theory is intertwined with the development of a distinct European social psychology. Since the late 1960s European social psychologists have considered themselves to have a slightly different social and theoretical agenda than North American social psychologists (e.g., Jaspars 1980, 1986; Tajfel 1972, l984)-one that recognizes metatheoretical and conceptual limitations of theoretical reductionism and instead seeks theories that articulate individual psychological processes and wider social forces (cf. Doise 1986; Lorenzi-Cioldi and Doise 1990). These goals also frame social identity theory and its more recent extension into self-categorization theory. The regional distinction between Europe and North America, however, is now blurred (Moreland, Hogg, and Hains 1994).

Social Identity Theory

Social identity theory is specified in detail elsewhere (e.g., Hogg 1992, 1993; Hogg and Abrams 1988; Tajfel and Turner 1979; J.C. Turner 1982). The basic idea is that a social category (e.g., nationality, political affiliation, sports team) into which one falls, and to which one feels one belongs, provides a definition of who one is in terms of the defining characteristics of the category—a self-definition that is a part of the self-concept. People have a repertoire of such discrete category memberships that vary in relative overall importance in the self-concept. Each of these memberships is represented in the individual member's mind as a social identity that both describes and

260

SOCIAL PSYCHOLOGY QUARTERLY

prescribes one's attributes as a member of that group—that is, what one should think and feel, and how one should behave. Thus, when a specific social identity becomes the salient basis for self-regulation in a particular context, self-perception and conduct become in-group stereotypical and normative, perceptions of relevant out-group members become out-group stereotypical, and intergroup behavior acquires competitive and discriminatory properties to varying degrees depending on the nature of relations between the groups. Social identities are not only descriptive and prescriptive; they are also evaluative. They furnish an evaluation (generally widely shared or consensual) of a social category, and thus of its members, relative to other relevant social categories. Because social identities have these important self-evaluative consequences, groups and their members are strongly motivated to adopt behavioral strategies for achieving or maintaining in-group/ out-group comparisons that favor the in-group, and thus of course the self.

To account for social identity phenomena, social identity theory invokes the operation of two underlying sociocognitive processes. 1) Categorization sharpens intergroup boundaries by producing group-distinctive stereotypical and normative perceptions and actions, and assigns people, including self, to the contextually relevant category. Categorization is a basic cognitive process that operates on social and nonsocial stimuli alike to highlight and bring into focus those aspects of experience which are subjectively meaningful in a particular context (see "Self-Categorization Theory" below). 2) Self-enhancement guides the social categorization process such that in-group norms and stereotypes largely favor the in-group. It is assumed that people have a basic need to see themselves in a positive light in relation to relevant others (i.e., to have an evaluatively positive self-concept), and that self-enhancement can be achieved in groups by making comparisons between the in-group and relevant out-groups in ways that favor the in-group (but see Hogg and Abrams 1993). For example, comparisons can be made on stereotypical dimensions that favor the in-group rather than on those which are less flattering to the in-group.

An important feature of social identity theory is that in order to explain group members¹ behavior, it formally articulates

these basic sociocognitive processes of categorization and self-enhancement with subjec-tive belief structures. The latter refer to people's beliefs about the nature of relations between their own group and relevant out-groups. These beliefs (which are not necessarily accurate reflections of reality because they can be, and often are, ideological constructs) concern the stability and legitimacy of intergroup status relations and the possibility of social mobility (psychologically passing from, one group to another) or social change (psychologically changing the self-evaluative consequences of existing in-group membership). Subjective belief structures influence the specific behaviors that group members adopt in the pursuit of self-enhancement through evaluative positive social identity. For example, a group that believes its lower status position is relatively legitimate and stable but that it is quite possible to pass psychologically into the dominant group (i.e., acquire a social identity as a member of the higher-status group) will be unlikely to show much solidarity or engage in much direct intergroup competition. Instead members will attempt, as individuals, to disidentify and gain psychological entry to the dominant group. In contrast, a group that believes its lower status position is illegitimate and unstable, that passing is not viable, and that a different social order is achievable will show marked solidarity and will engage in direct intergroup competition.

Self-Categorization Theory

Self-categorization theory (J.C. Turner 1985; Turner et al. 1987; also see Oakes et al. 1994; J.C. Turner 1991) is a recent development that elaborates in detail the operation of the categorization process as the cognitive basis of group behavior. The process of categorization accentuates both perceived similarities between stimuli (physical objects or people, including self) belonging to the same category and perceived differences between stimuli belonging to different categories. This accentuation effect occurs on dimensions that the categorizer believes are correlated with the categorization. Thus, for example, when feminists who believe that men are more aggressive than women categorize themselves as feminists, they will tend to exaggerate men's aggressiveness, to sec all men as more aggressive than all women, to

A TALE OF TWO THEORIES

261

see little difference in aggressiveness among men, and to see little difference in nonaggres-siveness among women (including self). The categorization-accentuation process as a whole serves an important function for the individual. It highlights incergroup discontinuities, ultimately renders experience of the world subjectively meaningful, and identifies those aspects which are relevant to action in a particular context.

