NOT The Story of Social Identity · identity research and ideas. The late Henri Tajfel started the story, writing the ﬁrst social identity paper in 1971 (or possibly 1970). Turner

11:31:06:01:10

Page 13

Page 13

Introduction: Part 2

The Story of Social Identity•

John C. Turner and Katherine J. Reynolds

(Department of Psychology, Australian National University, Canberra)

In this chapter we shall outline briefly the originsand development of the social identity perspective

in social psychology from its beginning in 1971 upto the present (around 2009). This is a long andcomplex journey. Social identity ideas came intobeing to make sense of unexpected experimentaldata having to do with the effects of social categor-ization on intergroup behaviour (Tajfel, 1972a;Tajfel, Flament, Billig, & Bundy, 1971; Turner,1972, 1975). But what started as a limited analysisof specific processes in intergroup behavior hasdeveloped over some 30 or more years into abroad-ranging and powerful new perspective onhuman social psychology, with relevance to almostevery significant problem, finding, or theory in thefield. In fact, its relevance to all the social scienceshas become increasingly clear. It is useful to try tosummarize the story of this development to providea context for understanding current research and togive some feeling for where both social identitywork and social psychology need to go in the future.Our emphasis will be on trying to show how coreprinciples and themes developed rather thanreviewing endless empirical studies, and also ondispelling the many misconceptions that exist. Aswe move toward the present, the discussion will

inevitably become more selective, thematic, andbrief.

Why should anybody want to make the effort tofollow and understand this story? The short answeris because it outlines a new vision of human beingsand human minds: one that is rich, productive, andempirically and theoretically coherent. This viewrejects the dominant individualistic conception ofthe human mind and argues that a defining featureis the social and psychological interdependence ofthe individual and the group. Human beings areneither merely individuals nor merely group mem-bers. Their individual and group selves, and per-sonal and group aspects, exist in an uneasy butcreative interplay, both collaborative and antagon-istic. The social identity tradition is the most recentand easily the most powerful attempt to date tocome to grips with the scientific reality of the psy-chological group. This issue is at the heart of socialpsychology, with the consequence that insightsinto the functioning of the psychological group andits relationship to the self-process are relevant toevery fundamental problem of social psychology.

Properly grasped, the social identity perspectiveoffers an opportunity for intellectual and theor-etical coherence and predictive power that the

13

NOT FOR •FO

John C. Tuohn C. Tu

(Department oDepartment

DISTRIBUTION

riginsiginsve

inevitainevitabrief.rie

Why shouWhy shoufollow and undefollow and unde

because it outlineause it outlinuman minds: onuman minds: o

lly and theoreticly and theoretdominant individminant individ

nd and argues thad and argues tpsychological intsychological in

he group. Humahe group. Humals nor merely grs nor merely gr

group selves, agroup selves, ast in an uneasyin an uneas

tive and antagontive and antagonhe most recenthe most recent

to date too date tothe psy-the psy-

socialociahts

11:31:06:01:10

Page 14

Page 14

science has not had since Lewin’s field theory(Lewin, 1948, 1952). Its implications extendbeyond social psychology to psychology at large(and especially the problem of cognition) and theother social sciences. One does not have to agreewith this view to find value in the perspective.Science progresses by testing ideas rather thanby closing one’s mind to them. Regardless ofwhether a researcher agrees or disagrees, thosewho are serious about understanding human beingsneed to be aware of the social identity perspectiveand the arguments it puts forward on the limita-tions of the current individualist orthodoxy and thealternative it offers.

The chapter will not be a summary of everythingthat has been done over some 35 years, nor will itbe a list of all the individuals or the research thathas contributed to the large body of work on socialidentity. This is impossible and perhaps not evenhelpful. It is easy to miss the forest for the trees.Our aim is to try to map the forest through insightsprovided by Turner, who was involved in the storyfrom the very beginning. The chapter provides ahistorical overview of the basic theoretical ideasand how they developed, so that the contributionsto this volume can be understood in the broaderhistorical context. It also highlights very brieflythe more specific contribution of these basic theor-etical developments in the areas of stereotyping,leadership and power, social change, and personal-ity—four areas (amongst others) that have been afocus of Turner and colleagues’ research since the1980s and 1990s until the present day. None of thismeans of course that there are not many otherswho have made important contributions to socialidentity research and ideas. The late Henri Tajfelstarted the story, writing the first social identitypaper in 1971 (or possibly 1970). Turner wrote thesecond, also in 1971. Tajfel died in 1982, havingled collaborative work with Turner on social iden-tity theory or SIT (Tajfel & Turner, 1979, 1986).Although Tajfel was supportive of the turn to self-categorization theory or SCT (Turner, 1978, 1982,1985), he was unable to participate in the signifi-cant developments to come.

Beginnings: Social Identity andPositive Distinctiveness

John Turner met Henri Tajfel in 1971 at the Uni-versity of Bristol in the UK when he applied forand then began a PhD there under his supervision.Tajfel had already had a complicated and difficultlife and achieved much in social psychology (seeTurner, 1996). Turner had just obtained a degreein social psychology at the University of Sussex(after some eventful years and delays). In terms ofbackground, the two were chalk and cheese. Tajfelwas a Polish Jew, of middle class background, whohad become French and then British, been edu-cated on the continent, and was committed tounraveling scientifically the causes of prejudice.Turner was English, working-class, and aLondoner, who got to university by accident, cour-tesy of a free state education and state support.Tajfel’s dominant social categorization was eth-nicity and Turner’s was class, but they understoodeach other implicitly, agreeing about the kinds ofanalyses of society and social psychology thatwere needed and those that were nonsense. Bothhad thought about the meta-theory of social psych-ology a lot. Both had accepted the reality of mindas an evolutionary product and human universal,but rejected a conception of social psychology as areductionist “psychologizing” of society (seeTajfel, 1972a; Turner & Oakes, 1986, 1997). Therewas an embrace of the interactionist position pion-eered by Kurt Lewin and others, where the psycho-logical nature of individuals had to be apprehendedwithin an understanding of groups and member-ship in society.

In 1971 in the first volume of the new EuropeanJournal of Social Psychology, Tajfel and colleagues(Tajfel et al., 1971) published seminal studies onthe effects of social categorization on intergroupdiscrimination. Their procedure became known asthe minimal group paradigm. In discussing thesedata in the journal article there was no mention ofsocial identity, but instead a relatively circularappeal to a “generic norm” of ingroup favouritismor ethnocentrism. By the time Turner got to Bristolin September 1971, Tajfel (1972a) had produced anew explanation. It appeared in a few pages at the

14 Introduction: Part 2

NOTothowith with Science Science by closing by closingwhether a resewhether a res

ho are serious abho are serious abto be aware oe aware o

argumeargc FOR

it puit puent individuent individ

ffers.fferll not be a summt be a sum

over some 35 yeer some 35 yeviduals or the reviduals or the r

e body of workbody of woand perhape

oreor DISTRIBUTION

not not or the treeor the tre

rough insightsrough insightsin the storyin the story

rovides avides aideasdea

onsons

ananwerewerehad thoud thouology a lotology a lotas an evolutioas an evolutiout rejected a cont rejected a con

uctionist oni ““psychpsych1972a; Turner &1972a; Turner &

mbrace of the interace of the intt Lewin and otheLewin and othe

individuals had dividuals had nding of groupsnding of groups

ume of the new me of the n E, Tajfel and colleTajfel and coll

seminal studies minal studies n on intergroupn on intergro

me known ase known assing thesthes

tiotio

11:31:06:01:10

Page 15

Page 15

end of a chapter for a French textbook on socialcategorization. The story of social identity beginswith this chapter.

In these few pages Tajfel introduces and definesthe concept of social identity and puts forward thehypothesis that people are motivated by a need fora positive social identity and the idea that topreserve, maintain, or achieve a positive socialidentity they must establish a positively valued dis-tinctiveness for their own groups compared toother groups. Tajfel used the data from the minimalgroup paradigm to illustrate what he meant by theidea that people are motivated by the desire toestablish positive distinctiveness for their ingroupscompared to outgroups. He also discussed how theeffects of social comparisons on intergroup rela-tions differ from their effects in intragroup relations,where Festinger (1954) had argued that theyproduced a drive for uniformity pressures. Theminimal group data had shown that social categor-ization into groups in isolation from and uncon-founded by all the variables normally thought tocause group formation and negative intergroupattitudes was sufficient for discrimination in whichthe ingroup was favored over the outgroup. AsTajfel put it, they were data in search of a theory:His ideas about social identity and positive distinct-iveness heralded what later became SIT into life.

It is true that these data themselves were theproduct of a longstanding interest in social categor-ization and prejudice, both in Tajfel’s academicand personal life (Turner, 1996), but the minimalgroup paradigm was novel, as were the ideas cre-ated in response to it. Rabbie and Horwitz (1969)had earlier asked the same question about the roleof social categorization (was it alone sufficient forintergroup discrimination?) but their paradigm wasnot fully minimal and their findings and explan-ation were different (Turner, 1975; Turner &Bourhis, 1996). Tajfel did not present his ideas as“SIT”. No such term or theory existed at this time.In discussing what he saw as the central explana-tory idea in this analysis, Tajfel was explicit that itwas the notion that social comparisons betweengroups were aimed at establishing positivelyvalued distinctiveness for one’s own group.

Turner’s first task was to review the role of

social categorization in intergroup relations and thefindings of the minimal group paradigm in order toflesh out the explanation of the minimal group datain terms of the need for a positive social identityand the drive for positive ingroup distinctiveness,which he did in a review paper written before theend of 1971. Turner showed how social identityprocesses could provide a systematic account ofminimal and other forms of intergroup discrimin-ation and ingroup bias (in terms of a process thathe called social competition) that was not based ona conflict of interests à la Sherif (e.g., Sherif,1967). In the paper, Turner developed the theor-etical implications of this account for both thestudy of intergroup relations and processes ofself-categorization. Tajfel liked this paper a lot andasked Turner to present it at the Small Group Meet-ing of the European Association of ExperimentalSocial Psychology (EAESP) on Intergroup Rela-tions, held at Bristol in February 1972. Thismeeting was the first at which social identity ideaswere publicly presented and it generated excite-ment of various kinds. Subsequently, after circulat-ing the paper for several years and with somecollection of data, it was published in 1975(Turner, 1972, 1975).

Social identity research thus began with anexplicit focus on problems of intergroup dis-crimination and ethnocentrism, not the nature ofthe psychological group. Its key hypothesis wasthat people need to achieve positive ingroup dis-tinctiveness to gain a positive social identity, notthat there is a distinction between personal andsocial identity. Neither paper argued that ethno-centrism was universal or that social categorizationautomatically and inevitably produced ingroupbias, and the analysis did not argue that someintrapsychic drive for self-esteem is the basicfactor in either group formation or intergroupdiscrimination.

