European Journal of Social Psychology Eur. J. Soc. Psychol. 28, 227±248 (1998)

Exploring temporal aspects of social identity: the concept of possible social identities MARCO CINNIRELLA* Department of Psychology, Royal Holloway, University of London, U.K. Abstract In the social identity model of reactions to negative social identity (Tajfel & Turner, 1979, 1986), the concept of cognitive alternatives focuses on individual and group perceptions of the possibility of changing group memberships or improving existing ones. In the current paper, the under-researched concept of cognitive alternatives is expanded so as to better encompass issues relating to the temporal dimension of social identity maintenance. Markus and Nurius' (1986) possible selves perspective is used as a starting point for exploring the manner in which social identity maintenance is in¯uenced by cognitions about, and social representations of, a group's past and possible future. It is proposed that the concept of cognitive alternatives be expanded to incorporate possible social identities, which represent individual and shared cognitions about possible past group memberships, possible future group memberships, and perceptions of the possible past and future for current group memberships. The consequences of perceiving positive and negative possible social identities are examined, and methodological issues which might facilitate their empirical study addressed. # 1998 John Wiley & Sons, Ltd.

INTRODUCTION There remain certain social phenomena which highlight gaps in our current knowledge of social identity processes. Ultra-right survivalist movements in North America, the con¯ict in the former Yugoslavia, and moves toward further European integration all, for example, highlight the need for a theory of social identity which adequately encompasses the temporal nature of identity maintenance and the quest *Address for correspondence: M. Cinnirella, Department of Psychology, Royal Holloway, University of London, Egham Hill, Egham, Surrey, TW20 0EX, U.K. Tel: (0)1784 443531. Fax: (0)1784 434347. e-mail: [email protected]. The author wishes to thank Chris Brewin, Miles Hewstone, Russell Spears, Jerzy Trzebinski and two anonymous reviewers for useful advice concerning the current manuscript.

CCC 0046±2772/98/020227±22$17.50 # 1998 John Wiley & Sons, Ltd.

Received 3 July 1996 Accepted 20 May 1997

228

M. Cinnirella

for coherence amongst past, present, and future identities (see Breakwell, 1986). In this paper, insights derived from the possible selves tradition of research (Markus & Nurius, 1984) are used to further elaborate and expand social identity theory's concept of cognitive alternatives (Tajfel & Turner, 1979, 1986). In doing so, it is argued that this carries with it the bene®t of expanding social identity theory's treatment of temporal aspects of social identity construction. Out of this review is identi®ed the need for a concept of possible social identities. This concept is described, its utility assessed, and possible methodological strategies for exploring it examined. During the course of the discussion, key theoretical arguments are set out in the form of hypotheses which are open to empirical exploration in the future. The Social Identity Perspective in Social Psychology One of the most in¯uential perspectives on the social psychology of group processes is the theory of social identity developed by the late Henri Tajfel (e.g. Tajfel, 1974; Tajfel & Turner, 1979, 1986; see also Hogg & Abrams, 1988; Abrams & Hogg, 1990, for reviews), and later revised and extended by John Turner (Turner, 1982, 1984, 1987). Tajfel's social identity theory (SIT) and Turner's associated self-categorization theory (SCT) both focus on the mechanisms which allow individuals to internalize category and group memberships as a part of the self-concept. Tajfel's theory outlines the role played by coherence and the need for categorization of social objects, and goes on to explore self-esteem maintenance and the drive for a sense of positive distinctiveness over outgroups. Turner's theory (Turner, 1982, 1985, 1987) focuses on the way in which self-categorization lies at the heart of most group processes, such as group polarization (Turner & Oakes, 1989; Turner, 1991) and stereotyping (e.g. Haslam, Turner, Oakes, & McGarty, 1992). The Self and Social Identity Tajfel was clear that his primary aim was to develop a theory of intergroup relations (Tajfel, 1974), and not a theory of the self. Although Turner's self-categorization theory does have more to say about the nature of self (cf. Abrams, 1992; Deaux, 1992; Hogg & Abrams, 1990; Schi€man & Wicklund, 1992), it retains Tajfel's focus on group processes and instances where ingroup members act in a relatively homogeneous manner. Recently, it has been suggested that the social identity paradigm requires a more elaborate treatment of the self and individual di€erences (see for example, Abrams, 1992; Breakwell, 1986, 1991; Deaux, 1992), and it is suggested here that one way in which this might be achieved is by introducing the notion of possible social identities. Possible Selves A useful approach to the self and motivation has been forwarded by Hazel Markus, Paula Nurius, and their colleagues (Markus & Nurius, 1984, 1986, 1987; Markus & Ruvolo, 1989; see also Cross & Markus, 1994), in the shape of their theory of possible # 1998 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 28, 227±248 (1998)

Exploring temporal aspects of social identity

229

selves. The latter essentially represent individuals' beliefs about what the self was in the past and might become in the future, together with some estimate of the probability that di€erent possible selves will be realized. The perspective constitutes a valuable contribution to the literature in its attempt to encompass symbolic interactionist (see, for example, Gecas, 1982; Shrauger & Schoeneman, 1979; Stryker, 1987) and social cognition (see, for example, Nurius, 1989) approaches to the self (see, for example, Day, Borkowski, Punzo, & Howsepian, 1994; Oyserman & Saltz, 1993). The crucial link to motivation in the theory is a€orded by the contention that individuals attempt to achieve positively valued (i.e. desired) possible selves, whilst hopefully avoiding other, negatively evaluated (i.e. feared) possible selves. Furthermore, it is supposedly through the process of manufacturing possible selves that individuals devise plans and strategies to achieve or avoid particular outcomes (Markus & Nurius, 1987; Markus & Ruvolo, 1989; Markus, Cross, & Wurf, 1990). Unlike the SIT and SCT literature, the possible selves literature is rich in work which has examined both personality/personal identities and social/group identities, being from the outset, a perspective with a speci®c goal of understanding self-concept, motivation and identity, in contrast to social identity theory's goal of understanding intergroup relations, and self-categorization theory's aim of developing a perspective on group processes (although SCT theorists do argue their perspective addresses issues of selfhood).

EXTENDING THE CONCEPT OF POSSIBLE SELVES The Issue of Level of Analysis: Where Should We Look for Possible Selves? When conceptualized as cognitive representations, possible selves may be thought of as elements of self-schemata (cf. Kato & Markus, 1993; Markus, 1977; Markus & Sentis, 1982; Markus & Zajonc, 1985; Nurius, 1989)Ðorganized knowledge structures stored in long-term memory, which include information pertaining to the individual's self-concept and especially those traits the individual perceives as most central to his or her self (termed self-schematic traits by Markus, 1977). Individuals will have a variety of possible selves available to them, and those currently salient are part of what has been termed the working self-concept, i.e. that part of the cognitive self-system currently activated (Markus & Kunda, 1986). Already there are two similarities apparent with Turner and associates' SCT modelЮrstly, the notion that elements of the self may be conceptualized as cognitive schemata, and secondly, that such cognitive structures vary in salience so that at any particular moment, only a sample (and perhaps just a single schema or self-categorization) will be currently salient. However, social cognition approaches to the self have sometimes been criticized as overly individualistic and insensitive to social construction (see, for example, Deaux, 1992; Forgas, 1981; Hogg & Abrams, 1988; Markus & Zajonc, 1985). The possible selves perspective is less vulnerable to such criticisms, but a detailed exposition of the interface between individuals and social groups within the framework has yet to be provided. Some of these weaknesses are less noticeable in recent applications of the possible selves approach. For example, Day and associates (1994) have explored the # 1998 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 28, 227±248 (1998)