Categorization of self and others into in-group and out-group defines people's social identity and accentuates their perceived similarity to people's cognitive representation of the defining features of the group (i.e., their group prototypicality, or normative-ness). People are essentially "depersonalized": they are perceived as, are reacted to, and act as embodiments of the relevant id-group prototype rather than as unique individuals. Depersonalization of self is the basic process underlying group phenomena— for example, social stereotyping, group cohesion and ethnocentrism, cooperation and altruism, emotional contagion and empathy, collective behavior, shared norms, and the mutual influence process. It has none of the negative implications of terms such as "dehumanization" or "deindividuation"; it simply refers to a contextual change in the level of identity (from unique individual to group member), not to a loss of identity. Through depersooalization, self-categorization effectively brings self-perception and behavior into line with the contextually relevant in-group prototype, and thus transforms individuals into group members and individuality into group behavior.

According to self-categorization theory, people cognitively represent social groups in (emu of prototypes. A prototype is a subjective representation of the defining attributes (e.g., beliefs, attitudes, behaviors) of a social category, which is actively constructed from relevant social information in the immediate or more enduring interactive context (cf. Fiske and Taylor 1991). Because members of the same group generally find themselves placed relatively similarly in the same social field (i.e., they are exposed to similar information from the same perspective), their prototypes usually are very similar—that is, shared. Prototypes ordinarily are unlikely to be checklists of attributes (though of course they can be elicited in this form by probing). Rather, they are fuzzy sets

that capture the context-dependent features of group membership, often in the form of representations of exemplary members (actual group members who embody the group) most fully or ideal types (a relatively nebulous abstraction of group features). People can assess the prototypicality of real group members, including self—that is, the extent to which a member is perceived to be close or similar to the group prototype.

Because group prototypes define groups as distinct entities, they are constructed as a dynamic balance between competing cognitive pulls со minimize intracategory differences and to maximize intercategory differences—a process governed by the principle of metacontrast. For this reason, prototypes are influenced strongly by what out-group is salient. Therefore relatively enduring changes in prototypes and thus in self-conception can occur if the relevant comparison out-group changes over time—for instance, if Catholics gradually come to define themselves in contradistinction to Muslims rather than to Protestants. Such changes are also very transitory insofar as they are tied to whatever out-group is salient in the immediate social context. Thus social identity is highly dynamic: it is responsive, in both type and content, to intergroup dimensions of immediate social comparative contexts.

This responsiveness of social identity to immediate social contexts is a central feature of social identity and self-categorization theory. The cognitive system, in seeking to maximize meaning in a specific context, engages whatever categorization is cognitively most readily available and best explains or fits the similarities and differences among people. For example, we might initially "try on" the readily available categorization of "man/woman" to make sense of a particular social context (e.g., what people are doing, saying, wearing). The category of "man" or of "woman," however, would not become fully activated as the basis of self-categorization and depersonalization. unless it made adequate sense of relevant similarities and differences (i.e., fit the data well). Once fully activated on the basis of perceived similarities and differences among stimuli, categories organize themselves around contextually relevant prototypes and are used as a basis for the perceptual accentuation of intragroup similarities and intergroup differences, thereby maximizing separateness and clarity. Self-categorization in terms of the activated in-group category then depersonalizes behavior in terms of the in-group prototype.

The subjective salience of social categories is governed not only by the mechanics of stimulus-category fit, but also by the motivated availability of social categories. That is, people engage actively in more or less competitive (and mote or less successful) renegotiation of the frame of reference in order to achieve a self-categorization that is more favorable for conceptualization of self in that context. For instance, a nontraditional male at a feminist meeting might try to avoid the contextually negative implications of self-categorization as male by drawing attention to contextually less negative self-categorizations.

262

SOCIAL PSYCHOLOGY QUARTERLY

SOME DIFFERENCES

Conceptually more interesting are the differences between identity and social identity theory. In general it is not appropriate to castigate theories for failing to do what they did not set out to do in the first place; a theory ought to be assessed against its self-proclaimed explanatory scope. Therefore in this respect it would be easier to justify an

The social identity and self-categorization models of group processes have a number of important features: 1) they are general theories of the social group, not constrained by group size, dispersion, and so forth; 2) they incorporate the role of both the immediate and the more enduring intergroup context in group behavior-, 3) they account for the range of group behaviors (e.g., conformity, stereotyping, discrimination, ethnocentrism) in terms of a limited number of theoretically integrated generative principles; 4) they are basically sociocognitive; and 5) they do not construct group processes from interpersonal processes. The process of self-categorization depersonalizes perception, feelings, and action in terms of the contextually relevant , self-defining in-group prototype. Behavior -thus is influenced by the categorical structure of society via the mediation of social identity and the accompanying process of self-categorization. The contextual salience of specific social identities rests on the extent to which they render maximally meaningful a particular context, and contextual factors influence the form taken by identity-contingent cognitions and behaviors. Because social identities are attached to value, a complex social dynamic exists in which groups vie for relatively positive social identity. Intergroup relations and social identity thus are dynamically intertwined.