In terms of the need for self-esteem, Turner(1975) explicitly derived it from the interactionbetween social comparisons between groups andthe social values that existed in society and thatmembers used to define their group identities(Turner & Reynolds, 2001). Tajfel (1972b) hadalready explained that he saw social values as

The Story of Social Identity 15

NOTconcecon

hypothesis thhypothesis ta positive socia positive soc

eserve, maintainerve, maintaity they must estathey must est

ess for their oess for their os. Tajfel used tjfel used

m to illum toa FOR

te whte wmotivated bmotivated

inctiveness for tinctiveness for tHe also discusselso discus

ons on intergrouns on intergroun intragroup relatin intragroup relat

argued that thgued that pressuresssu

ialia DISTRIBUTION

heheegor-egor

uncon-uncght toht to

up

meemewere pwere pment of vment of ing the paping the papcollection of collection of

urner, 1972, 197er, 1972, 197cial identity rescial identity re

focus on probfocus on proband ethnocentrnd ethnocentr

ical group. Its kcal group. Itsto achieve positto achieve pos

a positive sociala positive sociaion between pern between pe

per argued thatper argued thatsocial categorizaial categoriz

roduced ingrouroduced ingrougue that someue that some

the basicthe basictergrouptergroup

ner

11:31:06:01:10

Page 16

Page 16

derivatives of social ideologies. Thus, from thestart, the need for self-esteem, for positive self-evaluation, was seen as fully “social-psychological” (Tajfel, 1972b; Turner & Oakes,1986) rather than being independent from socialcontext and self-definition. Turner (1975) alsopointed out, in a sign of what was to come, thatthere were several strategies available to perceiversto achieve a positive self-evaluation other thanintergroup discrimination, including changingone’s self-categorization and redefining the dimen-sions and values that defined an identity.

These two papers (Tajfel, 1972a; Turner, 1972)provided an essential foundation for what was tocome, but were more appropriately described as“positive distinctiveness theory” at this time ratherthan what became “SIT”. Indeed Tajfel never likedthe term SIT, which was coined by Turner andBrown (1978) as a deliberately abbreviated title,because he thought it did not do justice to the posi-tive distinctiveness analysis. Tajfel thought itwould mislead and in this he has been proven right.

An Intergroup Theory on Three Legs:The Emergence of SIT

Tajfel summarized the basic processes at work inthe new analysis as the “social categorization–social identity–social comparison–positive dis-tinctiveness sequence”. This sequence gave us anunderstanding of important processes in grouppsychology that could shape intergroup behavior(rather than individual behavior) and could beused to supplement the processes already specifiedby Sherif in his realistic group conflict theory(e.g., Sherif, 1967). Human beings defined them-selves in terms of social categorizations that pro-vided them with social identities. These wereimportant aspects of their self-concepts based ongroup memberships. These identities were definedand evaluated by intergroup (not intragroup) com-parisons on dimensions associated subjectivelywith perceivers’ social values and hence there wasa motive to define social identity positively, mean-ing positively different from other relevant groups,where social and psychological conditions encour-

aged comparisons in terms of such identities. Themotive for positive ingroup distinctiveness, wheninstigated, could lead to competitive, ingroup-favoring intergroup responses under certain condi-tions and other responses under other conditions.

From 1971 to 1976 the work was done thatproduced the next two important “legs” of whatTajfel and Turner (1979) called a “conceptual tri-pod”. Turner’s empirical work (Turner, 1975,1978) examined what happened in the minimalgroup paradigm when a “self–other” categorizationwas superimposed on the ingroup–outgroup cat-egorization of anonymous others, so that peoplecould react to ingroup and/or outgroup otherscompared to the (personal) self. He found thatpeople would ignore the ingroup–outgroup cat-egorization where they could make decisions thatdirectly favoured self provided that the social cat-egorization had not been used previously to defineself and remained minimal. On the other hand, theyacted on an ingroup-favoring basis even at the sac-rifice of personal and direct self-interest underconditions where the social categorization hadbecome more meaningful and salient to thembecause they had used it previously. These findingsconfirmed several things: that social categoriza-tions had to be accepted and to some degreeinternalized by members to have an effect (dis-crimination was not automatic and inevitable), thatself–other and ingroup–outgroup categorizationscould be seen as competing alternative ways ofdefining the self, and that under the right condi-tions individual self-interest was less powerful thangroup identity.

Tajfel saw all this as well and, on reading theseconclusions, immediately was stimulated to makea big conceptual breakthrough (which had clearlycome from his life experience and been in hismind for many years; see Turner, 1996). This wasthe “interpersonal–intergroup continuum”. At onelevel the continuum clarified that the minimalgroup findings did not imply that people wereonly group members, that they always acted interms of social identity processes and showedethnocentrism just because they belonged to socialcategories, and so on. It became clear that socialidentity processes were only expected to have


NOT191contecontepointed opointed there were sthere were to achieve a po achieve a

tergroup discritergroup discris self-categoriz-categori

d valuesd vwo FOR

t det depers (Tajfepers (Tajfe

sential foundatsential foundatmore appropriate appropri

ness theoryss theory”” at tat t””. Indeed Tajfel . Indeed Tajfel

coined by Tcoined by ely abbra

jusu DISTRIBUTION

ted ted to the posto the po

el thought itel thought itproven right.roven right.

concobecombecombecause tcause conconfirmed smed stions had to btions had to

ternalized by mernalized by mination was not aation was not a

er and ingrouper and ingroueen as competieen as compe

elf, and that unelf, and that unelf-interest was lf-interest was l

well and, on reall and, on rwas stimulated was stimulated

h (which had clh (which had and been in hand been in

996). This was996). This wumm””. At one. At one

miniminim

11:31:06:01:10

Page 17

Page 17

effects in selected situations where conditions wereright and that there was more work to be donespecifying such conditions. At another level, it wasa big conceptual advance because it enabled Tajfeland Turner to make a qualitative psychologicaldistinction between individual and collectivebehavior. Acting as a group member was psycho-logically different from acting as an individual(as Sherif had said) and people were capable ofdoing both. The language of social categorizationwas important for specifying what this meantoperationally. It also highlighted that more workwas needed on the conceptualization of social cat-egorization itself (the task that created SCT).

SIT (Tajfel & Turner, 1979) offers specifichypotheses about the causes and effects of shiftsalong the interpersonal–intergroup continuum.Being a theory of what goes on in society betweengroups, it has to incorporate an account of whenpeople act or are likely to act collectively, and thisis the continuum. In fact, however, the continuumhas been replaced by the personal–social identitydistinction of SCT, which built on the earlierconceptualization but better formulated its keyinsights. SCT did this so successfully that manycompletely confuse the two theories, but this is animportant area where the two theories are in factvery different. We shall come back to this point,but for the moment it suffices to make clear thatTajfel referred to the continuum as “acting in termsof self” versus “acting in terms of group”, whereasthe essence of the personal identity versus socialidentity continuum was that both were acting interms of self (Turner, 1978, 1982). Did or wouldTajfel have denied this? Of course not. As soon asthe reformulation was made (in 1978) he recog-nized its value and embraced it. The personal–social identity distinction has become so successfulthat it has moved from academic to popular mediaand is used routinely without awareness of thetheoretical context.

In 1974 Tajfel gave a series of lectures in whichhe set out to explore the relevance of the basic-process analysis for real-world societies in whichintergroup relations are characterized by hierarch-ies of power, wealth, and prestige. He put the socialpsychology into the context of social structure—

the organized social environment. These lecturesbecame the basis for his chapters in Tajfel (1978)and the whole set of ideas formed the basis for amajor research program on social identity andintergroup relations funded by the British SocialScience Research Council from 1974 to 1977.Turner became the director of the experimentalwing of the program in 1974 and then joined Tajfelas co-director with Howard Giles in 1976. Giles,with Richard Bourhis, joined the project to lookat the role of language in social identity; TonyAgathangelou and Sue Skevington worked on thefield wing; and Turner and Rupert Brown pursuedexperimental studies. The first experiment con-ducted (Turner & Brown, 1978) clarified some ofthe issues in the extended analysis, in particularelaborating the notion of “insecure group relations”into perceived instability and illegitimacy, muchof which then found its way into Tajfel and Turner(1979).

Given the collective psychology and theinterpersonal–intergroup continuum, the final legof the tripod was to specify how the members ofgroups in different positions in society reacted toeach other as a function of that psychology. It wasimportant to take into account the degree to whichpeople were objectively and subjectively able tomove from one group to another, whether groupboundaries were seen to be permeable or imperme-able, since this was relevant to shift along thecontinuum. It was also important to consider fromboth the dominant/high status and the subordinate/low status point of view the extent to which “cogni-tive alternatives” to the status quo were perceived.As a function of the different status positions of thegroups, they had motives either to maintain orachieve positive distinctiveness and how they actedto do so varied with their collective understandingof the intergroup relationship as stable or unstableand/or legitimate or illegitimate. Each group couldfollow one or more different strategies (individualmobility, social creativity, or social competition;Tajfel & Turner, 1979; Turner & Brown, 1978)depending on their specific understanding of theirsituation and other relevant factors.

Thus, far from arguing that low self-esteem,negative social identity, or low social status always


NOTTurnTu

istinctionistinction behavior. Actinbehavior. Actin

gically differentcally differenherif had said) rif had said)

oth. The languagoth. The languagtant for specfor spec

It also It ae FOR

hlighthlighceptualizaticeptualiza

task that createtask that createner, 1979) offer1979) off

es and effects os and effects oergroup continuergroup contin

n society betwesociety betwcount of nt

elyly DISTRIBUTION

enend thisd this

tinuumtinuentityentity

ier

Ginterpeinterpeof the tripof the trgroups in digroups in dieach other as a each other as a mportant to take itant to take i

e were objective were objectiom one group tom one group

were seen to be pere seen to be pis was relevant s was relevan

s also important also importangh status and thegh status and the

the extent to whie extent to whitus quo were pertus quo were pe

tatus positions ous positions r to maintain or to maintain o

how they actedow they actedderstandingerstanding

unstableunstablecouldould

ual

11:31:06:01:10

Page 18

Page 18

predicted more ingroup favouritism, from thebeginning the theory was more complex andsophisticated. SIT was capable of predicting avariety of intergroup attitudes and responses(including leaving one’s own group) depending onan interaction between at least four factors: statusposition × acting individually versus collectively ×perceived stability/instability × perceived legitim-acy/illegitimacy. A high status group member whosaw their superiority as illegitimate and unstableand felt little attachment to their group could jointhe low status group to eliminate a “conflict ofvalues” (dissonance, guilt). A low status groupmember who saw their inferiority as stable andlegitimate and could not join the high status groupas an individual was likely to engage in social cre-ativity, finding superiority on alternative dimen-sions whilst accepting inferiority on the statusdimension (and downgrading the latter’s import-ance). SIT (as it now was) suggested a whole arrayof possibilities and interesting hypotheses toexplore (Turner, 1999). What a pity then that criticsof the theory have aimed their arrows only at a fewmisguided hypotheses that had never been in thetheory (e.g., that people should show more ingroupbias the lower their personal self-esteem or themore they identified with the ingroup; that the the-ory cannot predict outgroup favouritism and theeffects of legitimacy, that it ignores power, onlyever predicts universal ethnocentrism, etc.).

The final theory was completed in 1976 and pub-lished as Tajfel and Turner (1979, being updated in1986). Other work such as the edited book byTajfel (1978) contained ideas and studies (andoften revisions of already available material) thatwere already known when writing the Tajfel andTurner chapter, which was deliberately a system-atization of where they had got to by 1976. Therewere other articles published after 1979 butnothing superseded what was written then (as asubstantive statement of SIT).