230

M. Cinnirella

modi®ability of possible selves in children, and Oyserman & Saltz (1993) the relationship between delinquency and possible selves. Both of these studies explicitly recognize the socially constructed nature of possible selves, for example by examining communication skills and their links to the self. Yet even these studies leave underexamined the relationship between group processes and individual constructions of possible selvesÐit is this individual-group dynamic which requires further elaboration in the possible selves perspective, and which can be explored by introducing the concept of possible social identities. Possible Social Identities The issue of level of analysis (see also Doise, 1986, 1988; Rosenberg, 1988) becomes especially relevant when one addresses the question of social identities and their relation to possible selves. It is argued here that one particular kind of possible self is a possible social identityÐperceptions of current and possible group memberships. It is also useful to make a distinction between possible personal selves, and possible social identities. In contrast to possible social identities, which focus on group memberships, possible personal selves represent an individual's perceptions of his or her personality, unique traits and attitudes. The relevance of such a distinction will certainly vary cross-culturally (see, for example, Shweder & Bourne, 1984; Triandis, 1989; Triandis, Leung, Villareal, & Clack, 1985; Triandis, Bontempo, Betancourt, Bond, Leung, Brenes, Georgas, Hui, Marin, Setiadi, Sinha, Verma, Spangenberg, Touzard, & Montmollin, 1986; Triandis, Bontempo, Villareal, Asai, & Lucca, 1988; Triandis, McCusker, & Hui, 1990). This suggests that our ®rst hypothesis can be a crosscultural one: Hypothesis 1: Individualistic cultures will often engender a greater focus on possible personal selves, with possible social identities becoming more important in collectivistic cultures (see also Markus & Kitayama, 1991). One way that this hypothesis could be explored, is by comparing the degree to which individuals think about what might happen to ingroups in the future, with the degree to which they think about what might happen to themselves as an individual, perhaps employing a fairly non-reactive measure such as a modi®ed `Who am I?' (Kuhn & McPartland, 1954), which has, for example, been successfully employed as a measure of individualism±collectivism (Triandis et al., 1990). In making this distinction between possible personal selves and possible social identities, a parallel can be seen with Tajfel's notion of personal versus social identities (e.g. Tajfel, 1974), which is admittedly a simpli®cation (personal and social identities can surely become intertwined, for example), but nevertheless a useful one for our current purpose. Possible social identities include conceptualizations of the social categories and groups an individual might have been a member of in the past, and could become a member of in the future. In addition, they also represent predictions about how existing social group memberships might change over time, and thoughts about how groups might have been in the past. Thus possible social identities can pertain to potential group memberships (both past and future), as well as current group memberships and thoughts about how these might have been di€erent in the past and could develop in the future (this distinction between current and possible group # 1998 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 28, 227±248 (1998)

Exploring temporal aspects of social identity

231

memberships is explored in more detail later in the discussion). This acknowledges the fact that individuals might fantasize, not only about what might happen to current groups to which they belong, but also what it might be like to join other groups in the future, and what it might have been like to have been a member of a particular group in the past. Like possible personal selves, possible social identities have a perceived source, an a€ect associated with them (i.e. are generally desired or feared), a level of salience (including both current activation and inherent accessibility) and an associated perception of their likelihood of being realized. Possible social identities can be perceived to originate from the self, other ingroup members, outgroup members, and other sources, such as the mass media and opinion leaders (Lazarsfeld, Berelson, & Gaudet, 1944). Associated with the perceived source will therefore be perceptions of credibility which will impact on the acceptance of possible social identities and the choice made from competing alternatives, as outlined in hypothesis 2: Hypothesis 2: Source credibility e€ects and established mechanisms of social in¯uence (see, for example, Turner, 1991) will impact on individual and group reactions to available possible social identities. This highlights how possible social identities of an ingroup, in the form of social representations, might be disseminated by outgroupers, as well as ingroup members. Ultra right-wing organizations fostering notions of `racial purity' for example, might not assign any source credibility to mainstream mass media representations of their organization and its possible future. In order to explore this hypothesis, selfcategorization studies of social in¯uence (see Turner (1991) for an overview) could be modi®ed in order to explore the degree to which the salience of in- and outgroup categories impacts on the activation and endorsement of possible social identities. In addition, discursive and rhetorical approaches (see, for example, Reicher & Hopkins, 1996) could be employed to explore the manner in which competing visions of a group's futureÐi.e. contested possible social identitiesÐare manifested in everyday talk and conversation, as well as in mass media. Possible social identities are linked to possible personal selves, such that individuals might think, for example, about how being extroverted helps them in their local amateur dramatics groupÐthis represents the under-explored interface between personal and social identityÐand, to use the current terminology, possible personal selves and possible social identities. To give another example: as well as thinking about what might happen to his/her political party after a forthcoming election (i.e. considering a possible social identity), a politician might also wonder about his/her future within the party, whether s/he might ever become its leader, and so on (i.e. develop possible personal selves concerned with status and power within an ingroup). Future research could explore, for example, the extent to which the desire to realize personal possible selves (e.g. a desire to achieve promotion at work) might in¯uence the strategies for social identity adopted by individuals (e.g. a desire to become a member of particular subgroups at work, or to identify with the organization). Since cognitive representations of social category and group memberships include, as well as possible selves, exemplars, relevant episodic memories, norms, prototypes and stereotypes, then possible selves will exert in¯uence on most of these other elements. For example, which possible social identities are currently salient will in turn in¯uence which ingroup exemplars (cf. Smith & Zarate, 1990) and stereotypes are # 1998 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 28, 227±248 (1998)

232

M. Cinnirella

salient or accessible. This clearly has consequences for SCT's concept of prototypicality (Turner, 1987, 1991)Ðcurrent operationalizations of the latter tend not to explore the possibility that past-oriented social identities may lead to the derivation of prototypicality based on past comparisons and outgroups. Thus, it can be suggested that: Hypothesis 3: The activation or salience of cognitions about ingroup exemplars, norms, prototypes and stereotypes, will be a€ected by the nature of the temporal orientation adopted towards the ingroup. For example, when England played Germany in the semi-®nal of the soccer European Championship of 1996, the British tabloid media used images and discourses from the second world war, encouraging readers to conceptualize the soccer game as a re-enactment of a past military con¯ict, and relying on appropriate past-oriented stereotypes and possible social identities (employing images, for example, of English soccer players dressed in military attire). Thus, if past-oriented possible selves are dominant when thinking about a social group, the activated and accessible group prototypes are likely to be quite di€erent to those which would otherwise be active when the temporal focus is on the present or possible future. It is also the case that past and present exemplars and prototypes can become `active' simultaneouslyÐin the current example, members of the 1996 England soccer team might constitute exemplars, but also, as a result of the media use of 1940s imagery, be complemented by relevant exemplars from the second world war, such as Churchill. This hypothesis could be explored experimentally using attempts to manipulate temporal orientation adopted towards an ingroup, and observing changes in measures of ingroup stereotypes, prototypes, and so on. A related hypothesis pertains to the perception of outgroups: Hypothesis 4: The current salience of (ingroup-)relevant outgroups will be in¯uenced by the temporal orientation adopted towards the ingroup. If one persists with the example of British reactions to the 1996 soccer European Championships, it might be argued that the past orientation suggested by World War II imagery was also likely to encourage perception of Germany as an outgroup, and activation of social memories in which Germany was a military enemy of Britain. This means that when group members focus on past-oriented possible social identities, they may re-activate formally dormant social stereotypes and seek positive distinctiveness by making comparisons with outgroups as they were in the past, or past outgroups as they are in the present. In terms of individual di€erences and the personal±social identity interface, this also implies that ingroup members can adopt qualitatively di€erent perspectives on the same ingroup (for example by perceiving di€erent ingroup exemplars), depending on the temporal perspective adopted. Once again, this hypothesis could be explored using experimental manipulations which attempt to in¯uence the temporal orientation ingroup members adopt towards their group, looking for e€ects on the choice of outgroup for intergroup comparison, and the content of outgroup stereotypes. Where large-scale social categories such as racial or nationality groups are the target of possible social identities, then this makes it all the more likely that a wide variety of relevant possible selves will be circulating in society, including the mass media (see also Bruner, 1962). Furthermore, since groups have varying access to, and # 1998 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 28, 227±248 (1998)