SOME SIMILARITIES

Because identity and social identity theory are isolated scientifically from one another, ithas been necessary to provide a somewhat detailed overview of both perspectives. This review enables us to identify some of the principal similarities and differences between them—similarities and differences which must be understood in their wider disciplinary context as reflecting the fact that identity theory is ultimately a sociological theory and social identity theory a psychological theory.

Both theories address the structure and function of the socially constructed self (called identity or social identity) as a dynamic construct that mediates the relation- ship between social structure or society and individual social behavior.

Reciprocal links between society and self are acknowledged by both theories. Behavior is considered to be organized into meaningful units that are subsumed by specific self-definitions: identity theory discusses the organization of behavior in terms of roles, while social identity theory talks of norms, stereotypes, and prototypes. Just as behavior is organized into discontinuous clusters, the self is structured into discrete identities that are interrelated in various important ways.

Both theories also discuss the way in which identities are internalized and used to define self: social identity theory speaks of social identification and the process of self-categorization, while identity theory discusses the process of labeling or naming oneself as a member of a social category, or of commitment. '

One reason for these similarities may be that social identity theory is relatively distinct among rccent social psychqlogical theories, in ways that make it more comparable to sociological theories. Contemporary social psychological theories tend to focus only on

intrapsychic processes and interpersonal relations, while social identity theory attempts to explain group behavior in terms of concepts
that articulate societal and psychological processes and that recognize the primacy of society over individual.

A TALE OF TWO THEORIES

263

attempt to criticize social identity and self-categorization theories for failing to meet their goal of articulating psychological and social factors in the explanation of group processes than to justify a criticism of identity theory for failing to expand on the sociocog-nitive processes underpinning identity. We believe, however, that the coexistence of two such similar frameworks warrants some comment on the extent to which these theories can articulate society and individual and describe generative processes. We compare the theories from the standpoint of a social identity theorist; an identity theorist's standpoint might be expected to raise different issues or to place a different interpretation on issues.³

Level of Analysis

One of the most important sources of differences is the fact that identity theory is not essentially a psychological theory, and therefore does not place much emphasis on describing generative cognitive processes. In (his respect, social identity theory, as a psychological theory, may have some advantages over identity theory—advantages that stem from its more detailed specification of sociocognitive processes.

For instance, identity theory focuses on the process of labeling oneself as belonging to a particular social category, acknowledges the role that others may play in supporting this categorization, and relates self-conception to behavior via behavioral prescriptions embodied by roles. Yet it generally stops short of specifying in any detail the cognitive processes and structures (eg., categorization, prototypes) that may underlie identity dynamics and may produce conformity to norms. Burke (1991; Burke and Reitzes 1991), however, has described a dissonance-reduction process in which the self, as a cybernetic control mechanism, is motivated to bring

¹ Alternative interpretations might include the following: 1) Symbolic interactionisra place* "taking the role of the other" at center stage at a cognitive process; (bus identity theory, which is based oo. symbolic interaction-isra. does explore, cognitive processes. 2) Some identity theory perspectives (e.g., McCall and Simmons 1978) are explicit about the influence of situation*! factors on identity enactment*, therefore identity theory is situation-ally dynamic. 3) Depending on how rales are defined, general social attributes are viewed by some identity theorists as carrying behavioral expectations (e.g., Burke 1991).

self-conception into line with reflected appraisals (perceptions of self suggested by others' behavior) by modifying own. behavior. People behave in ways that are consistent with their role identities as a consequence of reducing or avoiding incongruency between internalized identity standards and others' perceptions of self. With the exception of this proposed mechanism, sociocognitive mechanisms do not occupy a central role in identity theory. Although the original symbolic inter-actionist emphasis on "taking the role of the other" actually invites a sociocognitive analysis, such as that proposed by Burke, role-taking processes are largely not examined empirically or elaborated by identity theorists, but rather are assumed. In contrast, such processes and structures form the theoretical and empirical core of social identity theory, particularly self-categorization theory, which specifies in detail a social psychological process that links identity to behavior via depersonalization and conformity.

Stemming from differences in emphasis on, and type of, sociocognitive process, identity theory only hints at the possibility that people may favorably evaluate others who have the same role identities as themselves and that this favorable evaluation may be stronger as a function of identity salience. This idea is explored more fully by social identity/self-categorization theory through the notion of depersonalized social attraction (Hogg 1992, 1993). Finally, identity theory's lesser emphasis on generative sociocognitive process may also be partly responsible for its tendency to underplay the role of the immediate context and instead to attribute identity changes to changes in role position (see below). Social identity theory, in contrast, has a somewhat more dynamic and more highly elaborated perspective, which explains contextual salience in terms of social comparative factors, self-esteem motivation, uncertainty reduction, and social explanation. This approach may be able to account more fully for the responsiveness of social behavior to the immediate context.