SIT was a new kind of theory in social psych-ology, which may be one reason why it has oftenbeen co-opted into the mainstream view of preju-dice it rejected. The revolution was begun bySherif and his colleagues (e.g., Sherif, Harvey,White, Hood, & Sherif, 1961), who argued that

intergroup attitudes were not the cause of inter-group relations but their effect, and moreover thatto understand how intergroup attitudes were gener-ated and changed one had to examine the relationsbetween groups at their own level, not reduceintergroup prejudices to the psychology of theindividual. Sherif’s specific theory of how inter-group relations generated intergroup attitudes wasa realistic conflict theory, emphasizing conflicts ofinterests, not lack of contact, familiarity, prejudice,etc. SIT was not a rejection of conflicts of interestbut complementary to it, expanding the role ofsocial structure, collective theories, and ideologiesand integrating an understanding of self-interestinto a richer and more collective view of the selfand social identity. Tajfel and Turner were inagreement with Sherif and felt that he would havebeen a social identity theorist—they were certainly,in principle, Sherifians.

Whereas the orthodox prejudice traditionfocuses on the role of a pathological, deviant,or irrational individual psychology in explainingsocial antagonism, SIT’s focus was on the collect-ive psychology of intergroup attitudes, producedwithin a social structure of intergroup relationshipsand mediated by people’s collective definition,perception, and understanding of those relation-ships. The theory rejected individualism andreductionism, arguing that all cognition was social-psychological and that the political and macro-societal complexities of intergroup relationshipswere socially shared, group-based interpretationsand were fundamental to racism and social con-flict. There is much more to say about the newthinking provided by the theory, but we shall returnto it later. For the moment, let us merely note whatwas distinctive and new about the theory:

1. It focused on group psychology, not individualpsychology, to explain racism, prejudice, andconflict.

2. It agreed with Sherif and colleagues that inter-group attitudes followed and did not causeintergroup relations, but added social identityprocesses to realistic goal relations to explainthe effects of intergroup relations.

3. It put intergroup relations into society, a social


NOT(in(an inan inposition position perceived stperceived sacy/illegitimacycy/illegitimacaw their superiorw their superio

felt little attacittle attacstatus st

is FORup toup to

ance, guilt)ance, guilsaw their infersaw their infer

uld not join the not join thelikely to engageikely to engage

ority on alternatority on alternanferiority on teriority on

g the latthe stest DISTRIBUTION

imimwhole arrawhole arr

ypotheses toypotheses ton that criticsn that critics

y at a fewat a fewin theth

oupup

socsoive psive pswithin a thin a and mediatand mediatperception, andperception, anhips. The theorps. The theo

ctionism, arguingonism, arguingogical and that ogical and tha

mplexities of inmplexities of shared, group-bshared, group-b

mental to racismntal to racismh more to say ah more to say a

e theory, but we eory, but we let us merely nolet us merely no

t the theory:the theo

not individualnot individejudice, andudice, and

ii

11:31:06:01:10

Page 19

Page 19

structure that determined the character ofintergroup relations.

4. It proposed that it was the cognitive interpret-ation of intergroup relations that shaped inter-group and individual behavior and that suchinterpretation was a collective, ideological, cog-nitive activity, not an isolated individual one.

5. It illustrated how processes that impinged on asocially shared, group self could affectbehavior. This last point—distinguishing a col-lective self from the self-concept in general—contained within it a powerful lever for the nextstep forward.

The Emergence of SCT

From 1971 onwards Turner was interested in theimplications of the minimal group paradigm forpsychological group formation. By 1978 this hadbecome his central research question. What was apsychological group? How did they form and whatwas the effect of their formation?

In 1978 the European Laboratory of SocialPsychology (LEPS) held a conference on socialidentity at the University of Rennes in France.Turner wrote and presented a paper “Towards acognitive redefinition of the social group” thatsummed up ideas he had been developing about thenature of the psychological group. This paper, sub-sequently published as Turner (1982) but also as aninvited paper in a French journal in 1981, was thebeginning of self-categorization theory (SCT), thenext stage in the social identity story.

As mentioned, SCT is a different theory fromSIT. It is not an extension or derivation, but in factis a more general account of the self and groupprocesses than SIT was ever intended to be. Theterm “SCT” did not appear in Turner (1978) but thefact that it was a new explanation of a differentproblem was obvious to everyone working withTajfel and Turner at the time. The confusion hasbeen that not everything that uses the words “socialidentity” is part of SIT. So why was it a new the-ory? And where did it come from? There have beenthree main steps in the creation of the contempor-ary theory. In the first in 1978, the distinction

between personal and social identity was made andthe hypothesis that social identity was the basis ofgroup processes was elaborated (Turner, 1978,1982). In the second in 1982–1983, whilst Turnerwas at the Princeton Institute of Advanced Studies(IAS), the distinction between personal and socialidentity was elaborated into the notion of levels ofself-categorization and the theory was formalized(Turner, 1985). In the late 1980s and early 1990s,working on the self-concept and stereotyping (thelatter with Penny Oakes and Alex Haslam), thehypothesis of self-categorizing as the activation offixed cognitive structures was rejected in favor ofself-categorizing as a reflexive process of socialcontextual judgment (Turner, 1988; Turner, Oakes,Haslam, & McGarty, 1994). This new dynamicperspective enabled a solution to the problemof the validity of stereotyping in terms of theveridical-because-contextual nature of self-categories and a new understanding of the nature ofself and its role in socializing and motivating cog-nition (Oakes, Haslam, & Turner, 1994; Turneret al., 1994; Turner & Oakes, 1997; Turner & Ono-rato, 1999). A new synthesis has been made pos-sible in relation to Tajfel’s original problem ofprejudice and theoretically related phenomena ofsocial change, power, personality, and the relativityof cognition.

The problem that stimulated SIT was why didsubjects discriminate in the minimal group para-digm? The theory went on to explain the condi-tions under which groups acted to change theirintergroup attitudes and actions in society. It was atheory of intergroup conflict, ethnocentrism, andsocial change and the main psychological hypoth-esis was that people sought to achieve or maintainpositive ingroup distinctiveness to gain a positivesocial identity. SCT addressed a different questionin the minimal group paradigm—why did subjectsidentify with the minimal groups at all and whydid they act as if they had group identities thatmattered to them? More generally, the questionwas how does psychological group formation takeplace and, indeed, is there actually a group processthat is psychologically real and distinctive, irredu-cible to the psychology of individuals?

The key hypothesis was that group processes


NOT

group grouinterpretainterpretanitive activititive activitIt illustrated hot illustrated h

cially shared, ally shared, vior. This last povior. This last po

elf from the srom the within it ith FOR

werfwerf

interested in interested inparadigara

7878 DISTRIBUTION

orors hads had

t was at wwhatwhat

selfelfnition nition et al., 199al., 19rato, 1999). rato, 1999). sible in relatiosible in relatio

ejudice and theodice and theol change, power, change, power

tion.ionblem that stimulam that stimul

minate in the mminate in the y went on to exwent on to e

groups acted to groups acted to actions in societctions in socie

ict, ethnocentrisict, ethnocentrissychological hyphological hy

hieve or maintaihieve or maintaigain a positiveain a positive

ent questiont questionsubjectssubjects

whywhyhat

11:31:06:01:10

Page 20

Page 20

emerged from a shift towards defining the self as asocial category, in terms of social identity, from theself as an individual person, in terms of personalidentity. What mattered was not the need for posi-tive social identity but the processes of depersonal-ization that emerged from self-categorization. Howone defined oneself, not positive interpersonalrelations, was basic to the group as a psychologicalprocess. This is a theoretical point, not a simpleempirical one. Of course interpersonal relations areoften related to group formation, but they were notconsidered necessary or sufficient, since what mat-ters is their relationship to the key psychologicalprocess of social identification (Turner, 1982,1984; Turner, Hogg, Oakes, & Smith, 1984). Thesepoints are explained in some detail in Turner andBourhis (1996). The distinction between personaland social identity and the specification of itsrelevance to the causal production of collectivepsychology was the beginning of the new theory.

After the great advances of the group dynamicstradition, inspired and led by Lewin and aided byother giants of Gestalt-influenced social psych-ology (Asch, Sherif, Festinger), North Americangroup research had gradually regressed to a neo-behaviorist individualism in which the concept ofgroup became superfluous because it had ceasedto have any explanatory power. By the 1970s,researchers were asking “whatever happened tothe group in social psychology?”. The answer wasclear—it had been reduced to a collection of indi-viduals interacting to satisfy personal motives andself-interest who had thereby become cohesiveand mutually influential. Every process (of inter-personal attraction, cooperation, and mutual influ-ence) when analyzed turned out to be explicableapparently as individuals affecting individualswithout any suggestion that the formation of agroup relationship could qualitatively change theseinterpersonal relations, that the relations betweengroup members could be causally affected by theirmembership in a joint unit. Indeed such sugges-tions were frowned on as unscientific. It is hardto believe now that in the 1960s and 1970s, talk ofpsychological products and processes such asgroup memberships or social norms being “shared”was dismissed by textbooks as poetic license at

best and the group mind at worst (Tajfel’s writingsstood out for his continual emphasis on the sociallyshared uniformities of attitude and behavior). Theproblem was that these interpersonal theories werenot actually supported by the data. They persistednevertheless because they fit the dominant ideol-ogy, the ideology of individualism.

One of these disconfirmations emerged fromsocial identity research. Why did low status groupsnot simply fall apart psychologically? Why did themembers not simply try to move upwards infantasy if they could not objectively? How didlow status, subordinate groups, lacking inresources, prestige, and power, manage to hold onto the loyalty of their members so that they couldwork to change the status quo? On closer examin-ation of the group cohesion literature (e.g., Lott &Lott, 1965), one found that supposedly they couldnot! People were attracted to groups that mediatedrewards for them. No rewards, no group! But thedata showed that groups did sometimes becomemore cohesive and unified as a function ofdeprivation, derogation, and defeat, contrary toreinforcement theory (Turner, 1981; Turner,Sachdev, & Hogg, 1983; Turner et al., 1984). Also,other data looking at the effects of social categor-ization or group formation on interpersonalrelations implied very strongly that group forma-tion was a psychological process that actuallychanged things. People were not just individuals,and being a group member was psychologicallydifferent (just as Sherif had highlighted yearsbefore). A new theory was needed and what form itcould take was evident in the minimal group para-digm and the interpersonal–intergroup continuum(see Turner, 1982; Turner & Oakes, 1989).