Exploring temporal aspects of social identity

233

control over, the mass media, the ability of group members to control the possible social identities associated with large-scale categories and groups will be associated with power di€erences in society (see also Deschamps, 1982; Tajfel, 1982). As outlined below, therefore, possible social identities are especially likely to be contested when they pertain to large-scale social groups: Hypothesis 5: The possible social identities associated with large-scale and di€use social categories and groups, such as nationalities and racial groups, will be diverse, subject to contestation, and in part di€used via mass media. This in turn suggests that, as a social group becomes the focus of mass media attention, so the possible social identities associated with it enter the public domain, and there is a sense in which freedom to generate positive and unchallenged social representations of the ingroup, is wrested from the hands of ingroupers, to be replaced instead by an ideological struggle as alternative, and perhaps negative, possible social identities for the ingroup emerge from outgroup sources, and are conveyed in the mass media. For example, a category such as nationality, since it represents an `imagined community' (Anderson, 1991), can be construed in a wide variety of sometimes contradictory ways, and it is, therefore, to be expected that this scale of social identity will be associated with a heterogeneous base of possible social identities from which an individual comes to make some choice. For example, those Britons who are pro-European may endorse quite di€erent possible national identities compared to those who vehemently oppose European integration (Cinnirella, 1996, 1997). This hypothesis that the scale of a social group will have an impact on the availability of competing possible social identities, could be explored empirically by comparing the content of, and processes a€ecting, possible social identities of small social groups, compared to those which pertain to large-scale categories and groups. It would be predicted that when considering large-scale categories and groups, individuals will be more concerned with socially shared and contested possible social identities, compared to when they think of smaller social groups, such as co-workers at the oce. In terms of level of analysis, it should therefore be clear that possible social identities exist in both individual cognitions and in society (as what Moscovici (1984, 1988) has termed social representations). This means that when engaging in the process of negotiating favourable identities and reputations (Baumeister, 1982; Emler & Hopkins, 1990; Go€man, 1959b; Swann, 1987; Wicklund & Gollwitzer, 1982), individuals must be at least partly constrained by the context set by possible social identities discussed by other group members, contained in the mass media, and so on (see also Rosenberg & Gara, 1985). What is particularly important for the social psychology of group processes is to develop a greater understanding of how shared possible social identities might be related to collective attempts at social identity maintenance. Where a substantial number of ingroupers become motivated to avoid or attain a shared possible social identity, then this is likely to lead to collective action. To pursue the example of European and national identities a step further, it might therefore be suggested that if dominant mass media social representations present a possible European identity as something undesirable, and to be feared (perhaps because it threatens national identityÐsee Cinnirella, 1997), then individual and group endorsement of such representations might lead to collective action in the form of voting against further European integration in elections and referenda. # 1998 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 28, 227±248 (1998)

234

M. Cinnirella

Methodological Issues It remains unusual for social identity and self-categorization approaches to include in-depth interview studies exploring the mechanisms through which individuals construct and maintain their personal and social identities (but see Reicher & Hopkins (1996) for an example of how useful such an approach can be). This represents a methodological divergence between the current perspective and that of mainstream social identity and self-categorization work, since it is suggested here that in-depth interview methodologies are extremely useful tools for exploring possible social identities. Data from such studies can be analysed using a variety of techniques, such as discourse analysis (Wetherell & Potter, 1988), rhetorical analysis (Billig, 1987) or other qualitative approaches. However, there is no reason why more quantitative approaches might not also be employed, perhaps using self-completion survey techniques to elicit perceptions of possible social identities from reasonably large samples of group members. This kind of approach would allow, for example, the exploration of the degree to which possible social identities are shared amongst ingroupers, as well as an examination of how heterogeneous such perceptions are. Furthermore, more traditional experimental techniques could be usefully employed, as outlined in the hypotheses, to explore some of the proposed di€erences between small- and large-scale groups, and the strategies adopted by individuals and groups when faced with di€erent possible social identities. There is no reason, for example, why experiments cannot try to manipulate the nature of possible social identities salient to group members and then test hypotheses about the e€ects of such manipulations. It is suggested that any single possible social identity will possess at least the following key properties, all of which could be explored in empirical investigations, although any single investigation might only focus on a selection: (1) Di€usion i.e. degree to which awareness of the possible social identity has di€used within both the ingroup and relevant other groups. (2) Degree of acceptance/validity within the ingroup. (3) A€ect i.e. whether desired/feared/neutral. (4) Perceived likelihood of realisationÐ(for future-oriented) possible social identities. (5) Perceived source e.g. ingroup versus outgroup. (6) Salience and inherent accessibility. (7) Temporal focusÐ whether it focuses on past, present, future, or some combination of these. (8) Qualitative content of possible social identityÐfor example, a description of the speci®c past or future scenario(s) involved, including on the discourses and rhetoric employed by ingroupers. Empirical investigations of possible social identities require a `triangulation' of data emerging from such methods as self-concept measures (e.g. repertory gridsÐsee, for example, Weinreich, 1986; `Who am I?', self-narrative approaches, etc.), quantitative and qualitative measures of social identities, and discursive and social representations approaches to shared representations of a group's possible pasts and futures. In particular, it is suggested that possible selves methodology, wherein individuals are requested, for example, to outline the content, a€ect, control, and probability associated with a possible self, is expanded so as to be used to explore possible social identities, and that the data emerging from such analyses be examined both quantitatively and qualitatively.

# 1998 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 28, 227±248 (1998)

Exploring temporal aspects of social identity

235

TEMPORAL PERSPECTIVE In as much as possible social identities encompass notions of the self in the past, present, and future, they involve the kind of manipulations of a temporal sense of self discussed with some elegance by Go€man (1959a, b, 1968). The similarity between Go€man's perspective and the current one, is that both accept that individuals and social groups attempt to manipulate the image of the ingroup, and it is argued here that one aspect of this image manipulation concerns the manipulation of the possible social identities perceived for an ingroup: Hypothesis 6: Ingroup members are concerned to persuade both other ingroupers and also outgroupers, to endorse desired possible social identities of the ingroup i.e. to accept positively evaluated `visions' of what might happen to the ingroup in the future, or alternatively, positively evaluated constructions of the ingroup's history. The degree to which ingroup members are concerned about how their group is perceived by outsiders will vary as a function of: the scale of the group (being more likely with large-scale groups); the degree to which identity-maintenance is based on intergroup comparison (with groups not engaging in frequent comparisons being less likely to worry about outgroup perceptions), and the degree to which the group is `in the public eye'Ðfor example, the focus of mass media attention (with groups represented in the media caring more about their image in the eye of outsiders). These hypotheses about the conditions under which groups will attempt `imagemanipulation' can all be explored using experimental methodologies whereby the antecedent conditions are manipulated. The current perspective also has similarities with that of Breakwell (1986), who suggests that identity construction is often motivated by a desire for continuity over time. Both Breakwell (1986) and Go€man (1959a, b, 1968) suggest the importance of re-interpreting memories of the past and predictions of the future in order to maintain a degree of coherence and continuity of the self-concept. Similar arguments are forwarded by symbolic interactionist perspectives on self-narratives (see, for example, Gergen & Gergen, 1983, 1987, 1988; Mancuso & Sarbin, 1983). Here, it has been argued that culturally embedded traditions of storytelling and recurring narrative structures a€ect the manner in which individuals recount their life-histories. Further research in the future might endeavour to explore the degree to which such narrative structures are also employed by individuals when they re-construct ingroup histories (i.e. past possible social identities), and also when they talk about the possible futures for ingroups (i.e. about future possible social identities). Where the current perspective diverges to an extent from the self-narrative approach, is in the current focus on the degree to which perceptions of ingroup history and possible futures contribute to self-narratives, and the interest in the degree to which groups might collectively create shared self-narratives pertaining to the group. This suggests the next hypothesis: Hypothesis 7: Social groups will create shared `life stories' or narratives of the group which tie past, present and predicted futures into a coherent representation. # 1998 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 28, 227±248 (1998)