We believe that one of the strengths of social identity/self-categorization theory, among social psychological theories, is that it tries systematically to articulate (cf. Doise 1986; Lorenzi-Cioldi and Doise 1990) the psychological level of analysis (sociocognitive processes) with the "sociological" level (socio-

264

SOCIAL PSYCHOLOGY QUARTERLY

historical dimensions of intergroup relations). Id this respect, social identity is a social construct that mediates individual and society. Practically, however, researchers tend to put their conceptual energy into psychological, principally cognitive, factors; they lean, if anything, toward the psychologization of behavior. Although a great deal of detail is provided on self-categorization and depersonalization, currently there is less work on how social structural variables or social belief structures really enter the picture (but cf. Taylor and Moghad-dam 1987; van Knippenberg and Ellemers 1993). Some critics have been led to ask to what extent the theory in fact meets its meta-theoretical objectives (e.g., Condor 1990; Wetherdl and Potter 1992). Identity theory does not confront this problem because it does not rest so explicitly on sociocognitive processes. As a more sociological perspective, it perceives a direct reciprocal link between individual and society mediated by the social construct of role identity. In doing so it does not reduce the social to the individual; nor, on the other hand, does it fully elaborate intervening sociocognitive mechanisms. Can social identity theory perhaps help provide the missing sociocognitive dimension, and identity theory help keep social identity theory away from the jagged rocks of psychological reductionism?

Intergroup Behavior

Another important source of differences between the theories is that social identity theory is about intergroup relations and group behavior, while identity theory concerns role behavior. Identity theory thus is focused differently than social identity theory. It concentrates on role behavior and role identities, and does not consider in any direct sense the impact of other social attributes on self. These "other attributes" are mainly large-scale category memberships such as ethnicity, sex, race, and nationality. For social identity theory these are the most significant sources of social identity; social identity dynamics are contextualized by the social relations between such categories.

Social identity theory places emphasis on intergroup relations and thus on the role played by out-groups; identity theory does not. Instead identity theory addresses counter-roles (e.g., father-daughter), which are not necessarily the same thing as out-groups (i.e.,

Burke and Tully 1977). In addition, courtter-i roles are considered relevant only insofar as they help to clarify the meanings of role identities. Identity theory places little emphasis on the impact of people's identities on their relations with out-group others, to contrast, social identity theory triesto specify the effects of salient social identity on people's perceptions of and conduct toward others, particularly out-group others. Social identity theory therefore goes further than identity theory. Not only does it explicate a person's individual. behavioral choices, as does identity theory; it also explicates people's relations with out-group others and consequently allows some understanding of intergroup behavior. Building on a characterization of society as hierarchically structured in terms of relations between (large-scale) social categories, social identity theory is actually able to specify how a person's position in the social world (mediated by self) affects social behavior. (From a sociological perspective, however, research has focused mainly on individual and group rather than on individual and society.)

Paradoxically, then, the more psychological perspective of social identity theory, because of its intergroup analysis, may come closer to achieving the more sociological goal of identity theory to address the dynamic impact of society on self. The "psychological" perspective not only has more to say about underlying psychological processes, but also may have more explanatory utility in regard to truly social outcomes. In contrast, the "sociological" perspective seems to be concerned more strongly with individualistic outcomes of identity such as role behavior and, recently, with affective outcomes such as psychological well-being.

Roles and Groups

Social identity theory does not explicitly discuss roles, though it would probably consider roles to refer to positions in a given group (e.g., leader, comic, bureaucrat; cf. Hogg 1995, forthcoming). From this perspective one might argue that roles provide a sense of distinct individual identity within a group, perhaps satisfying a need for intra-group differentiation (cf. Brewer's (1991, 1993] notion of optimal distinctiveness) or even a need for personal identity, but that they do not provide a social identity in the

A TALE OF TWO THEORIES

265

strict sense of the term. This idea contrasts quite sharply with identity theory, which considers self-definition to derive principally from roles, via role identities, rather than from the broad range of wider social attributes that social identity theory considers to be the basis of social identity. Social identity theory therefore permits a conceptual differentiation between roles (differential behavioral prescriptions within a group) and identity based on group membership. In contrast, identity theory's notion of roles has many properties of both group membership and differential behavioral prescriptions within a group; in this sense, group membership and roles may not be distinguished from one another.