What did the theory say? The essence of Turner(1978) was that psychological group formationwas a matter of social identification rather thangroup cohesion. People became a group not insofaras they developed positive interpersonal attitudeson the basis of mutual need satisfaction but insofaras they defined themselves in terms of a sharedsocial category membership. A shared social iden-tity emerged on the basis of cognitive criteria suchas shared fate, shared situation, or shared attributes(positive or negative). Turner hypothesized that


NOTtivtizationizatioone deone defirelations, warelations, wprocess. This irocess. This mpirical one. Of mpirical one. Of

related to groted to grored neceed

r FORry or ry or

tionship totionship tocial identicial identifificatcat

gg, Oakes, & SmOakes, & Sd in some detail n some detail

istinction betweeistinction betwethe specihe specifificatica

duction otiof thf t DISTRIBUTION

olleolleew theory.ew theory

oup dynamicsoup dynamicsnd aided bynd aided by

al psych-psychericanca

eo-eo-

depdereinforeinfoSachdev,achdev,other data loother data lization or grization or g

lations implied ations impliedwas a psycholas a psychol

d things. People things. Peopla group membea group memb

t as Sherif hadas Sherif hadeory was neededory was needed

ent in the miniment in the minimonal––intergroup intergroup

& Oakes, 1989).& Oakes, 1989).The essence of Tuhe essence of

group formatiogroup formation rather thanon rather th

p not insofarnot insofarattitudtitud

11:31:06:01:10

Page 21

Page 21

people defined themselves as either individual per-sons or as social categories, in terms of personal orsocial identity, and that as their self-perceptionshifted from personal to social identity they wouldperceive themselves as the relatively interchange-able members of a shared social category. Thisprocess of depersonalization or self-stereotypingcould be used to explain how the fundamentalgroup processes emerged from social identity.Thus as social identity became salient and peopledefined themselves in terms of the same sharedidentity, they would tend to see themselves as morealike in terms of the defining attributes of the iden-tity, giving rise to group-based attraction reflectinggroup-derived similarities rather than attraction topersonal characteristics (the group-based attractionversus personal attraction distinction). Similarlycooperation and mutual influence between mem-bers reflected shared social identities. A criticalpoint here is that this formulation allows one to seegroup formation as an adaptive process that makesgroup behavior, cohesion, cooperation, and influ-ence possible, and thus does not merely followbut actually enhances the chances of successfullyreaching goals.

The paper did various other things such asreviewing the evidence for the role of social cat-egorization in group formation, the effects of thelatter, and the idea that social identity varied insalience in a highly situation-specific way. All theseideas have since been pursued with great vigorempirically and have become widely acceptednotions in the science. Is there anyone left in socialpsychology who has not heard of the personal ver-sus social identity distinction and does not knowthat the salience of the latter changes behavior inpredictable ways to make it more collective andgroup oriented? People now think this is commonsense, and yet before 1978 the notion that socialidentity was fundamentally implicated in the verynature of group behavior was unknown. And evennow it is not properly understood, it being widelythought, for example, that ingroup identificationis just another individual difference variable ratherthan a process that, under given conditions, worksto eliminate individual differences in a givensituation.

Early Applications: Social Influence,Group Polarization, the Crowd, SocialCohesion, and the Problem of Salience

In 1978 Turner applied and received funding forresearch on the new theory and his research groupof research assistants and PhD students (Wetherell,Smith, Reicher, Oakes, Hogg, Colvin), some ofwhom played both roles, began applying the ideasto different fundamental areas. Some members ofthis group were already at Bristol (Oakes, Reicher,Smith), whereas others arrived from elsewhere(Colvin, Hogg, Wetherell). The initial research (upto the formalization of the theory in 1982–1983)looked at social influence from different angles (inthe Asch conformity paradigm, group polarization,influence within the crowd), psychological groupformation and the distinction between personal andgroup-based attraction (trying to show how groupcohesion was a function of social identificationrather than interpersonal attraction), and the prob-lem of the salience of social categories. In 1982(the same year that Tajfel died) Turner beganworking on a more systematic statement of thetheory that resolved some of the basic conceptualand empirical issues that had emerged (Turner,1985). Then, having moved to Australia after hisyear in the USA, he prepared the book that wouldpresent the theory and allow his PhD students topresent their research on the theory (Turner, Hogg,Oakes, Reicher, & Wetherell, 1987). To understandthe work from 1978 on, it is best to summarizesome key issues in the new statement of the theoryand then briefly summarize the research directionsas they emerged.

Levels of Self-Categorization

In the year Turner spent at IAS Princeton (1982–1983) he had an opportunity to think about howto conceptualize the categorization processes atwork in the personal–social identity distinction.Finding the work of Rosch and her colleagues par-ticularly useful, he recast the relationship betweenpersonal and social identity as one of differentlevels of self-categorization. There were several


NOTceive ceive

able membeable membeprocess of depprocess of dep

uld be used to ld be used top processes emprocesses em

social identity bsocial identity bmselves in teves in te

would tenouhe FOR

o seeo seefifining attribning attri

up-based attractup-based attractes rather than atther than

e group-based atgroup-based atstinction). Simistinction). Sim

ce between mebetween mies. A cA

s os DISTRIBUTION

alalo seeo see

makesmaininflflu-u-

ow

ratlem olem (the samhe samworking onworking ontheory that resotheory that resnd empirical issmpirical iss

). Then, having ). Then, havingthe USA, he prehe USA, he pre

theory and alloeory and alloesearch on the thesearch on the t

& Wetherell, 198Wetherell, 1988 on, it is best 8 on, it is best

new statement ofw statement oe the research die the research di

98282–ow

11:31:06:01:10

Page 22

Page 22

critical points. Self-concepts were self-categories;self-categorizations were organized hierarchicallyby means of class inclusion and the different levelswere functionally antagonistic in their perceptualeffects at any given time but nevertheless mutuallydependent. Lower order self-categories wereformed inter alia from social comparisons withinhigher order ones, and higher order ones wereformed inter alia on the basis of lower level ones.Self-categorization and social comparison wereprocesses that required each other: All categoriza-tion reflected comparison and all comparison tookplace between stimuli categorized as having anidentity at some higher level.

These notions were formalized carefully andparticular hypotheses derived from them. Animportant point is that they explain how personaland social identities can be salient at the same timeand still be distinct and antagonistic (see Turner,Reynolds, Haslam, & Veenstra, 2006). A great dealof thought also went into the problem of salienceand the principles of meta-contrast and relativeprototypicality, which are explained in more detailshortly. In addition, a focus was developing thedetailed explanation of specific phenomena interms of self-categorization processes, especiallygroup formation and cohesion, cooperation, andsocial influence. The notion of meta-contrast andthe idea that different individuals would be more orless influential within a group as a function of theirrelative prototypicality emerged from the attemptto explain group polarization from within the newtheory whilst Turner was at the IAS.

The problem of salience emerged as soon asone understood that group behavior was based onsocial identity and that the influence of socialidentities was situation specific. Penny Oakes andTurner decided to address it in Oakes’ PhD. Thefirst step was adapting Bruner’s (1957) formula ofrelative accessibility × fit to describe the conditionsunder which a stimulus was captured by a category.Bruner argued that certain categories would behighly accessible (or likely to be activated) as afunction of contextual factors and the currentgoals, needs, and purposes of the perceiver. Turnerand Oakes, in thinking about “fit”, never thought interms of the isolated attributes of some stimulus

and the match between these features and categoryspecifications. They were influenced by Tajfel’s(1957, 1969) accentuation theory and in particularthe Tajfel and Wilkes (1963) study showing per-ceptual accentuation effects. When the line lengthof eight lines and the category membership werecorrelated (all short lines were in category A andall longer lines were in category B), the judgeddifference between the shortest of the longer fourlines and the longest of the shortest four lines wasexaggerated. There was an accentuation of similar-ities within the classes and differences between theclasses on length. Thinking of the fit of socialcategorizations in terms of peripheral–focal correl-ations between group membership and members’positions on some response dimension led Turnerand Oakes to distinguish between comparative andnormative fit. They originally defined normativefit as the degree to which perceived similaritiesand differences between group members correlatedwith group memberships in a direction consistentwith the normative meaning of the group identities(e.g., men and women differ in relation toindependent–dependent characteristics). Definingcomparative fit (or the degree of peripheral–focalcorrelation) arose from Turner’s attempt atPrinceton to provide a quantitative principle thatwould allow the prediction of when and whygroups would polarize as a function of individuals’pretest views on any issue in any given context.

Turner was supervising Margaret Wetherell’sPhD, trying to find a social identity explanation ofpolarization. He knew that one could explain grouppolarization as convergence on the normativeposition of the group if one could explain whygroup norms sometimes were group averages andsometimes were more extreme. Turner was con-vinced that there had to be some way of explainingnorm formation in terms of social categorizationprocesses that predicted how and why ingroupprototypes formed where they did. He tried to findsome way of understanding the peripheral–focalcorrelation between individuals’ group member-ship and their responses on a dimension so that hecould predict exactly which person or positionwould become most prototypical of the group as awhole and when that prototype would or would not


NOTeffedependepenformed formedhigher ordehigher ordformed or inter anter a

elf-categorizationelf-categorizatioesses that requthat requ

ected coectwe FOR

arisoarisostimuli catstimuli ca

e higher level.e higher level.were formalizee formali

es derived fros derived frothey explain hothey explain ho

e salient at the salient at thgonistic nis

2000 DISTRIBUTION

e Tue TuA great deA great d

m of saliencem of salienceand relativeand relative

ore detaile detailng theth

inin

(e.(eindepeindepcomparatmparacorrelation) correlation)Princeton to prPrinceton to p

ould allow the ould allow theps would polarizwould polariz

views on any issuviews on any iswas supervisingwas supervisin

fifind a social idend a social idknew that one couew that one co

nvergence on tnvergence on tif one could exone could e

were group averawere group averame. Turner wasme. Turner wa

way of explaininway of explainicategorizationcategorizatwhy ingrouphy ingroup

ed to to fifi

11:31:06:01:10

Page 23

Page 23

be polarized. Eventually Turner succeeded byinventing the principle of meta-contrast (Turner,1985; Turner & Oakes, 1986, 1989). A collectionof individuals tend to be categorized as a group tothe degree inter alia that the perceived differencesbetween them are less than the perceived differ-ences between them and other people (outgroups)in the comparative context. Ingroup–outgroup cat-egorization occurs when the differences perceivedbetween categories are larger than the differenceperceived within categories. Furthermore, any spe-cific person tends to be seen as more prototypicalof the group as a whole, to the degree that theperceived differences between that person andother ingroup members are less than the perceiveddifferences between that person and outgroupmembers.

The same principle that predicted categorizationof stimuli also predicted the relative prototypicalityof instances of any class. This analysis provided aconcrete refutation of reductionism and individual-ism in social influence and transformed the persua-sion–norm relationship. People did not producenorms from influence but were able to persuadebecause they embodied norms, and the normexpressed the identity of the group as a whole incontrast to other groups, not an individual property,averaged or summed, but a Gestalt property of themembers in a context. The principle generalizes toall categorization, not merely social categorization(Turner et al., 1994).

Research based on this conception of anidentity-based influence process stimulated thedevelopment of a more formal analysis of socialinfluence (Turner, 1985, 1991). SIT understood therole of group influence processes in intergroupbehavior and social change but did not provide anew theory of influence. SCT did, however. Build-ing on the insights from meta-contrast and proto-typicality, one can see that those that embody thegroup norms and are most representative of thegroup will be most persuasive or influential. It isthis expectation that certain others should perceivethe world in the same way as self (perceived differ-ence within and between categories) that makesthese others a valid source of information aboutreality (Moscovici, 1976; Turner, 1991). There is a

shift away from a distinction between physical andsocial reality testing and ideas of normative andinformational processes specified by Deutsch andGerard (1955) to a single process of influence—referent informational influence (Turner, 1982).Individuals always act in ways informed by under-standings of how similar others would respond inthe same situation.