236

M. Cinnirella

To use the current terminology, this means that ingroup members will be motivated to re-interpret and re-construct past, present and future-oriented possible social identities so that a sense of temporal continuity is perceived to exist. This hypothesis could be tested, for example, by focusing on `real-life' (as opposed to ad-hoc laboratory) groups which have recently been forced to re-examine their group's history in order to make it more compatible with new future directions they wish to take the group in. Discursive approaches, using interview and focus group techniques to generate data, would be especially useful when exploring such issues. One way in which the perspective adopted here diverges slightly from that of traditional possible selves perspectives, is in the current argument that past-oriented thinking can be just as consequential for the self-concept as future-thinking. Markus and Nurius (1986) acknowledge that conceptions of past selves might constitute possible selves, if the individual perceives the possibility of their re-emergence in the future. However, it is suggested here that conceptions of what our groups might have been like, and also of what group memberships we could have held in the past, are both important elements of possible social identities. These representations of the past always have the potential to a€ect social identity construction in the present and the future. An ethnic group, for example, through archaeological or historical research, could re-discover or re-activate what one might call a `dormant' possible social identity. To take another example, the past orientation typical of many national and ethnic identities (see, for example, Crick, 1991; Hewstone, 1986; Samuel, 1989; Smith 1991) is often maintained by social representations or myths of a glorious past or `golden age', enshrined in folklore, national festivals, and museums (De Vos, 1983; Smith, 1991). In Cinnirella's study of national and European identities in Britain (Cinnirella, 1996, 1997), he suggests that British national identity is often past-oriented (partly due to Britain's former dominance of world a€airs and colonial empire), and that this acts as a potential barrier to the adoption of a European identity, since Britain's past is often construed as under threat from European integration (see also Hewstone, 1986; Sotirakopoulou, 1991). One thing this demonstrates is how potential future changes to identities and groups are often collectively evaluated in terms of whether the proposed change is compatible with the group's past. Perceived incompatibilities can lead to collective resistance to change or re-interpretation of the past to construct a revised and coherent collective memory. One important implication of these arguments is that any adequate theory of change in social identity, must take into account the degree to which reactions to change are conditioned by social representations of the group's past. However, despite the fact that some social groups appear to have a relatively stable temporal orientation towards the ingroup, other social groups will adopt varying temporal orientations as outlined in the following hypothesis: Hypothesis 8: For most groups, the temporal orientation adopted towards the ingroup, and in turn, whether or not relevant possible selves are past- or futureoriented, will ¯uctuate depending on contextual factors. Some of the mechanisms or cues involved are likely to be: reactions to mass media portrayals of the ingroup; meeting past in- or out-group members; and re-visiting locations previously loaded with signi®cance for the ingroup. These may all serve to activate, or make more accessible, past or future possible social identities. Future # 1998 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 28, 227±248 (1998)

Exploring temporal aspects of social identity

237

empirical research could examine the degree to which these and other cues (many of which could be manipulated by the researcher), a€ect temporal orientation, so that in time, it may be possible to develop typologies of the kind of environmental and other cues which lead to past versus future temporal orientations. The ¯exibility to de®ne possible social identities in di€erent ways and to adopt di€erent temporal orientations in di€erent situations would suggest that it is not wise to try and develop a stage theory of the temporal orientations adopted by groups or their membersÐfor example, in a similar manner to the stage model of intergroup relations forwarded by Taylor and McKirnan (1984; see also Eiser (1986) for a critique). However, the concept of possible social identities can certainly help enrich such theories. For example, Taylor and McKirnan discuss consciousness raising which arises when individuals who have failed at their attempt to `pass' from the ingroup try to persuade other ingroup members to adopt collective action. It could well be argued that such e€orts could be conceptualized as attempts to disseminate possible future social identitiesÐin this case, representations which suggest a more acceptable future for the ingroup if collective action to change intergroup relations is initiated. Conceptualizing the consciousness-raising process as an attempt to disseminate possible social identities has methodological implications, suggesting a need to investigate the generation and di€usion of relevant social representations and discourses, with a particular focus on predictions of the group's future(s).

CONSEQUENCES OF DESIRED POSSIBLE SOCIAL IDENTITIES Just as possible selves are expected to have a direct impact on goal-directed behaviour, so possible social identities will also in¯uence behaviours surrounding social identity maintenanceÐat both an individual and collective level. When individuals strive to obtain a desired possible social identity, they will, to some extent, have to recruit other group members in order to validate the identity (see also Stryker, 1987), by default necessitating some form of shared possible social identity. For example, an individual who wishes to perceive a very positive future for one of his or her social groups would face diculties receiving social validation of this perception if the vast majority of group members instead adopt a pessimistic and fearful prediction of their group's future (i.e. a feared possible social identity). If we attempt to incorporate some of the ideas contained in the Tajfel±Turner model of reactions to negative identity (Tajfel & Turner, 1979, 1986), this allows us to suggest when possible social identities might focus on current group memberships and when, in contrast, they are likely to focus instead on possible or hypothetical group memberships: Hypothesis 9: In situations where group memberships are seen as relatively ®xed (group boundaries are perceived as impermeable) and social beliefs tend to promote collective actions (as in Tajfel & Turner's (1979, 1986) notion of the social change belief system), then possible social identities are likely to be largely based on projections of current social group memberships. Hypothesis 10: However, in societies where individualism is high, group boundaries are perceived as permeable, and a social mobility belief system promotes the # 1998 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 28, 227±248 (1998)

238

M. Cinnirella belief that movement between groups is acceptable, possible social identities are more likely to also include simulations of what it might be like to join other social groups in the future.

For example, individuals might develop possible social identities which focus on predictions of what it might be like to emigrate and develop a di€erent national identity. Hypotheses 9 and 10 could both be explored using experimental and/or survey methodologies which, could, for example, examine the degree to which scores on measures of individualism±collectivism and subjective belief structures (Tajfel & Turner, 1979, 1986) can predict the degree to which possible social identities focus on current versus possible, social group memberships. When individuals or subgroups attempt to gain a new, desired social identity, and perhaps to abandon an old social identity at the same time, it is often the case that a degree of anticipatory identi®cation will take place (see, for example, Hyman, 1942; Hyman & Singer, 1968; Merton & Rossi, 1968). Ellemers and Van Knippenberg have demonstrated experimentally the existence of anticipatory identi®cation where group boundaries are perceived to be permeable (Ellemers, Van Knippenberg, de Vries, & Wilke, 1988; Ellemers, Van Knippenberg, & Wilke, 1990; Van Knippenberg, 1978, 1984). This is highly congruent with the ®nding that anticipated future selves can elicit information-seeking and in some way advance preparation for identity change (Deutsch, Ruble, Fleming, Brooks-Gunn, & Stangor, 1988; Fiske & Taylor, 1991). Group boundaries must be perceived to be permeable in order to allow for anticipatory identi®cation, but this process also depends crucially on the availability of desired possible social identities and the perception that they are attainable. Where the groups concerned are large-scale di€use entities, such as ethnic or national categories, media social representations will in¯uence individuals' perceptions of whether the identity is attainable, and thus, indirectly, their degree of anticipatory identi®cation. This has important implications for the social psychology of group processes. For example, it raises the possibility that individuals might come to endorse the norms and stereotypes of a desired social identity prior to adopting that identity. Thus, the notion of anticipatory identi®cation/socialization can o€er an insight into how desired possible social identities might facilitate changes in social identity construction and group membership. Additionally, by incorporating the concept of possible social identities into work on anticipatory identi®cation, it should be possible to gain predictive power and in time develop a model of how and when anticipatory identi®cation takes place. The current framework also raises the possibility of anticipatory identi®cation based on feared possible social identitiesÐ Cinnirella's survey research on European integration (Cinnirella, 1993, 1996) suggests, for example, that some British citizens have accepted, with regret, their likely development of a European identity in the future, as if it were being imposed upon them by higher authorities. As measured on standard quantitative social identity measures, such individuals show signs of anticipatory identi®cation as `European', but at the same time manifest an embryonic European identity largely devoid of a€ective connotations or epistemological signi®cance for the self (Cinnirella, 1996, 1997). The nature of the interaction between possible social identities and anticipatory identi®cation is set out in the hypothesis below: Hypothesis 11: The degree of anticipatory identi®cation manifested by an individual is contingent on the endorsement and availability of desired possible # 1998 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 28, 227±248 (1998)