Social Context and Identity Salience

Finally, we believe, the two theories differ in how contextually responsive and how dynamic they consider (he self-concept to be. Both formulations consider the self to be structured into relatively discrete identities, but identity theory, particularly Stryker's formulation, regards this structure as relatively stable, changing primarily in response to changes in role positions (e.g., Serpe 1987). Others, such as McCall and Simmons (1978) and Burke (e.g., 1980, 1991), view identities as more responsive to context. Roles themselves, however, are dynamically constructed and reconstructed through interpersonal interaction. The chronic relative salience of identities within the self-concept is considered to be relatively stable; except in rare circumstances, the chronic salience of a person's identity determines his or her behavioral responses. For instance, the impact of identities on affective outcomes directly reflects the chronic salience of a person's identity. Identity theory acknowledges that sttuational factors may be important (e.g., McCall and Simmons 1987), certainly in construction and reconstruction of roles, but places less emphasis than does social identity theory on elaboration of sociocognitive processes that cause self to be highly responsive to immediate contextual cues. Burke (1991), however, suggests ways in which a cybernetic model of identity can explain the "rare" occasions when perceived incongruence produces identity change rather than behavior change.

In contrast, although social identity theory views social identity as an enduring construct

that changes with changing intergroup relations, it also places at center stage the view that the content of social identity is dynamically responsive to immediate contextual factors: different contexts may prescribe different contextually relevant behaviors contingent on the same social identity. Being Australian in the United States, for instance, can vary in chronic importance from person to person, and the meaning and behavioral prescriptions of this identity can vary as a function of changing intergroup relations between Australia and the United States. Furthermore, immediate contextual factors (the situation and the interactants) will influence what aspect of Australian identity is prescribed; a colloquium presentation and a cocktail party might elicit very different "Australian" behaviors. This, we believe, is a more dynamic treatment of the relationship between self and identity (on the one hand) and immediate social context (on me other) than is offered by identity theory. In addition, self-categorization theory, because of its more highly elaborated cognitive emphasis, explores in greater detail than identity theory the sociocognitive generative mechanisms associated with transitory identity salience.

Identity theory, however, goes further than social identity theory in describing the conditions under which particular identities -will be "chronically" salient, and perhaps has gone further toward theoretically and empirically considering the impact of chronic levels of identity salience. It also places greater emphasis on analysis of interpersonal social interaction as an influence on enaction and modification of roles, and thus on identity dynamics.

CONCLUSIONS AND PROSPECTS

The aim of this article has been to compare identity theory with social identity/self-categorizatton theory as two perspectives on the dynamic mediation between individual social behavior and society (or social structure) of the socially constructed self. Identity theory originates in the discipline of sociol-ogy, and deals with the structure and function of people's identity as related to the behavioral roles they play in society. Social identity/self-categorization theory originates in the discipline of psychology, and deals with the structure and function of identity as related to people's membership in groups.

266

SOCIAL PSYCHOLOGY QUARTERLY

These two perspectives are remarkably similar; yet, to our knowledge, no attempt has been made to formally contrast and compare them. Such a comparison is long overdue as the first step in a debate on conceptual integration, demarcation, and differentiation of the two perspectives.

Both theories consider social behavior to be structured into meaningful units that are subsumed by specific self-definitions (identities), which themselves are interrelated parts of a differentiated and structured self-concept. Emphasis is placed on explicating the processes responsible for internalizing identities and for making different identities the salient bases for self-conception and conduct in particular contexts. Against the background of these broad conceptual and metatheoretical similarities, from the standpoint of a social identity theorist, a number of significant differences exist.

First, identity theory is a perspective on the relationship between the roles people play in society and the identities that such roles confer. The focus is on individual behavior as it is mediated by role identities. In contrast, social identity theory concerns intergroup relations and group processes, with a focus on the generative role of identity in group and intergroup aspects of behavior (e.g., conformity, collective action, stereotyping, group solidarity, ethnocentrism). We believe that because of this difference in emphasis, social identity theory may be better placed to link individual social behavior to dynamic features of social structure.

Second, social identity theory, particularly its recent extension into self-categorization theory, goes further than identity theory in elaborating the sociocognitive generative processes that underlie the operation of identity. This may be an advantage that allows social identity theory to specify in greater detail than identity theory how identities are internalized, how contextual factors nuke different identities salient, and how identities produce identity-consistent behavior. It has been suggested, however, that social identity theory, especially its recent extension into self-categorization theory, may have become too strongly concerned with cognitive processes alone. This problem does not apply to identity theory, which has the advantage of focusing more explicitly on interindividual social interaction as an influence on identity.

Third, identity theory concerns behavioral roles and role identities rather than broader

social category membership; the opposite is true of social identity theory. Fourth, social identity theory views identity as a dynamic construct that responds to changes in both long-term intergroup relations and immediate interactive contexts, and elaborates the underlying sociocognitive mechanism. Identity theory tends more to view identity as a relatively static property of roles, and focuses on the dynamics of interpersonal social interactive contexts that influence the construction and reconstruction of roles.