Self-Categorization as RelationalFlexible Judgment

Understanding meta-contrast, prototypicality,and polarization led to the next big step. If self-categories were inherently comparative then theywere infinitely variable, contextual, and relative.This meant that they could not be stored as fixed,cognitive structures in some mental system beforethey were used, waiting to be prodded into action(as Turner, along with many others, had thoughtoriginally). The cognitive structural view of stereo-typing and the self-concept had to be abandoned.It also meant that the very variability of self-categories was not an indication of distortion andbias but the very opposite—an indication of theirveridicality and orientation to reality. The com-parative flexibility of self-categories arose becausethey were contextual representations of people.They were selective and variable because theframes of reference within which people definedthemselves were always changing. If reality keepschanging and self-categories reflect reality, thenthey must change and be flexible in order to beaccurate. Moreover the basic principle at workhere, of meta-contrast, is a fully rational one. Thesalience of self-categories is a function of motives,expectations, knowledge, and reality working increative synthesis. Self-categorizing is variable,flexible, and selective, is based on past experience,knowledge, and theories, and is reality oriented.These notions have big implications for categor-izing, the self-concept and stereotyping, the valid-ity and relativity of perception, and for the natureof human cognition.

The idea of the self-category as an on-the-spotjudgment was also facilitated by the work of Medin


NOTdegredegr

between thebetween theences between ences between

the comparativehe comparativzation occurs whion occurs wh

categories are lcategories are within categorin categori

nds to bndsa FOR

een aeen aole, to theole, to the

s between thats between thatare less than theess than th

person and ouperson and o

ted categorizatd categorizprototypitot

prp DISTRIBUTION

tytyded aded a

vidual-vidersua-rsua-

ce

(as asoriginaorigintyping antyping aIt also meaIt also meacategories was categories was

as but the very out the very oicality and oriencality and orie

flflexibility of selfexibility of secontextual reprentextual repr

elective and vaelective and vce within whichce within whi

ays changing. Ifays changing. Ifgories reries reflflect reect re

flflexible in ordeexible in ordc principle at wrinciple at

rational one. Thrational one. Thon of motives,on of motives,

working inorking invariable,variable,

ence,nceed.

11:31:06:01:10

Page 24

Page 24

and his colleagues (e.g., Medin & Wattenmaker,1987), showing that categories are expressions oftheories and knowledge about how things gotogether rather than simple similarities, and byBarsalou’s (1987) work on the variability of proto-typicality judgments, which argues against con-cepts as fixed mental models. In general, SCT andthe way it deals with salience has been much influ-enced by many significant figures, not merelyTajfel, but also Lewin, Asch, Sherif, Rosch, andMedin. The reworking of self-categorizing as aprocess of reflexive and comparative social judg-ment rather than as the activation of pre-existingstored self-concepts meant that the idea of relativeaccessibility based on Bruner had to be replaced byperceiver readiness, since the former implied that astored, ready-made category was more or less closeto usage rather than being created in the processof use as and when needed (see Oakes, Haslam,& Turner, 1994; Turner et al., 1994; Turner &Onorato, 1999).

None of this implies that self-categories varyarbitrarily unconstrained by continuities of reality,experience, motives, and knowledge. On the con-trary, it is the fact that there is a systematic way ofrelating such variation to changes in reality thatenables the theory to argue for the veridicality ofself-categorizing, that relativity is not relativism(Turner & Oakes, 1997). The vantage point of theperceiver varies because individuals and groups arecontinually interacting and these interactions haveimplications for the self-categorization processes(and emergent similarities and differences) as anindividual and group member. A flexible, relationalself-process allows for this variation to be repre-sented and acted upon psychologically.

Major Elaborations and ThematicDevelopments

The story of social identity so far and the theor-etical elaborations and developments it offers cap-ture the interplay between psychological processesand the socially structured system of relations inwhich individuals and groups are defined and func-tion. This perspective provides answers to the

quintessential problem of social psychology: themind–society interaction. Individuals, groups, andintergroup relations exist objectively. People aredefined in society as members of these groups(e.g., religion, gender, ethnicity, age, class, roles,and so on) that have a particular history and socialrelationship. They have a social location that isshared with others. It is recognized that thesegroups have a psychological aspect; the norms,values, beliefs, and ideologies are socially trans-mitted through influence and internalized, funda-mentally affecting one’s psychology—creatingsocially-shared regularities that affect the content,structure, and functioning of the mind (see Turner& Oakes, 1997; this volume).

In this view, social psychology is not the exten-sion of general psychological processes to socialstimuli or social contexts—it integrates social andcultural patterns and products with psychologicalexplanation. The emphasis on social structure andlarge-scale social processes, the distinctionbetween personal and social identity, the recogni-tion of different levels of self-categorization andinterdependent higher and lower order selves,and the flexible, variable nature of the self-categorization process are the building blocks thatgive substance to this alternative interactionist view.

This alternative view and its implications for thefield perhaps can be appreciated more fully whenexamining its specific and unique contribution inmajor thematic areas. What follows can only be abrief overview. There is so much to say in terms ofthe progress that has been made, so much detailthat is needed to explicate the larger ideas, and somuch research still waiting to be done (see the con-cluding chapters in this volume). In the remainderof this chapter four thematic developments aredescribed: stereotyping, leadership and power,social change, and personality. These areas(amongst others) also have been major themes ofresearch for Turner and colleagues throughout thelate 1980s and 1990s until the present day. In com-bination they highlight the diversity of topics towhich social identity core ideas are relevant andthe exciting possibilities that the perspective offersin developing a more integrated understanding ofthe person. Some have been a focus for research


NOTBaBtypicatypiccepts as cepts as the way it dethe way it denced by mannced by man

ajfel, but also Lajfel, but also Lin. The reworThe rewo

of reof fler FOR

e ande andn as the acn as the ac

cepts meant thacepts meant thad on Bruner had Bruner ha

since the formernce the formeregory was more oegory was more

g created in thcreated in (see Oake

1 DISTRIBUTION

HaHa; Turner ; Turner

egories varygories varyof reality,realit

e con-onofof

tiotiinterdinterdand thed thecategorizatiocategorizatigive substance tgive substance

This alternativeThis alternativeperhaps can be erhaps can be

ng its specing its spec fific acmatic areas. Whaatic areas. Wh

w. There is so mu. There is so muhas been madehas been made

plicate the largerplicate the largering to be done (sto be done

olume). In the relume). In the reic developmentsc developmen

rship and powehip and powThese areasThese ar

r themes ofthemes ofghout thout th

11:31:06:01:10

Page 25

Page 25

over many years and some over the past 5 years orso. As highlighted by Haslam et al. (the two con-cluding chapters of this volume), though, over thesame time period there are other research groupsand research areas where SIT/SCT is having a sig-nificant impact.

Stereotyping and the Relativity ofPerception

In the 1980s and 1990s a research focus was onstereotyping. In answering the question “Why dowe stereotype?”, the dominant view is because ofshortcomings of the human cognitive system. Itis argued that the cognitive system is limited incapacity and therefore cannot apprehend others asindividuals in all their detail. Categorization, sim-plification, and overgeneralization are functionaloutcomes of these cognitive shortcomings, leadingto others being judged as more similar and alikethan they really are. The core assumption here isthat stereotypes are false and erroneous becausethey come into existence through weaknesses inhuman cognitive abilities.

The SCT analysis of stereotyping is differentfor three main reasons:

1. Stereotypes are not erroneous but represent lifeas group members. It is argued that humanbeings are group members as well as indi-viduals, and it is accepted that there is collectivepsychology as well as individual psychology.When human beings define themselves as groupmembers and act as group members, they takeon shared beliefs, goals, and attitudes thatdefine “who we are” and “who we are not” in agiven context. They build up knowledge,experiences, and theories as group membersabout groups and group relations. They areperceptually ready to perceive the world in par-ticular ways. Stereotypes reflect these groupproperties (e.g., norms, values, beliefs), whichcannot properly be understood by looking at theindividual group members in isolation or asaggregates. Stereotyping first and foremost isunderstood as an outcome of these groupprocesses.

2. Stereotyping is tied to the structure of inter-group relations. It is recognized that peoplehold the attitudes that they do toward the mem-bers of particular groups because they are in aspecific kind of relationship with those groups.People are embedded in social structures, theyhave a particular social position (e.g., high orlow status), they have qualities that have certainexisting social meaning (e.g., gender), and theyoccupy certain social roles that denote status andvalue (e.g., men versus women). To understandstereotyping one must look at the social–struc-tural realities of the intergroup relationships atwork in any system of human relations (i.e.,who is advantaged and disadvantaged, who hasstatus and who does not, are group goals com-petitive or cooperative, do the groups haveshared interests or not).

3. Collective theories about intergroup relationsare central to stereotyping. The impact of thestructural reality of intergroup relations ismediated by collective theories, ideologies, andbeliefs about group life (e.g., what it means tobe a man versus a women). People act in termsof intergroup relationships as mediated by theircollective definition, explanation, perception,and understanding of their situation and theirrelationship to others.

There have been two main responses to thesearguments. The first is that stereotyping is distortedby the goals, values, motives, ideologies, andexpectations of the perceiver and therefore cannotbe useful or valid. In SCT, though, it is argued thatall perception is relative to the perceiver andaffected by his or her knowledge, motives, andexpectations, and in particular those that come tothe fore to define the self in a given social context.An essential distinguishing feature of the SCTanalysis is that both individuation and stereotyping(and all perception) are outcomes of self–other cat-egorization processes (see the concluding chaptersof this volume). The shift from individuation tostereotyping is simply a shift in the level of self-categorization from personal identity to socialidentity.

The other main misunderstanding is that the


researeseiifificant impacant imp

eotyping and ttyping and tiontion

and 1990nd answFOR

rereng the qng the q

dominant viewdominant vhuman cognitivecognitiv

ive system is lisystem is liot apprehend otht apprehend oth

ategorization, sigorization, are functiare func

ings, ng DISTRIBUTION

dingdingd aliked alihere ishere is

usememebeliefsbeliefbe a man be a manof intergroupof intergroupcollective delective defifi

d understandingd understandinonship to others.nship to others

een two main een two mainrst is that stereotyst is that stereo

es, motives, idees, motives, ideceiver and therefver and theref

though, it is arguthough, it is argu the perceiver he perceiver

ge, motives, ange, motives, ane that come tothat come to

cial context.al context.he SCThe SCT

ypingpingat-

11:31:06:01:10

Page 26

Page 26

SCT analysis of stereotyping essentially embracesand supports relativism. The idea here is thatbecause stereotyping and other forms of cognitionare argued in SCT to represent features of realityfrom the vantage point of the perceiver, the stereo-types that people hold must be equally valid. SCTdoes argue for relativity but not relativism. Socialinfluence and consensus at higher levels of self-categorization can correct for relativism by seekingagreement across individual and group perspec-tives. There are social mechanisms to validate andinvalidate certain views. Australians, for example,through consensus, discussion, rules, norms, andlaws, influence and shape which perceptions andactions are considered right and valid. These socialarrangements can vary (or stay the same) as a func-tion of disagreements and arguments about what iscorrect and true and attempts to seek change or not.As a function of these processes, for example,there have been changes in gender stereotypes andassociated laws regarding equal opportunity anddiscrimination.