Exploring temporal aspects of social identity

239

social identities which suggest that attainment of the group membership is both possible and desirable. Explorations of this hypothesis could focus on the relationship between the content of available possible social identities and the degree to which anticipatory identi®cation is evident, such that, in time, predictive models might help explain the circumstances under which anticipatory identi®cation is more or less likely. CONSEQUENCES OF FEARED POSSIBLE SOCIAL IDENTITIES There is some evidence to suggest that individuals engage in self-serving biases when they construct their sense of social identity (see, for example, Cinnirella, 1993; Marques & Yzerbyt, 1988). In as much as possible social identities are also represented in the cognitive structures of individuals, then they are open to various selfserving biases documented in the social cognition literature (see Baumeister (1995) and Fiske (1995) for excellent reviews). These mechanisms, which serve to protect the self from negative social identity, will also protect the self from negative possible social identities. Much depends on whether feared social identities are associated with the past, present, or future, suggesting that any model of reactions to negative or threatened social identities must take into account the temporal dimension of identity construction. Current operationalizations of social identity theory appear somewhat static in their approach to negative social identity and the temporal aspects of social identity construction. For example, the Tajfel±Turner macro-level model of identity maintenance (Tajfel & Turner, 1979, 1986) seems best suited to explaining reactions to currently negative social identities, and seems more problematic when applied to predicted negative social identity (i.e. feared possible social identitiesÐfor example, the fear that one's political party might lose an election). The incorporation of feared possible social identities into the Tajfel±Turner model also expands the treatment of self-esteem maintenance: for example, self-esteem could be linked to the perceived likelihood of attaining desired and feared possible social identities, as well as the capability of a social identity to convey positive distinctiveness. Thus, for example, the potentially puzzling scenario of a current social identity failing to provide positive self-esteem despite multiple opportunities for positive distinctiveness, might be understandable if members of the relevant social group are focused on a widely shared perception of their group's future demise, i.e. a feared possible social identity. This important notion is outlined in the hypothesis below: Hypothesis 12: The degree to which a social identity satis®es motivational needs such as positive self-esteem, distinctiveness and control, is dependent on the temporal orientation currently adopted towards the social group in question. This implies that current motivational needs can be satis®ed or, alternatively, threatened, by focusing on the past or predicted future for an ingroup, as well as the ingroup's current situation. By taking measures of the degree to which social identities satisfy motivational needs, and exploring the associations of such measures with measures of temporal orientation and the existence of possible social identities, it should be possible to explore this hypothesis empirically. # 1998 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 28, 227±248 (1998)

240

M. Cinnirella

The work on possible selves suggests that in general, respondents tend to have more positive than negative possible selves available to them, and that this is likely to be due to the general bias to maintain a positive self-concept (Markus & Nurius, 1986; see also Taylor & Brown, 1988). This highlights another similarity between possible selves and social identity perspectivesÐa posited link between cognitive structures and self-esteemÐwith both traditions stressing the utility of self-schemata which allow a sense of positive self-esteem. Extending this notion to the concept of possible social identities allows for the formulation of the next hypothesis: Hypothesis 13: Individuals will prefer to focus on possible social identities which help maintain a positive sense of social identity. This suggests that, for example, if a group's past seems to be more conducive to maintaining positive identity than its predicted future, then ingroup members are likely to adopt a past orientation to the ingroup and focus on past-oriented possible social identities. It should, however, be remembered that the degree to which ingroup members are able to collectively resuscitate the ingroup's past, for example, will depend on the ingroup's power and status within society, access to and control over, mass media, and the degree to which negative social identity has been internalized and accepted as legitimate due to acceptance of hegemonic ideologies which justify the ingroup's position. This posited desire to maintain a positive social identity also has implications for how individuals and social groups react to threatened social identity: Hypothesis 14: When a possible social identity is perceived to be undesirable or feared, then individuals will either (a) challenge the legitimacy of the possible social identity; or (b) where the latter is not possible, then focus on alternative, more positively evaluated social identities, if available. For example, if social identity is threatened by an outgroup which seeks to paint the ingroup's history in negative terms, then ingroupers might try to challenge this possible social identity by denying its legitimacy or accuracy, or where such a strategy fails, instead attempt to move the focus to positively evaluated future possible social identities, when available. An example of this can be seen in the desire of many Germans to look towards the future and escape the shadow of Nazism, and in the recent debate concerning whether Japan should formally apologize to nations such as the United Kingdom, for its treatment of enemy prisoners during the second world war. Those feared possible social identities associated with the future would, in turn, tend to encourage a focus on the past or the present. An example of this e€ect was discovered by Cinnirella (1993, 1996) who found that when British respondents perceived European integration to pose a threat to British identity, they often activated past-oriented possible British identities, associated with the former power of the British empire and Britain's past industrial and military dominance. This highlights how even when a group's future is uncertain, and there are few opportunities for fashioning positive social identity by focusing on the current state of the group, a positive sense of social identity can still be created by focusing on positively evaluated reconstructions of the group's past. This perspective is compatible with that of Moreland, Levine and colleagues, who emphasised in their model of group socialisation, how individuals engage in an ongoing evaluation of the degree to which social groups are perceived to satisfy various goals, and how this evaluation encompasses consideration of the group's past, # 1998 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 28, 227±248 (1998)

Exploring temporal aspects of social identity

241

present and future (Moreland & Levine, 1982, 1984; Moreland, Levine, & Cini, 1993). Combining the current perspective with the ideas of Moreland et al. and Tajfel and Turner, we can predict when a negatively evaluated possible social identity will lead an individual to actually consider leaving the ingroup, or else focusing self-perception on a di€erent social identity: Hypothesis 15: When an individual's sense of social identity is not entrenched, there is low commitment to the group (Moreland et al., 1993), and group boundaries are perceived to be permeable, then perception of a feared possible social identity may lead s/he to focus on other social identities in their repertoire (see Steele, 1988), attempt to `pass' as a member of another group (cf. Tajfel & Turner, 1986), or even leave the group in order to join a more satisfying one. Such strategies yet again depend on the cooperation of others in our social environmentÐfor example, not to perceive us as a member of group x any longer. One aspect of escaping from delinquent labels, for example, involves developing the communication skills required to negotiate a more favourable reputation and possible self which is accepted by others (Oyserman & Saltz, 1993; see also Ito, 1994; Oyserman & Markus, 1990). Studies of juvenile delinquency using the possible selves approach can, in some way, also be conceptualized as studies of social identity maintenance and, in particular, cognitive alternatives (Tajfel & Turner, 1979, 1986). Such studies are useful in that they demonstrate how cognitive alternatives can be empirically investigated, and the utility of conceptualizing of them as possible social identities which are subject to negotiation within social interaction. This is, therefore, more than simply substituting possible selves terminology into the social identity modelÐit is acknowledging that possible selves-type approaches can enrich social identity models by suggesting empirical and theoretical means for incorporating the negotiation of identities. The notion of possible social identities can only enrich that of cognitive alternatives if the methodological and theoretical issues outlined above are addressedÐin particular, more varied, multi-level methodologies are needed if the various elements of possible social identities are to be encompassed. It is extremely important to ascertain how feared possible social identities are explained by both individuals and groups. It is suggested here that causal explanations (see Nisbett & Ross (1980) for a classic overview) of the past and possible future will be important and, furthermore, that social groups will develop shared explanations of possible selves. Current knowledge of attributional processes and the selfconcept suggests the following hypothesis: Hypothesis 16: Given the tenacity of self-serving biases in attributions (Bradley, 1978; Miller & Ross, 1975), it could be predicted that feared possible selves will often be explained externally. Where shared and feared possible social identities are concerned, outgroups are likely targets for blameÐfor example, members of a group may fear that the group might lose its status in society, and blame this on a speci®c outgroup. This is one way in which feared possible social identities might impact on intergroup relationsÐin this example, by actually serving to encourage hostile perceptions of an outgroup. Similarly, Cinnirella (1996) found evidence to suggest that many British respondents blame nations such as France and Germany for maintaining the process of European integration, and in some sense can be said to also be blaming these countries for what # 1998 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 28, 227±248 (1998)