Generally it is inadvisable to attempt to integrate very different theories (cf. Billig's [1976] critical analysis of efforts to integrate Marxist with Freudian perspectives on the explanation of prejudice). Often it is preferable to pit one theory against the other in an empirical or conceptual attempt to establish which is better. One way in which the two theories discussed here could be pitted against one another might be through research into underlying sociocognitive processes. Contrasting predictions could be examined empirically to compare the self-categorization and social comparison processes specified by social identity theory and self-categorization theory with the cybernetic mechanism suggested by Burke (1991; Burke and Reitzes 1991)- Another approach might be through research into the predictive utility of intergroup analyses. Contrasting predictions could be examined empirically to compare the intergroup analysis specified by social identity theory with the role analysis suggested by identity theory.

Identity theory and social identity theory differ, we believe, in the degree and type of contextual, responsiveness that they assign to identity. It would be worthwhile to devise empirical tests in which the more static conception of identities, as envisaged by identity theory, could be pitted against the more contextually responsive conceptualization of social identity/self-categorizacion theory (cf. Oakes et al. 1994). Social identity theory, however, could benefit from consideration of identity theory's more detailed specification of the dynamics of chronically salient identities, and its fuller attention to interpersonal social interactive factors. Finally, identity theory links self-attitude (identity) to behavior fairly automatically via the notion of roles, while hardly specifying how this happens, whereas social identity theory specifies quite exactly the processes that link self-attitude (identity) with normative behavior. Per-

A TALE OF TWO THEORIES

267

haps the two theories can be coordinated to help explain the general relationship between attitudes and overt behavior (cf. Hogg forthcoming; Terry and Hogg forthcoming)

We have tried here to show that identity theory is useful in its own domain; it has particular strengths in its analysis of the impact of chronic identities on (mostly individualistic) outcomes, and in its emphasis on interpersonal social interactive contexts. Social identity theory is also useful in its own domain; it is particularly strong in its elaboration of sociocognitive processes and in its emphasis on intergroup relations. These strengths, we suggest, may allow one to actually link society with individual social behavior more effectively. It may be possible in some way to integrate or articulate identity theory with social identity theory. For example, it may be possible to locate the concept of role and identity theory's attendant analysis of interpersonal social interaction within social identity theory's broader intergroup analysis and its more fully elaborated sociocognitive analysis. In the first instance, however, it may be more useful to explore, conceptually and empirically, the difference between role identities and social identities. What are the differences between identities that arise from behavioral roles within groups, identities that arise from group membership, and identities that arise from membership in large-scale social categories?

REFERENCES

Abrams, Domioic and Michael A. Hogg. 1990. Social Identity Theory: Constructive and Critical Advances. London: Harvester/Wheatsheaf.

Billig, Michael. 1976- Social Psychology and Intergroup Relations. Loadon: Academic Press.

Blumer, Herbert. 1969- Symbolic Interaction: Perspective and Method. Englewood Cliffs, NJ: Prentice Hall.

Brewer, Marilynn. 1991. "The Social. Self: On Being the Same and Different ac the Same Time." Personality and Social Psychology Bulletin 17:475-82.

____ 1993. "The Role of Distinctiveness in Social

Identity and Qroup Behaviour." Pp. 1-16 in Group Motivation: Social Psychological Perspectives, edited by Michael A. Hogg and Dominic Abrams. London: Harvester/Wheatsheaf.

Burke, Peter. I. 1980. "The Self: Measurement Requirements from an Interactionist Perspective." Social Psychology Quarterly 43:18-29.

____ 1991. "Ideotily Processes and Social Stress."

American Sociological Review 56: 836-49.

Burke, Peter J. and Donald С Rcitaes. 1981. "The Link between Identity and Role Performance." Social Ptychology Quarterly 44:83-92.

____ 1991. "An Identity Approach to Commitment."

Social Psychology Quarterly 54:280-86.

Burke, Peter J. and Judy Tully. 1977. "The Measurement of Role/Identity." Social Forces 55:880-97.

Callero, Peter. 1985. "Role-Identity Salience." Social Psychology Quarterly 48:203-15.

CaUero, Peter L., Judith A. Howard, and Jane A. Pilavin. 1987. "Helping Behavior as Role Behavior. Disclosing Social Structure and History in the Analysis of Prcsocial Action." Social Psychology Quarterly 50:247-56.

Condor, Susan. 1990. "Social Stereotypes and Social Identity. " Pp 230-49 in Social Identity Theory: Constructive and Critical Advances, edited by Dominic Abrams and Michael A. Hogg. London: Harvester/Wheatsheaf.

Cooley, Ovaries. 1902. Human Nature and Social Order. New York: Scribners

Dcise, Willem. 1986. Levels of Explanation, in Social Psy-Siology. Cambridge, UK: Cambridge University Press.