The central message is that stereotypes are notrigid or erroneous but reflect perceptions of grouprelations from the perceiver’s vantage point. Asothers have argued, the one way you can alwaysfind “erroneous” stereotypes is to look from outside.Often, what it means is “we don’t agree with you”.Agreement and disagreement between individualsand groups are an outcome of life as individualsand group members. Acceptance or rejection ofstereotypes is informative about the nature ofintergroup relations and can motivate social changeand the emergence of new shared understandingsof what is acceptable and unacceptable.

Social Influence, Leadership, and Power

Beginning in the 1990s also was a more specificapplication of the SCT analysis of social influenceto the issue of leadership and then later power. SCTdoes not argue for monolithic conformity where allgroup members will necessarily agree and be inter-changeable in their views. Group members can anddo disagree and they discuss, argue, and exchangeviews. In groups, members can and do leave, create

schisms, and come to new understandings of “whowe are”. The SCT analysis of social influence rec-ognizes that challenging people’s current under-standing of the world, the creation of uncertainty,and the resolution of disagreements occur amongstthose where at some level there is perceived psy-chological similarity. Disagreement with outgroupmembers, on the other hand, reinforces self–otherdifferences and affirms one’s views as right andcorrect.

There is disagreement because there is anasymmetry within the structure of the group iden-tity (see discussion on meta-contrast). There arepeople who will be considered better representa-tives of the group, a “truer” group member as afunction of these processes. Schisms partly relateto boundaries being drawn amongst certain groupmembers in the interests of making claims aboutwho best embodies the norms, values, and beliefsof the group as a whole. There are more central andmore peripheral members, people who betterembody definitions of “who we are” compared to“who we are not”. In this way, it is possible toargue that social influence occurs in groups andthat one or more people will be more influentialthan others.

The SCT analysis has direct implications for theunderstanding of leadership (Turner & Haslam,2001) and power (Turner, 2005; Turner, Reynolds,& Subasic, 2008). The idea that group membersare more or less representative of a group meansthat leadership is distributed in groups and thereis no clear divide between leaders and followersin terms of some leadership characteristic or“essence”. The implication is that leadership suc-cess and failure fundamentally depend on groupidentity and being able to shape and create sharedunderstandings of “who we are” and “what we do”.In this analysis there is recognition of the relation-ship between leaders and other group members in abroader context of group relations.

When it comes to the concept of power, tensionsremain between the SCT analysis of influence andsocial-psychological explanations of power. Putsimply, power as defined by control over resourcesis purported to override social influence processesas described in SCT and related work. The


NOTfroftypes typesdoes argudoes argininflfluence anuence acategorization categorization greement across greement across

There are socre are sote certaie c

n FORiews.iews

us, discussus, discuse and shape wh and shape wh

dered right and vright and ry (or stay the say (or stay the sa

nd arguments abnd arguments abpts to seek chans to seek ch

cesses, fseerer DISTRIBUTION

examexameotypes aneotypes a

portunity andportunity and

s are notare notgroupou

AsAs

““wwargue arguethat one at one than others.than others

The SCT anahe SCT annderstanding of derstanding of

) and power (Tuand power (Tusic, 2008). The sic, 2008). Th

r less representar less represenp is distributed p is distributed

e between leadbetween leadleadership chaleadership chaation is that leadn is that lea

ntally depend ontally depend oape and create shpe and create

and nd “what we doat we dof the relation-of the relati

members in aembers in a

11:31:06:01:10

Page 27

Page 27

traditional interdependence model of power arguesthat control over resources leads to power (liking,influence, dependence) and power leads to influ-ence. Turner (2005) in his three-process theoryof power turns this process on its head. The SCTanalysis of power argues that the foundation ofpower is getting others to carry out one’s will—theexercise of such power through people is the ultim-ate resource (and allows groups to harness thethings/resources they need to achieve their goals).Persuasion, authority, and coercion are outlined tobe the three processes of power. All three rest onidentity and the influence processes that flow fromit in the following ways:

1. Persuasion emerges from the perception thatpeople share an ingroup membership: there areshared norms, values, beliefs, and goals. Theleaders and followers therefore willinglyengage in activities that will further theirgroup’s goals and aspirations.

2. Authority emerges from ingroup membershipand structure where there is the perception thatthe system through which power relations aredefined is legitimate. The leaders and followerswillingly engage in activities because theyaccept the relevant roles and responsibilities inthe group.

3. Coercion emerges when there is not a sharedingroup membership between parties and assuch there is a need to bring about compliancethrough one’s capacities to provide positive andnegative outcomes. Coercion also requires per-suasion and authority over those who bringothers (against their will) to act in particularways.

Power as coercion is argued to be the weakest formof power as it reveals a lack of genuine ability toinfluence through group identity. It also is likely tocreate further disidentification, weakening persua-sion and authority and making these less likely toemerge in the future. Thus, leaders, leadershipgroups, institutions, elites, and authorities have realsocial power when they are able to achieve actionthrough influence and persuasion rather thancontrol.

It is through the SCT analysis of social influence,

leadership, and power that it is possible to under-stand concretely that SCT is a different theory ofthe group from social interdependence, where it isliking and dependence that underpin group forma-tion. In the three-process theory it is argued thatidentity confers power and it is not grounded inmaterial resources. Power is an emergent propertyof specific social and psychological relationsbetween people. The value of resources isindeterminate outside of the framework of a sharedsocial identity (e.g., who and what is valued), andonly within it can leaders persuade others or exer-cise legitimate authority over them. Even coercionultimately depends on this dynamic: Coercion willnot succeed if a leader cannot rely on others to dotheir bidding. Power is always socially constrainedand conferred and is not separate from influencebut is itself evidence of influence at work.

From Prejudice to Social Change

SIT and SCT came out of the prejudice traditionwithin social psychology but it has been necessaryto recast the paradigm in order to understandprejudice and social change. The traditionalapproach, referred to by Turner in the FreilichFoundation Eminent Lecturer Series in 2001 as the“prejudice orthodoxy” (Turner, 2001), argues thatthe causes of prejudice are:

1. Specific quasi-pathological personalities orindividual difference factors that, more or less,directly predispose people to generalized preju-dice against vulnerable minority (and other)outgroups.

2. General psychological processes that, oncetriggered by some stimulus, state, or event,inevitably and automatically lead to hostilitytowards outgroups (e.g., frustration, limitedcognitive capacity, group formation, andethnocentrism).

3. The meaninglessness of intergroup attitudes andstereotypes where, as outlined above, any groupjudgment or stereotype, any social categoricalrepresentation of a person or group of people, issomehow faulty, biased, and misleading.


NOTpower powe

analysis of analysis of power is gettingpower is getting

ercise of such porcise of such pesource (and alource (and al

sources they neesources they neauthorityhority, andan

rocessesocefl FOR

powepowce processece process

ys:ys:

from the percepm the percmembership: themembership: the

fs, and goals. Tand goals.efore williefore wil

furtheurt DISTRIBUTION

yyheirheir

bershipbershipthath t

FromFrom

SIT and SSIT and Swithin social pwithin social o recast the pao recast the p

udice and socdice and socch, referred to referred t

on Eminent Lectn Eminent Lecrthodoxyhodoxy” (Turn (Turn

ejudice are:ejudice are:

athological perathological perfactors that, moactors that, mo

ple to generalizedple to generalizeminority (and otority (and

ess that, once that, once, or event,or event,

hostilityhostilitymitedmited

nd

11:31:06:01:10

Page 28

Page 28

4. The passivity of cultural and social learningwhere, through socialization, prejudice islearned from one’s society. The assumptionseems to be that what culture teaches, one hasno choice but to learn (e.g., automatic process-ing and implicit attitudes).

SIT/SCT and related work over many yearstogether serve as a detailed rejection of thesecauses. More recently, the focus of thinking andresearch has been on how prejudices conceptual-ized as a function of personality, general psych-ology, deficit psychology, and/or socializationdirectly impact on social psychological theories ofsocial change. If the causes of current prejudicesare unavoidable and deeply rooted in our psych-ology (even our unconscious), then it is difficult toenvisage the possibility of real and genuine socialchange. Is it any surprise that within social psych-ology over so many years there has been a focus onexplaining system stability and maintenance of thestatus quo rather than processes of social change?(Turner & Reynolds, 2003).

SIT and SCT offer a different view where preju-dice is not inevitable and social change and socialprogress are possible. Prejudice is an outcome ofpeople’s understanding as group members of a par-ticular intergroup relationship. In this intergrouprelations perspective, it is necessary to take intoaccount people’s group memberships, the real rela-tionships between them, the social structure withinwhich they function, the ideologies that they use todefine themselves, and the social system and theirrelationships. This analysis means that if grouprelations change through social action, if there is achange to people’s social relationships, then preju-dice towards particular groups can also change(and this change can be represented in all aspectsof cognition—in implicit and explicit processes). Itis possible through forms of social organization,conventions, and practices to impinge on, modify,and change people’s group identities and associ-ated social norms and affect how they think, feel,and are likely to act.

These ideas offer a much less pessimistic viewof humanity and human social relations. Funda-mental to social change is the degree to which the

current structure of relations (e.g., dominant andsubordinate, high and low status, advantaged anddisadvantaged) is widely accepted or rejected. Insocial identity theory and related work there hasbeen a focus on legitimacy, stability, and permea-bility as factors that account for system stabilitythrough social mobility and social creativity andsocial change through social competition andconflict (see Tajfel & Turner, 1979). What has beenmissing is a detailed exploration of the processesby which some action, outcome, or relationship istransformed from one perceived as legitimate toone perceived as illegitimate, and specifically withrespect to social change how a subordinate groupredefines its relationship with the dominant groupfrom legitimate to illegitimate. Over the last10 years or so this has been a particular focus ofSCT work (Turner & Reynolds, 2002). At the cen-ter of this analysis is the recognition that the self ishierarchically organized and that it is possible toshift from intragroup (“we”) to intergroup (“us”versus “them”) and vice versa.

Legitimacy is understood in this work, in linewith the thinking of many authors, as a state ofaffairs in accord with (or not) a rule, a law, a beliefabout what is right, proper, or moral. FollowingSCT, it is action that is in line with or that violatessome norm, value, rule, procedure, right, or obliga-tion associated with a shared ingroup membership,where some perpetrator is in violation of a specificrule that “we are supposed to share”. The conceptof legitimacy itself is intertwined with group iden-tity. A legitimate actor (leader, authority, govern-ment, institution), by definition, embodies a sharedsocial identity; in acting in line with a shared socialidentity, in accord with ingroup norms, identifica-tion with the group is enhanced (along with percep-tions of fairness, trust, leadership satisfaction,endorsement, and influence). Reciprocally, anyactor with whom one identifies is thereby morelikely to be seen as legitimate. When some systemactor behaves legitimately, one is more likely toidentify with and seek membership in it and is morelikely to endorse and justify this group. Underthese conditions, there is system identification andsystem justification.