242

M. Cinnirella

are perceived to be negatively evaluatedÐi.e. fearedÐpossible social identities, in which Britain's sovereignty is perceived to be threatened. In addition, it is also possible for outgroups to be blamed for acting as barriers to the attainment of desired possible social identities: `poor white racism', for example, can be conceptualized as the struggle for desired possible social identities and the scapegoating of relevant outgroups when the failure to achieve such identities must be explained. This suggests the next, and ®nal hypothesis: Hypothesis 17: Ingroup members will develop explanations for failure to attain desired possible social identities, and these are likely to be biased in the same way that explanations of ingroup failures tend to be biased (in the so-called `ultimate attribution error'Ðsee Pettigrew, 1979), with a tendency to explain such failures externally. Even past-oriented possible selves could be a factor in intergroup con¯ict, if an outgroup challenges the legitimacy of a past-oriented and desired possible social identity. In this way, and because the current approach adopts a temporal perspective which includes the past as well as the future, the concept of possible social identities is more ¯exible than competing concepts, such as cognitive alternatives (Tajfel & Turner, 1979, 1986) or fraternalistic relative deprivation (Runciman, 1966). This hypothesis could be explored empirically by seeking explanations from ingroup members when desired possible social identities are not attainedÐfor example, when a sports team fails to win a major competition that its supporters believed it would win.

CONCLUSIONS The concept of possible social identities outlined above could be encompassed within social identity and self-categorization theories and thus enrich, rather than threaten, their treatment of identity and group processes. An important element of the Tajfel± Turner model of social identity maintenance is the idea that individuals may have perceptions of cognitive alternatives (Tajfel & Turner, 1979, 1986) to the current intergroup situation. However, the concept of cognitive alternatives, and some other elements of the Tajfel±Turner (1986) model remain inadequately de®ned (see, for example Abrams, 1992): it is not clear at what level of analysis (individual, group, both?) cognitive alternatives operate, i.e. where to look, empirically, for cognitive alternatives. In addition, the mechanisms through which they a€ect social identity maintenance are under-described, and there is little or no discussion of how perceptions of cognitive alternatives are arrived at. It is suggested here that possible social identities constitute an important element of cognitive alternatives, and furthermore that the Tajfel±Turner model would bene®t from expanding the notion of cognitive alternatives to the status quo so that it also encompasses wider perceptions of a group's possible pasts and futures, and the individual's role within the group. This necessitates the acceptance that cognitive alternatives are created by individuals but can also come to be shared, and contained in cultural mediums such as television. If social identity researchers adopt an approach to cognitive alternatives similar to that used by other researchers to examine possible selves, it would open the way for a greater understanding of how cognitive alternatives are negotiated by # 1998 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 28, 227±248 (1998)

Exploring temporal aspects of social identity

243

individuals and also how they might come to be shared and disseminated within and between groups. The essence of this argument is that it would be useful to consider cognitive alternatives to be possible social identitiesÐthis would necessitate a broadening of the cognitive alternatives concept to include the variety of perceptions which comprise possible social identities, as outlined above. Furthermore, whilst the Tajfel±Turner (1979/1986) model is largely geared towards understanding reactions to negative and/or threatened social identity, in contrast, the argument forwarded in the current paper is that cognitive alternatives, especially when re-conceptualized as possible social identities, play an important role in everyday identity maintenance, and are not solely relevant to situations in which identity is threatened. Adding the concept of possible social identities to SIT and SCT would strengthen both perspectives' capacity to encompass temporal issues pertaining to social identity maintenance, an aspect of social identity which remains under-explored and not fully articulated in either paradigm. One particularly noticeable omission is a lack of theorising about past social identities and the manner in which past, present and future may be re-constituted to create meaningful `stories', at both the individual and group (i.e. shared) levels. This is, therefore, not simply a methodological lacuna associated with the inherent diculties of encompassing the temporal dimension in laboratory experiments on social identityÐit also represents an important theoretical lacuna. Additionally, the possible social identities concept enhances the ability of the Tajfel± Turner model to predict when intergroup con¯ict might occurÐby suggesting that some intergroup con¯icts may be partly caused by struggles between in- and outgroups over the de®nition of possible social identities. Addition of the possible social identities concept to social identity and self-categorization models might require increased ¯exibility in the methodologies typically employed, with the realization that qualitative approaches, such as in-depth interviews analysed in terms of discourse and rhetoric, can usefully complement traditional laboratory experiments. Recent work within the SCT tradition suggests that such methodological open-mindedness may become more common in the future (see, for example, Turner, Oakes, Haslam, & McGarty, 1994). In the 17 hypotheses outlined above, the concept of possible social identities has been mapped out and its consequences for personal and social identity explored. This concept extends the notion of possible selves to include hypothetical thinking about social groups, and has both methodological and theoretical implications for the study of social identity. In particular, it has been suggested that social identity maintenance is often concerned with past and (predicted) future scenarios, as well as the current situations of social groups, and that exploring the degree to which possible social identities are past- or future-oriented will provide invaluable insights into social identity construction. One of the primary reasons why possible social identities are so crucial to group processes and identity maintenance is due to their direct impact on behaviour and motivation, with the notion that individuals and collectives are driven to attain desired possible social identities and avoid the realization of feared possible social identities. Exploration of the hypotheses outlined above would best proceed in a programmatic fashion so that knowledge of possible social identities and their functioning can be enhanced in a systematic manner and forged into a coherent body of knowledge. The development of a theory of social identity better able to address the temporal dimension should lead in time to better social psychological understanding of complex intergroup con¯icts, such as the bitter struggle for identity in the former Yugoslavia, which is steeped in issues pertaining to past, present, and future identities. # 1998 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 28, 227±248 (1998)

244

M. Cinnirella REFERENCES

Abrams, D. (1992). Processes of social identi®cation. In G. M. Breakwell (Ed.), Social psychology of identity and the self concept. London: Surrey University Press/Academic Press. Abrams, D., & Hogg, M. A. (Eds) (1990). Social identity theoryÐconstructive and critical advances. Hemel Hempstead: Harvester Wheatsheaf. Anderson, B. (1991). Imagined communities (2nd ed.). London: Verso. Baumeister, R. F. (1982). A self-presentational view of social phenomena. Psychological Bulletin, 91, 3±26. Baumeister, R. F. (1995). Self and identity: an introduction. In A. Tesser (Ed.), Advanced social psychology. New York: McGraw-Hill. Billig, M. (1987). Arguing and thinking. Cambridge: Cambridge University Press. Bradley, G. W. (1978). Self-serving biases in the attribution process: A reexamination of the fact or ®ction question. Journal of Personality and Social Psychology, 36, 56±71. Breakwell, G. M. (1986). Coping with threatened identities. London: Methuen. Breakwell, G. M. (1991). Identity, social representations and action: sex and politics. Presented at the British Psychological Society (Social Psychology section) annual conference. University of Surrey, Guildford, September. Bruner, J. S. (1962). Myth and identity. In J. S. Bruner (Ed.), On knowing: Essays for the left hand. Cambridge, MA: Harvard University Press. Cinnirella, M. (1993). Social identity perspectives on European integration: a comparative study of national and European identity construction in Britain and Italy. PhD thesis, University of London. Cinnirella, M. (1996). A social identity perspective on European integration. In G. M. Breakwell, & E. Lyons (Eds), Changing European identities: Social psychological analyses of social change. Oxford: Butterworth±Heinemann. Cinnirella, M. (1997). Towards a European identity? Interactions between the national and European identities manifested by university students in Britain and Italy. British Journal of Social Psychology, 36, 19±31. Crick, B. (1991). The English and the British. In B. Crick (Ed.), National identities: The constitution of the United Kingdom. Oxford: Basil Blackwell. Cross, S. E., & Markus, H. R. (1994). Self-schemas, possible selves, and competent performance. Journal of Educational Psychology, 86, 423±438. Day, J. D., Borkowski, J. G., Punzo, D., & Howsepian, B. (1994). Enhancing possible selves in Mexican-American students. Motivation and Emotion, 181, 79±103. De Vos, G. A. (1983). Adaptive con¯ict and adjustive coping: psychocultural approaches to ethnic identity. In T. R. Sarbin and K. E. Scheibe (Eds), Studies in social identity. New York: Praeger. Deaux, K. (1992). Personalizing identity and socializing self. In G. M. Breakwell (Ed.), Social psychology of identity and the self concept. London: Surrey University Press/Academic Press. Deschamps, J.-C. (1982). Social identity and relations of power between groups. In H. Tajfel (Ed.), Social identity and intergroup relations. Cambridge/Paris: Cambridge University Press, Maison des Sciences de l'Homme. Deutsch, F. M., Ruble, D. N., Fleming, A., Brooks-Gunn, J., & Stangor, C. (1988). Information-seeking and maternal self-de®nition during the transition to motherhood. Journal of Personality and Social Psychology, 55, 420±431. Doise, W. (1986). Levels of explanation in social psychology. Cambridge: Cambridge University Press. Doise, W. (1988). Individual and social identities in intergroup relations. European Journal of Social Psychology, 18, 99±111. Eiser, J. R. (1986). Social psychology: Attitudes, cognitions, and social behaviour. Cambridge: Cambridge University Press. Ellemers, N., Van Knippenberg, A., de Vries, N., & Wilke, H. (1988). Social identi®cation and permeability of group boundaries. European Journal of Social Psychology, 18, 497±513. # 1998 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 28, 227±248 (1998)