Ftske, Susan T. and Shelley E. Taylor. 1991. Social Cognition. 2nd ed. New York: McGraw-Hill.

Koelter, John W. 1983. "The Effects of Role Evaluation and Commitment on Identity Salience." Social Psy-chahgy Quarterly 46:140-47.

Hogg, Michael A. 1992. The Social Psychology of Group Coltesiveness: From Attraction to Social Identity. London: Harvestec/Wheatsheaf.

. 1993. "Group Cohesiveness: A Critical Review and Some New Directions." European Review of Social Psychology 4:85-111.

.. Forthcoming. "Social Identity, Self-Categoriza-

tion and the Small Group." In Understanding Group Behavior, Vol. 2: Small Group Processes and Interpersonal Relations, edited by James Davis and Eric Witte. Hillsdale, NJ: Erlbaum.

Forthcoming. "Intragroup Processes, Group

Structure and Social Identity." In Social Identity: The Developing Legacy of Henri Tajfel, edited by W. Peter Robinsion. Oxford: Butterworth-Heinemann.

Hogg, Michael and Dominic Abrams. 1988. Social Identifications: A Social Psychology of Intergroup Relations and Group Processes. London: Rout ledge.

.____ 1993 "Towards a Single-Process Uncertainty-
Reduction Model of Social Motivation in Groups." Pp.
173-90 in Group Motivation: Social Psychological
Perspectives, edited by M.A. Hogg and D. Abrams.
J-ondon: Harvestec/Wheatsheaf.

Hogg, Michael A. and Craig McGarty. 1990. "Self-Categorization and Social Identity." Pp. 10-27 in Social Identity Theory: Constructive and Critical Advances, edited by D. Abrams and M.A. Hogg. London: Harvester Wheatsheaf.

James William [1890) 1950. The Principles of Psychology. New York: Dover.

Jaspars, Joseph M.F. 1980. "The Coming of Age of Social Psychology in Europe." European Journal of Social Psychology 10:421-29.

____ 19S6. "'Forum and Focus: A Personal View of

European Social Psychology." European Journal of Social Psychology 16:3—15

Lindesmith. Alfred and A. Strauss 1956. Social Psychology. New York: Holt, Rinehart and Winston.

Lotenzi-Cioldi. Fabio and Willem Doise. 1990. "Levels of Analysis and Social Identity." Pp. 71-88 in Social Identity Theory: Constructive and Critical Advances, edited by Dominic Abrams and Michael A. Hogg. London: Harvester/Wheatsheaf.

McCall, George and Jerry Simmons. 1978. Identities and Interactions. Revised ed. New York: Free Press.

Mead, George. 1934. Mind, Self, and Society: From the Standpoint of a Social Behaviorist, edited and with an

г

268

SOCIAL PSYCHOLOGY QUARTERLY

introduction by C.W. Morris Chicago: University of Chicago Press.

Morcland, Richard L., Michael A, Hogg, and Sarah С Halns. 1994. "Back ю the Future: Social Psychological Research on Groups." Journal of Experimental Social Psychology 30:527-55.

Nuttbrock, Larry and Patricia Freudigcr. 1991 "Identity Salience and Motherhood: Л Test of Strykcr's Theory." Social Psychology Quarterly 54:146-57.

Oakes, Penelope J., S Alexander Haslam, John С Turner. 1994. Stereotyping and Social Realm. Oxford: BlaclcweU.

PiUvin, Jane A., Dorcas Evans and Peter Callero. 19S4. "Learning to Give to 'Unnamed Strangers': The Process of Commitment U Regular Blood Donation." Pp. 471-92 in Tkt Development and Maintenance of Protocial Behavior: International Perspectives, edited by B. Suub, Daniel Ваг-Tel, Jerzy Karylowski, and Januaz Reyjowski. New York: Plenum.

Rosenberg, Morris. 1979. Conceiving the Self. New York: Basic Books.

_____ 1981. "The Self-Concept: Social Product and

Social Force." Pp. 562-92 in Social Psychology: Sociological Perspectives, edited by Morris Rosenberg and Ralph H. Turner. New York: Basic Books

Serpe, Richard T. 1987. 'Stability and Change in Self: A Structural Symbolic Inteactiontst Explanation." Social Psychology Quarterly 50:40-55.

Simon, Robin W. 1992. '"Parental Role Strains, Salience of Parental Identity, and Gender Differences in Psychological Distress." Journal of Health and Social Behavior 33:25-35.

Stryket, Sheldon. 1968. "Identity Salience and Rote Performance: The Importance of Symbolic Interaction Theory for Family Research." Journal of Marriage and the Family 30:558-64

____ 1980. Symbolic Interactianism: A Social Struc
tural Version. Palo Alto: Benjamin/Cummings.

1987. "Identity Theory: Developments and

Extensions." Pp. 89-104 in Self and Identity, edited by K. Yardley and T. Hooess. New York: Wiley

Stryker, Sheldon and Richard T. Serpe. 1982. - Commitment, Identity Salience, and Role Behavior " Pp. 199-218 in Personality, Roles, and Social Behavior, edited by W. Ickes and E.S. Knowlcs. New York: Springer-Veriag.