Conversely, when some system actor behaves


NOT ingin

SIT/SCT aSIT/SCT together serveogether serveauses. More receauses. More rec

arch has been has beena functa f

fic FORof pof p

psychologypsychologon social psychon social psych

the causes of ccauses of d deeply rooted deeply rooted

cious), then it iscious), then it isf real and genureal and gen

t within sthasas DISTRIBUTION

al psal pn a focus on a focus

tenance of thetenance of thecial change?cial change?

preju-ejucialial

with twith affairs infairs inabout what about what SCT, it is actioSCT, it is actiome norm, value,me norm, value

associated with aociated with aome perpetrator ome perpetrato

we are supposed e are supposeitself is intertwintself is intertwin

actor (leader, aactor (leader, ay dey defifinition, embnition, emb

g in line with a shline with a sgroup norms, idgroup norms, id

d (along with perd (along with prship satisfactiohip satisfactio

ciprocally, anyciprocally, aereby morereby more

e systesyste

11:31:06:01:10

Page 29

Page 29

illegitimately, in violation of the shared higherorder ingroup identity “we are supposed to share”,this facilitates its recategorization as an outgroup.Identification with the higher order ingroup isreduced and it is more likely that there will beinterest and sympathy (and ultimately identifica-tion) with those that contrast with, and are inopposition to, the actor. Using more system lan-guage, through these processes one comes to deni-grate and reject the system and its representativesand advance and justify a contrasting (sub-)group.

Along with this analysis there has been anattempt to identify the factors that move systemmembers to interpret, explain, and understand theillegitimacy of their own and others’ experiences(e.g., prejudice) not as an isolated occurrence but asan outcome of the illegitimacy of the system itselfand the dominant actors within it. Four of the cen-tral factors can be summarized briefly as follows:

1. An illegitimate act is interpreted as an inter-group rather than interpersonal experience (i.e.,victims and perpetrators act toward each otheras group members rather than as individuals).

2. An illegitimate act is attributed to the sharedinternal characteristics of the dominant group(i.e., perceiving it to be deliberate, reflecting trueattitudes rather than some accident or mistake).

3. An illegitimate act is explained in terms of a“category essence”, a theory of what makes it asocial entity (we assume that factors affectingthe perceived entitativity of the dominant groupinclude its degree of organization and homo-geneity and the perceived extent to whichillegitimate acts are intended, repeated, andsystematic).

4. The illegitimate “essence” or identity of thedominant group is linked to the social system asa whole through the development and dissemin-ation within the subordinate group of a collect-ive theory or ideology (i.e., the emergence ofradical, subversive beliefs that explain theillegitimate “essence” and power of the domin-ant group in terms of the illegitimate nature ofthe system and become normative within thesubordinate group). At the extreme the veryexistence of the dominant group will come to be

seen as illegitimate, as an irreversible threat tothe very principles it was once assumed to shareand embody.

This analysis of illegitimacy and the emergence ofa shift from a higher order ingroup “we” to “usversus them”, to a definition of “us” that requiresthe removal or even elimination of “them”, pro-vides a more systematic analysis of both socialchange and stability. It helps to explain how systemlegitimacy and justification or system illegitimacyand system rejection (and vice versa) emerge.

The Nature of the Self-Processand the Person

In thinking about stereotyping as a flexible, relativeprocess of self-categorization that oriented the per-son to group realities, it became obvious that therewere more general implications for self and iden-tity. In Turner (1982) there was a view about howthe self-concept was structured and organized thathas been developed and refined within SCT. In facta major theme of SCT has been the reconceptuali-zation of the self. Its major contributions through-out the 1980s and 1990s have been to flesh out whatthis implies for group behavior such as stereotyp-ing, homogeneity, influence, and prejudice. Duringthe past 5 years or so, attention has turned to whatthis implies for individuality, personality, and inparticular for personal identity. Traditionally therehas been a personality model of the self-conceptwhere the personal is the baseline, defining what isreal and accurate. The SCT alternative is that per-sonal identity is just one of multiple aspects of selfthat are all more or less important depending oncontext. The central arguments are that:

1. Self-categorization processes are variable.Individuality, like social identity, is not fixedand stable, but is variable and context depend-ent. It is not a ready-made set of predispositionsor traits, but a flexible definition of the self inrelation to current realities. The personal (justlike the social) self is an outcome of theperson × situation interaction. In SCT the “per-son” in this equation relates to the knowledge,


NOTuced auced

nterest and nterest andtion) with thostion) with tho

position to, the osition to, thee, through these phrough these

d reject the systed reject the syste and justify a justify a

this athfy FOR

ysis ysis factors thfactors t

explain, and uexplain, and uwn and othersd others’’ e

olated occurrencated occurrency of the system iy of the system

. Four of the cFour of thefly as follos f DISTRIBUTION

n inter-n in(i.e.,(i.e.,

her

tityitythe selthe selhas been has beena major thema major themzation of the sezation of the se

t the 1980s and e 1980s and mplies for groupmplies for grou

mogeneity, inogeneity, influenueears or so, attentrs or so, atten

or individuality, or individualitysonal identity. Tronal identity. T

ity model of thety model of thee baseline, debaseline, defifinin

T alternative is thT alternative is tultiple aspects ofple aspects o

ant depending oant depending ohat:at:

variablevariable..fifixedxed

d-

11:31:06:01:10

Page 30

Page 30

experiences, expectations, goals, ideology, andtheories that a person brings to a situation (i.e.,perceiver readiness). The “situation” in thisequation is the events that occur and the waythey are given meaning by the perceiver (includ-ing about the self through processes of com-parative and normative fit). As a function ofthese interactive processes a particular self-categorization will become salient and deter-mine one’s behavior in a given social context.As such, individuality (like group identity) is acontemporary creation and therefore there canbe continuity and change depending on whetherthe factors that underpin self-categorizing staythe same or vary. If there is stability in back-ground knowledge and the situations a personconfronts on a day-to-day basis, one wouldexpect relatively stable self-categorizations to(re)emerge (e.g., “who am I”). If person factorsor situational factors change, however, onewould expect self-conceptions to change also.

2. There are interdependencies between differentlevels of self-categorization. Group identities,beliefs, and outcomes such as norms, values,and goals shape the meaning of individuality.Understanding how we differ from others in aparticular situation is interpreted through thesocial and political processes that shape ourhigher order selves (what is valued, consideredappropriate, and so on, amongst those we con-sider most psychologically similar to us at agiven time). The SCT analysis of social influ-ence is central here because it reveals how atti-tudes, behavior, and cognition can change as afunction of group processes and intergroup rela-tions. In this way the individual and group areinterdependent and personality has to be under-stood as an outcome of both.

3. Social change can bring personal change.Along with social change comes the potentialfor change in people’s background knowledge,theories, beliefs—their interpretive resources—and the structural position and meaning of theirgroup memberships. With social change oftencomes the emergence of new group member-ship and allegiances (e.g., as a function of newpatterns of intergroup relations, changes in

the salient level of self-categorization, andinfluence processes). Along with the emergenceof new groups come different norms, values andbeliefs, and understandings of appropriate andvalued social conduct. In these processes liesthe potential for different individual and groupaspirations that present the opportunity for newpersonalities (while others may be modified andattenuated).

This SCT view of the person in which the personaland social are integrated highlights the interplaybetween individual and collective life. There ismuch work that is being done on individuality,personality, and the personal self but all of it flowsfrom the different understanding of the selfadvanced by SCT.

Glimpses of the Future: Social Psychologyas a Special Science

The social identity perspective has always seensocial psychology as a special science. We rejectthe idea that it is simply a branch of general psych-ology applied to social stimuli. The whole trend inindividualistic psychology is towards a reifyingreductionism—the cause is found in some inneressence—but in social identity research theemphasis should be on the unified psychological-social field, in a much more Lewinian spirit. Peopleare transformed by their social relationships andactivities, and this includes their minds, and theorymust make such transformation possible andexplicable.

Group life and social context are not factors thatmoderate the functioning of the mind, they funda-mentally shape it. Social-psychological processesarise from the functioning of the human mind in anorganized social environment. The social contextqualitatively affects their mental functioning. Thushumans are not merely individuals and neither areour minds. We have both collective selves aswell as personal selves. The group is an emergentpsychological process that makes possible collect-ive products, which in turn become psychologicalforces, social values, ideologies, and power struc-tures. All aspects of human psychological func-


NOTinginparatiparatithese intthese incategorizatioorizatimine onemine o ’s behs beh

s such, individh, indiviemporarmp

nuFOReatioeatio

and changeand changthat underpin sethat underpin se

vary. If there is If there isge and the situatand the situa

y-to-day basis, y-to-day basis, e self-categorizself-catego

I”). If pe. Igege DISTRIBUTION

n fan fawever, onwever, o

change also.change alsoeen differenten different

dentities,ntitiealues,ue

ity.ity.

socsothe idthe idology apogy apindividualisindividualisreductionismreductionism——

sencesen ——but in but inhasis should be is should be

eld, in a much meld, in a much mrmed by their smed by their

this includes thethis includes theh transformatiotransformatio

context are not ntext are noof the mind, theyof the mind, they

ychological procchological prohuman mind in auman mind in

social contextsocial contextoning. Thusning. Thus

either aher a

11:31:06:01:10

Page 31

Page 31

tioning, from the cognitive to the emotional,motivational, and behavioural, are affected by themind–society interaction. Whatever the nature ofthe observation, they are all shaped by the sociallydefined self-process. This is a view of humanpsychology that is not limiting of what we can beand how humans can live. It acknowledges howpeople can redefine their social relationships and,in the process, their personhood.

In work on social identity these ideas have beendeveloped in great detail, in particular in researchon social influence and social norms, prejudice, theself-process, personality, and cognition. There aremany who have shared in the enterprise to date andwho have helped to make the leap forward. Thestory of social identity is that it generates a newdistinctive vision that has fundamental implica-tions for social psychology, other branches ofpsychology, and other disciplines. The scientificimplications of a truly non-reductionist view of themind will be socially radical, contradicting manycurrent assumptions within science. In many waysthe work has hardly begun.

Acknowledgments

This chapter was supported by funding from theAustralian Research Council, including an Austral-ian Professional Fellowship to Professor Turner.We would like to thank Boris Bizumic, CarolynNewbigin, and Emina Subasic for comments onearlier drafts of this manuscript. Thanks also tothe editors of this volume, Nyla Branscombe andTom Postmes, for their helpful suggestions andpatience.

REFERENCES

Barsalou, L. W. (1987). The instability of graded structure:Implications for the nature of concepts. In U. Neisser (Ed.),Concepts and conceptual development: Ecological andintellectual factors in categorization (pp. 101–140).Cambridge, UK: Cambridge University Press.

Bruner, J. S. (1957). On perceptual readiness. PsychologicalReview, 64, 123–152.

Deutsch, M., & Gerard, H. B. (1955). A study of normativeand informational social influences upon individual judg-ment. Journal of Abnormal and Social Psychology, 51,629–636.

Festinger, L. (1954). A theory of social comparison processes.Human Relations, 7, 117–140.

Lewin, K. (1948). Resolving social conflicts. New York:Harper & Brothers.

Lewin, K. (1952). Field theory in social science. London:Tavistock.

Lott, A. J., & Lott, B. E. (1965). Group cohesiveness as inter-personal attraction: A review of relationships with ante-cedent and consequent variables. Psychological Bulletin,64, 259–309.