Exploring temporal aspects of social identity

245

Ellemers, N., Van Knippenberg, A., & Wilke, H. (1990). The in¯uence of permeability of group boundaries and stability of group status on strategies of individual mobility and social change. British Journal of Social Psychology, 29, 233±246. Emler, N., & Hopkins, N. (1990). Reputation, social identity and the self. In D. Abrams, & M. A. Hogg (Eds), Social identity theoryÐconstructive and critical advances. Hemel Hempstead: Harvester Wheatsheaf. Fiske, S. T. (1995). Social cognition. In A. Tesser (Ed.), Advanced social psychology. New York: McGraw-Hill. Fiske, S., & Taylor, S. (1991). Social cognition (2nd ed.). New York: McGraw-Hill. Forgas, J. P. (Ed.) (1981). Social cognition: Perspectives on everyday understanding. London: Academic Press. Gecas, V. (1982). The self-concept. Annual Review of Sociology, 8, 1±33. Gergen, K. J., & Gergen, M. M. (1983). Narratives of the self. In T. R. Sarbin, & K. E. Scheibe (Eds), Studies in social identity. New York: Praeger. Gergen, M., & Gergen, K. J. (1987). The self in temporal perspective. In R. Abeles (Ed.), Lifespan social psychology. Hillsdale, NJ: Erlbaum. Gergen, K. J., & Gergen, M. M. (1988). Narrative and self as relationship. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 21). San Diego, CA: Academic Press. Go€man, E. (1959a). The moral career of the mental patient. Psychiatry, 22, 123±142. Go€man, E. (1959b). The presentation of self in everyday life. Garden City, NY: DoubledayAnchor. Go€man, E. (1968). Asylums. London: Pelican. Haslam, S. A., Turner, J. C., Oakes, P. J., & McGarty, C. (1992). Context-dependent variation in social stereotyping 1: the e€ects of intergroup relations as mediated by social change and frame of reference. European Journal of Social Psychology, 22(1), 3±20. Hewstone, M. (1986). Understanding attitudes to the European Community: A socialpsychological study in four member states. Cambridge: Cambridge University Press. Hogg, M. A., & Abrams, D. (1988). Social identi®cations. London: Routledge. Hogg, M. A., & Abrams, D. (1990). Social motivation, self-esteem and social identity. In D. Abrams, & M. A. Hogg (Eds), Social identity theoryÐconstructive and critical advances. Hemel Hempstead: Harvester Wheatsheaf. Hyman, H. H. (1942). The psychology of status. Archives of Psychology, 38, Columbia University, Whole No. 269. Hyman, H. H., & Singer, E. (Eds) (1968). Readings in reference group theory and research. New York: The Free Press. Ito, M. (1994). The psychological traits of juvenile delinquents from the viewpoint of di€erentiation and internalization of ideal images. Japanese Journal of Educational Psychology, 42(4), 363±372. Kato, K., & Markus, H. R. (1993). The role of possible selves in memory. Psychologia, 36(2), 73±83. Kuhn, M. H., & McPartland, T. S. (1954). An empirical investigation of self-attitudes. American Sociological Review, 19, 68±76. Lazarsfeld, P. F., Berelson, B., & Gaudet, H. (1944). The people's choice. New York: Duell, Sloan, and Pearce. Mancuso, J. C., & Sarbin, T. R. (1983). The self-narrative in the enactment of roles. In T. R. Sarbin, & K. E. Scheibe (Eds), Studies in social identity. New York: Praeger. Markus, H. (1977). Self-schemata and processing information about the self. Journal of Personality and Social Psychology, 35, 63±78. Markus, H. R., & Kitayama, S. (1991). Culture and the self: implications for cognition, emotion, and motivation. Psychological Review, 98, 224±253. Markus, H., & Kunda, Z. (1986). Stability and malleability of the self-concept. Journal of Personality and Social Psychology, 51, 858±866. Markus, H., & Nurius, P. (1984). Possible selves. Presented at British Psychological Society Self and Identity Conference, University College, Cardi€, Wales, July. Markus, H., & Nurius, P. (1986). Possible selves. American Psychologist, 41, 954±969. # 1998 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 28, 227±248 (1998)