Stryker, Sheldon and Anne Stathsm. 1985. "Symbolic Interaction and Role Theory." Pp. 311-78 in The Handbook of Social Psychology, 3rd ed., edited by G. Lindzey &. E. Arooson. New York: Random House.

Tajfel, Henri. 1959. "Quantitative Judgment in Social Perception." British Journal of Psychology 50:16-29.

____ 1963. "Stereotypes." Race 5:3-14.

____ 1969a. "Social and Cultural Factors in Percep
tion." Pp. 315-94 in Handbook of Social Psyclwlogy,
Vol. 3, edited by G. Undzey and E. Aronson.
Reading, MA: Addisoo-Wesley.

____ 1969b. "Cognitive Aspects of Prejudice."

Journal of Social Issues 25:79-97.

_____ 1970 "Experiments in Iniergroup Discrimina
tion." Scientific American 223:96-102

1972. "Some Developments In European Social

Psychology." European Journal of Social Psychology 2:307-22.

1974. "Social Identity and Intcrgroup Behav-

iour " Social Science Information 13:65—93.
____ , ed 1978. Differentiation between Social

Groups. London: Academic Press.

., ed. 1982. Social Identity and Iniergroup Rela-

tions, Cambridge, UK: Cambridge University Press. ., ed. 1984. 77m Social Dimension: European

Developments in Social Psychology. Cambridge, UK: Cambridge University Press.

Tajfel, Henri and John С Turner. 1979. "An Integrative Theory of Intergroup Conflict." Pp. 33-47 in The Social Psychology of Intergroup Relations, edited by W.G. Austn andS. Worchel. Monterey: Brooke-Cole.

Taylor, Donald M. and Fathali Moghaddara. 1987. Theories of Intergroup Relations. New York: Praeger.

Terry, Deborah J. and Michael Hogg. Forthcoming. "Group Nonns and the Attitude-Behavior Relationship: A Role for Group Identification." Personality and Social Psychology Bulletin.

Thoits, Peggy A. 1991. "'On Merging Identity Theory and Stress Research." Social Psychology Quarterly 54:101-12

Turner, John С 1982. "Towards a Cognitive Redefinition, of the Social Group." Pp. 15-40 in. Social Identity and Intergroup Relations, edited by H. Tajfel. Cambridge, UK: Cambridge University Press.

____ 1985. "Social Categorization and the Self-
Concept: A Social Cognitive Theory of Group
Behaviour." Pp. 77-122 in Advances in Group
Processes Tlieory and Research, Vol. 2, edited by
E.J. Lawkr. Greenwich, CT: JAI.

- 1991 Social Influence. Milton Keyo.es: Open

University Press.

Turner, John C. and H Giles, eds. 1981. Intergroup Behaviour. Oxford: Blackwell.

Turner, John С , Michael A Hogg, P.J. Oakes, S.D. Reicher, and M.S. Wetherell. 1987. Rediscovering the Social Group: A Self-Categorization Theory. Oxford', Blackwell.

Turner, Ralph H. 1978. "The Role and the Person." American Journal of Sociology 84:1-23.

van Knippenberg, Ad and Naomi Elleraers. 1993. "Strategies in Intergroup Relations." Pp. 17-23 in Group Motivation: Social Psychological Perspectives, edited by Michael A. Hogg and Dominic Abracns. London: Harvester/Wheatsheaf.

Wetherell, Margaret S. and Jonathan Potter. 1992. Mapping the Language of Racism. London: Harvester/ Wheatsheaf.

White, Clovis L and Peter J. Burke. 1987 "Ethnic Role Identity among Black and White College Students." Sociological Perspectives 30:310-33.

Wiley. Mary G. 1991. "Gender, Work, and Stress: The Potential Impact of Role-Identity Salience and Commitment." Sociological Quarterly 32:495-510.

Michael A. Hogg is Reader in Social Psychology and Director of the Centre for the Study of Group Processes at //w University of Queensland. His research interests are group processes and intergroup relations from the perspective of social identity theory and self-categorization theory. He is currently researching group cohesion, structure, motivation and attitudes, and is revising his 1988 social identity text Social Identifications.

A TALE OF TWO THEORIES 269

Deborah J. Terry is Senior Lecturer in Psychology at the University of Queensland. Her research interests are attitude-behavior relations and attitude change.. She is currently researching the effects of group membership and mode of behavioral decision-making on attitude phenomena

Katherine M. White is pursuing her doctoral studies at the Universin of Queensland. Her thesis explores the role of group membership in attitude-behavior relations, and is pan of a wider interest in attitudinal and normative phenomena.