Medin, D. L., & Wattenmaker, W. D. (1987). Categorycohesiveness, theories, and cognitive archeology. InU. Neisser (Ed.), Concepts and conceptual development:Ecological and intellectual factors in categorization.Cambridge, UK: Cambridge University Press.

Moscovici, S. (1976). Social influence and social change.London: Academic Press.

Oakes, P. J., Haslam, S. A., & Turner, J. C. (1994). Stereotyp-ing and social reality. Oxford, UK: Blackwell.

Rabbie, J. M., & Horwitz, M. (1969). Arousal of ingroup–outgroup bias by a chance win or loss. Journal of Personal-ity and Social Psychology, 13, 269–277.

Sherif, M. (1967). Group conflict and co-operation: Theirsocial psychology. London: Routledge & Kegan Paul.

Sherif, M., Harvey, O. J., White, B. J., Hood, W. R., & Sherif,C. W. (1961). Intergroup conflict and cooperation: theRobbers Cave experiment. Norman, OK: University ofOklahoma Book Exchange.

Tajfel, H. (1957). Value and the perceptual judgement of mag-nitude. Psychological Review, 64, 192–204.

Tajfel, H. (1969). Cognitive aspects of prejudice. Journal ofSocial Issues, 25, 79–97.

Tajfel, H. (1972a). Social categorization. In S. Moscovici(Ed.), Introduction a la psychologie sociale (Vol.1). Paris:Larouse.

Tajfel, H. (1972b). Experiments in a vacuum. In J. Israel& H. Tajfel (Eds.), The context of social psychology(pp. 69–119). London: Academic Press.

Tajfel, H. (Ed.). (1978). Differentiation between socialgroups: Studies in the social psychology of intergrouprelations. London: Academic Press.

Tajfel, H., Flament, C., Billig, M. G., & Bundy, R. F. (1971).Social categorization and intergroup behaviour. EuropeanJournal of Social Psychology, 1, 149–177.

Tajfel, H., & Turner, J. C. (1979). An integrative theory ofintergroup conflict. In W. G. Austin & S. Worchel (Eds.),The social psychology of intergroup relations (pp. 33–47).Monterey, CA: Brooks/Cole.

Tajfel, H., & Turner, J. C. (1986). The social identity theory ofintergroup behaviour. In S. Worchel & W. G. Austin (Eds.),Psychology of intergroup relations (pp. 7–24). Chicago, IL:Nelson-Hall.

Tajfel, H., & Wilkes, A. L. (1963). Classification and quantita-tive judgement. British Journal of Psychology, 54,101–114.

Turner, J. C. (1972, February). Social comparison and socialidentity: Some prospects for intergroup behaviour. Paper


NOTned sned

psychology psychology and how humaand how huma

ople can redeple can redefifinprocess, their perocess, their p

rk on social idenrk on social idenn great detail,eat detail

ence andncon FOR

cial ncial ny, and cogny, and cog

d in the enterprid in the enterpriake the leap forhe leap fo

that it generatesat it generatesundamental impundamental imp

ther branches her brancheThe sciee

viv DISTRIBUTION

fificcof theof the

g manyg mwaysways

SheheC. WCRobberRobberOklahomaOklahoma

Tajfel, H. (1957Tajfel, H. (1957nitude. nitu PsycholoPsycholofel, H. (1969). CogH. (1969). Co

ial Issuesal Issues, , 2525, 7979––9H. (1972a). Social (1972a). Soci

roduction a la psyction a la psyc

). Experiments in Experiments i.), ), The context ofThe context

Academic Press.Academic PDifferentiation bfferentiation bcial psychology of ial psychology of

ress.ress, & Bundy, R. F. (1Bundy, R. F.

behaviour. behaviour. EuropeaEurope7

ative theory ofve theory ofrchel (Eds.),chel (Eds.)

. 3333–47).47).

of

11:31:06:01:10

Page 32

Page 32

presented at the EAESP Small Group Meeting on Inter-group Relations, Bristol, UK.

Turner, J. C. (1975). Social comparison and social identity:Some prospects for intergroup behaviour. European Jour-nal of Social Psychology, 5, 5–34.

Turner, J. C. (1978). Social categorization and social dis-crimination in the minimal group paradigm. In H. Tajfel(Ed.), Differentiation between social groups: Studies in thesocial psychology of intergroup relations (pp. 235–250).London: Academic Press.

Turner, J. C. (1981). The experimental social psychology ofintergroup behaviour. In J. C. Turner & H. Giles (Eds.),Intergroup behaviour (pp. 66–101). Oxford, UK:Blackwell.

Turner, J. C. (1982). Towards a cognitive redefinition of thesocial group. In H. Tajfel (Ed.), Social identity and inter-group relations (pp. 15–40). Cambridge, UK: CambridgeUniversity Press.

Turner, J. C. (1984). Social identification and psychologicalgroup formation. In H. Tajfel (Ed.), The social dimension:European developments in social psychology (Vol. 2,pp. 518–538). Cambridge, UK: Cambridge UniversityPress.

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 (Vol. 2,pp. 77–122). Greenwich, CT: JAI Press.

Turner, J. C. (1988). Comments on Doise’s “Individual andsocial identities in intergroup relations”. European Journalof Social Psychology, 18, 113–116.

Turner, J. C. (1991). Social influence. Pacific Grove, CA:Brooks/Cole.

Turner, J. C. (1996). Henri Tajfel: An introduction. In W. P.Robinson (Ed.), Social groups and identity: Developing thelegacy of Henri Tajfel (pp. 1–23). Oxford, UK:Butterworth-Heinemann.

Turner, J. C. (1999). Current issues in research on social iden-tity and self-categorization theories. In N. Ellemers, R.Spears, & B. Doosje (Eds.), Social identity: Context, com-mitment, content (pp. 6–34). Oxford, UK: Blackwell.

Turner, J. C. (2001, October). Rethinking the nature of preju-dice from psychological distortion to socially structuredmeaning. The Second Freilich Foundation Eminent Lec-turer Series, Canberra.

Turner, J. C. (2005). Explaining the nature of power: A three-process theory. European Journal of Social Psychology,35, 1–22.

Turner, J. C., & Bourhis, R. Y. (1996). Social identity, inter-dependence and the social group: A reply to Rabbie et al. InP. Robinson (Ed.), Social groups and identities: Develop-ping the legacy of Henri Tajfel (pp. 25–63). Oxford, UK:Butterworth-Heinemann.

Turner, J. C., & Brown, R. J. (1978). Social status, cognitivealternatives and intergroup relations. In H. Tajfel (Ed.), Dif-ferentiation between social groups (pp. 201–234). London:Academic Press.

Turner, J. C., & Haslam, S. A. (2001). Social identity, organ-izations and leadership. In M. E. Turner (Ed.), Groups atwork: Advances in theory and research (pp. 25–65).Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.

Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., &Wetherell, M. S. (1987). Rediscovering the social group:A self-categorization theory. Oxford, UK: Blackwell.

Turner, J. C., Hogg, M. A., Oakes, P. J., & Smith, P. M. (1984).Failure and defeat as determinants of group cohesiveness.British Journal of Social Psychology, 23, 97–111.

Turner, J. C., & Oakes, P. J. (1986). The significance of thesocial identity concept for social psychology with referenceto individualism, interactionism and social influence. Brit-ish Journal of Social Psychology, 25, 237–252.

Turner, J. C., & Oakes, P. J. (1989). Self-categorization theoryand social influence. In P. B. Paulus (Ed.), The psychologyof group influence (pp. 233–275). Hillsdale, NJ: LawrenceErlbaum Associates, Inc.

Turner, J. C., & Oakes, P. J. (1997). The socially structuredmind. In C. McGarty & S. A. Haslam (Eds.), The messageof social psychology (pp. 355–373). Oxford, UK:Blackwell.

Turner, J. C., Oakes, P. J., Haslam, S. A., & McGarty,C. A. (1994). Self and collective: Cognition and social con-text. Personality and Social Psychology Bulletin, 20,454–463.

Turner, J. C., & Onorato, R. (1999). Social identity, personal-ity and the self-concept: A self-categorization perspective.In T. R. Tyler, R. Kramer, & O. John (Eds.), The psychologyof the social self (pp. 11–46). Mahwah, NJ: LawrenceErlbaum Associates, Inc.

Turner, J. C., & Reynolds, K. J. (2001). The social identityperspective in intergroup relations: Theories, themes andcontroversies. In R. Brown & S. Gaertner (Eds.), Handbookof social psychology: Vol 4: Intergroup processes(pp. 133–152). Oxford, UK: Blackwell.

Turner, J. C., & Reynolds, K. J. (2002). From the inevitabilityof prejudice to the origins of social change: The emergenceof perceived illegitimacy in intergroup relations. Fundedgrant application, Australian Research Council.

Turner, J. C., & Reynolds, K. J. (2003). Why social domin-ance theory has been falsified. British Journal of SocialPsychology, 42, 199–206.

Turner, J. C., Reynolds, K. J., Haslam, S. A., & Veenstra, K.(2006). Reconceptualizing personality: Producing indi-viduality by defining the personal self. In T. Postmes &J. Jetten (Eds.), Individuality and the group: Advances insocial identity (pp. 11–36). London: Sage.

Turner, J. C., Reynolds, K. J., & Subasic, E. (2008). Identityconfers power: The new view of leadership in social psych-ology. In P. ’t Hart & J. Uhr (Eds.), Public leadership:Perspectives and practices (pp. 52–72). Canberra, Australia:ANU E-press.

Turner, J. C., Sachdev, I., & Hogg, M. A. (1983). Social cat-egorization, interpersonal attraction and group formation.British Journal of Social Psychology, 22, 227–239.


TurTcrimrim(Ed.), (Ed Dsocial psysocial psyLondon: AcaLondon: A

Turner, J. C. (1981urner, J. C. (198intergroup behaviouintergroup behavio

ergroup behaviop behaviwell.ell.

. (19FORTowards Towards

H. Tajfel (Ed.)H. Tajfel (Ed(pp. 15(pp 15–40). Camb40). Camb

cial idential fication acation aajfel (Ed.), ajfel (Ed.), The socThe so

social psychologocial psychoUK: CambridgK: Cambrid

za DISTRIBUTION

and the seand the sp behaviour. In E.p behaviour. In E.

cessescesses (Vol. 2, (Vol. 2,

dual anddual anournalrna

TTityyIn T. In T. of the sof the sErlbaum AErlbaum A

Turner, J. C., & Turner, J. C., &perspective in intperspective inontroversies. In R. Bversies. In R. B

social psychologycial psychology3––152). Oxford, UK152). Oxford, U

& Reynolds, K. J. (& Reynolds, K. J.o the origins of sociathe origins o

gitimacy in intergrmacy in intergrustralian Research Cstralian Research C

s, K. J. (2003). Whs, K. J. (2003). Wsified. d. British JourBritish Jou

lam, S. A., & Veenm, S. A., & Veenality: Producing iy: Producing i

f. In T. Postmes &f. In T. Postmesoup: Advances inup: Advances in

8). IdentIdentl pp

Documents

NOT The Story of Social Identity · identity research and ideas. The late Henri Tajfel started the story, writing the ﬁrst social identity paper in 1971 (or possibly 1970). Turner