246

M. Cinnirella

Markus, H., & Nurius, P. (1987). Possible selves: the interface between motivation and selfconcept. In K. Yardley, & T. Honess (Eds), Self and identity: Psychosocial perspectives. Chichester: John Wiley & Sons. Markus, H., & Ruvolo, A. (1989). Possible selves: personalized representations of goals. In L. A. Pervin (Ed.), Goal concepts in personality and social psychology. Hillsdale, NJ: Erlbaum. Markus, H., & Sentis, K. (1982). The self in social information processing. In J. Suls (Ed.), Psychological perspectives on the self (Vol. 1, pp. 41±70). Hillsdale, NJ: Erlbaum. Markus, H., & Zajonc, R. B. (1985). The cognitive perspective in social psychology. In G. Lindzey, & E. Aronson (Eds), Handbook of social psychology (Vol. 1, pp. 137±230). Hillsdale, NJ: Erlbaum. Markus, H., Cross, S., & Wurf, E. (1990). The role of the self-system in competence. In R. J. Sternberg, & J. Kolligian, Jr. (Eds), Competence considered. New Haven, CT: Yale University Press. Marques, J. M., & Yzerbyt, V. Y. (1988). The black sheep e€ect: judgmental extremity towards ingroup members in inter- and intra-group situations. European Journal of Social Psychology, 18, 287±92. Merton, R. K., & Rossi, A. K. (1968). Contributions to the theory of reference group behaviour. In H. H. Hyman, & E. Singer (Eds), Readings in reference group theory and research. New York: The Free Press. Miller, D. T., & Ross, M. (1975). Self-serving biases in the attribution of causality: Fact or ®ction? Psychological Bulletin, 82, 213±225. Moreland, R. L., & Levine, J. M. (1982). Socialization in small groups: temporal changes in individual±group relations. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 15, pp. 137±192). New York: Academic Press. Moreland, R. L., & Levine, J. M. (1984). Role transitions in small groups. In V. Allen, & E. van den Vliert (Eds), Role transitions: Explorations and explanations. New York: Plenum. Moreland, R. L., Levine, J. M., & Cini, M. (1993). Group socialization: the role of commitment. In M. A. Hogg, & D. A. Abrams (Eds), Group motivation: Social psychological perspectives. London: Harvester Wheatsheaf. Moscovici, S. (1984). The phenomenon of social representations. In R. M. Farr, & S. Moscovici (Eds), Social representations. Cambridge/Paris: Cambridge University Press/ Maison des Sciences de l'Homme. Moscovici, S. (1988). Notes towards a description of social representations. European Journal of Social Psychology, 18, 3, 211±250. Nisbett, R. E., & Ross, L. (1980). Human inference: Strategies and shortcomings of social judgment. Englewood Cli€s, NJ: Prentice-Hall. Nurius, P. S. (1989). The self-concept: a social-cognitive update. Social-Casework, 70 5, 285±294. Oyserman, D., & Markus, H. (1990). Possible selves in balance: implications for delinquency. Journal of Social Issues, 46 2, 141±157. Oyserman, D., & Saltz, E. (1993). Competence, delinquency, and attempts to attain possible selves. Journal of Personality and Social Psychology, 652, 360±374. Pettigrew, T. F. (1979). The ultimate attribution error: extending Allport's cognitive analysis of prejudice. Personality and Social Psychology Bulletin, 5, 461±476. Reicher, S., & Hopkins, N. (1996). Seeking in¯uence through characterizing self-categories: an analysis of anti-abortionist rhetoric. British Journal of Social Psychology, 35 2, 297±311. Rosenberg, S. (1988). Self and others: studies in social personality and autobiography. In L. Berkowitz (Ed.), Advances in experimental social psychology, 21. New York: Academic Press. Rosenberg, S., & Gara, M. A. (1985). The multiplicity of personal identity. Review of Personality and Social Psychology, 6, 87±113. Runciman, W. G. (1966). Relative deprivation and social justice. Berkeley, CA: University of California Press. Samuel, R. (1989). Exciting to be British. In Introduction to R. Samuel (Ed.), Patriotism: The making and unmaking of British National Identity (3 vols.). London: Routledge. Schi€man, R., & Wicklund, R. A. (1992). The minimal group paradigm and its minimal psychology. Theory and Psychology, 2, 1, 29±50. # 1998 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 28, 227±248 (1998)

Exploring temporal aspects of social identity

247

Shrauger, J. S., & Schoeneman, T. J. (1979). Symbolic interactionist view of self-concept: through the looking glass darkly. Psychological Bulletin, 86, 549±573. Shweder, R. A., & Bourne, E. J. (1984). Does the concept of the person vary cross-culturally? In R. A. Shweder, & R. A. LeVine (Eds), Culture theory: Essays on mind, self and society. New York: Cambridge University Press. Smith, A. D. (1991). National identity. London: Penguin. Smith, E. R., & Zarate, M. A. (1990). Exemplar and prototype use in social categorization. Social Cognition, 8, 3, 243±262. Sotirakopoulou, K. P. (1991). Processes of social representation: A multi-methodological and longitudinal approach. Unpublished PhD thesis, University of Surrey. Steele, C. M. (1988). The psychology of self-armation: sustaining the integrity of the self. In L. Berkowitz (Ed.), Advances in experimental social psychology, (21). San Diego, CA: Academic Press. Stryker, S. (1987). Identity theory: developments and extensions. In K. Yardley, & T. Honess (Eds), Self & identity: Psychosocial perspectives. Chichester: John Wiley & Sons. Swann, W. B. (1987). Identity negotiation: where two roads meet. Journal of Personality and Social Psychology, 53, 1038±1051. Tajfel, H. (1974). Social identity and intergroup behaviour. Social Science Information, 13 2, 65±93. Tajfel, H. (1982). Instrumentality, identity and social comparisons. In H. Tajfel (Ed.), Social identity and intergroup relations. Cambridge/Paris: Cambridge University Press, Maison des Sciences de l'Homme. Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup con¯ict. In W. G. Austin, & S. Worchel (Eds), The social psychology of intergroup relations. Monterey, CA: Brooks/Cole. Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behaviour. In S. Worchel, & W. G. Austin (Eds), Psychology of intergroup relations. Chicago, IL: Nelson-Hall. Taylor, D. M., & McKirnan, D. J. (1984). A ®ve-stage model of intergroup relations. British Journal of Social Psychology, 23, 291±300. Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: a social psychological perspective on mental health. Psychological Bulletin, 103, 193±210. Triandis, H. C. (1989). The self and social behavior in di€ering cultural contexts. Psychological Review, 96, 506±520. Triandis, H. C., McCusker, C., & Hui, C. H. (1990). Multimethod probes of individualism and collectivism. Journal of Personality and Social Psychology, 59, 5, 1006±1020. Triandis, H. C., Bontempo, R., Betancourt, H., Bond, M., Leung, K., Brenes, A., Georgas, J., Hui, C. H., Marin, G., Setiadi, B., Sinha, J. B. P., Verma, J., Spangenberg, J., Touzard, H., & Montmollin, G. (1986). The measurement of etic aspects of individualism and collectivism across cultures. Australian Journal of Psychology, 38, 257±267. Triandis, H. C., Bontempo, R., Villareal, M. J., & Asai, M., & Lucca, N. (1988). Individualism and collectivism: cross-cultural perspectives on self-ingroup relationships. Journal of Personality and Social Psychology, 54, 2, 323±338. Triandis, H. C., Leung, K., Villareal, M., & Clack, F. (1985). Allocentric vs. idiocentric tendencies: convergent and discriminant validation. Journal of Research in Personality, 19, 395±415. Turner, J. C. (1982). Towards a cognitive rede®nition of the social group. In H. Tajfel (Ed.), Social identity and intergroup relations. Cambridge/Paris: Cambridge University Press, Maison des Sciences de l'Homme. Turner, J. C. (1984). Social identi®cation and psychological group formation. In H. Tajfel (Ed.), The social dimension: European developments in social psychology (Vol. 2). Cambridge: Cambridge University Press. Turner, J. C. (1985). Social categorization and the self-concept: a social cognitive theory of group behaviour. In E. J. Lawler (Ed.), Advances in group processes: theory and research, (Vol. 2). Greenwich, CT: JAI Press. Turner, J. C. (1987). A self-categorization theory. In J. C. Turner, M. A. Hogg, P. J. Oakes, S. D. Reicher, & M. S. Wetherell (Eds), Rediscovering the social group: a self-categorization theory. Oxford: Basil Blackwell. # 1998 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 28, 227±248 (1998)

248

M. Cinnirella

Turner, J. C. (1991). Social in¯uence. Buckingham: Open University Press. Turner, J. C., & Oakes, P. J. (1989). Self-categorization theory and social in¯uence. In P. B. Paulus (Ed.), Psychology of group in¯uence. Hillsdale, NJ: Lawrence Erlbaum. Turner, J. C., Oakes, P. J., Haslam, S. A., & McGarty, C. (1994). Self and collective: cognition and social context. Personality and Social Psychology Bulletin, 20 5, 454±463. Van Knippenberg, A. F. M. (1978). Status di€erences, comparative relevance and intergroup di€erentiations. In H. Tajfel (Ed.), Di€erentiation between social groups: Studies in the social psychology of intergroup relations. London: Academic Press. Van Knippenberg, A. F. M. (1984). Intergroup di€erences in group perception. In H. Tajfel (Ed.), The social dimension: European developments in social psychology. Cambridge/Paris: Cambridge University Press, Maison des Sciences de l'Homme. Weinreich, P. (1986). Manual for identity exploration using personal constructs. Warwick: Centre for Research in Ethnic Relations. Wetherell, M., & Potter, J. (1988). Discourse analysis and the identi®cation of interpretative repertoires. In C. Antaki (Ed.), Analysing everyday explanation: A casebook of methods. London: Sage. Wicklund, R. A., & Gollwitzer, P. M. (1982). Symbolic self-completion. Hillsdale, NJ: Erlbaum.

# 1998 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 28, 227±248 (1998)