The concept of blended identity
This article uses the term ‘blended identity’ to convey the concept that many different aspects or dimensions of self-definition can be salient simultaneously. In other words, it includes the situation where more than one identity simultaneously shapes a person’s self-definition in a particular social context. The study of blended identity is controversial, perhaps mainly due to the fact that some critics believe that this theoretical notion can be overstated. Some scholars reject talk of blended identity as naïve or unrealistic, concerned that some theorists suggest that the adoption of a blended identification is virtually automatic, common and universally desired. [7] Of course, some migrants may wish to stop identifying in terms of an identity that, for example, represents a tyrannical regime from which they have escaped. Other critics of the notion of blended identity suggest that claiming that blended identification is common is simply another form of essentialism that suggests that migrants always identify in terms of these complex identities. This universalism is overstated in our view, and fails to explain the psychological processes determining when particular identifications occur. [8] Our use of this concept, alongside empirical, process-based descriptions of blended identity salience, should reassure those critics.
The concept of blended identity has been discussed by researchers from a number of different disciplines. Other common terms include: dual nationality, dual citizenship, ethnic national identity, [9] hyphenated identity or ethnicity, [10] or simply mixed identities. In particular, social psychologists have sometimes labelled blended identities as ‘crossed categorisations’ to describe the combination or ‘crossing’ of many identities to form a complex identity. There have been three decades of social psychological research into phenomena flowing from blended identifications. This has mainly involved research into how crossed categorisations shape perceptions of other group members. For example, being biased in favour of your in-group (that is, the identity perspective made salient for you by a social comparison) and distinguishing self from and/or thinking negatively of out-groups (that is, groups and their perspectives with which you do not feel a sense of belonging in the context of a particular social controversy). [11]
The term ‘crossed categorisation’ reflects theoretical ideas about social identity salience described above. In the case of blended identity salience, the social context makes salient at least two identities, rather than simply one identity. This more complex categorisation or definition of the self is the perspective from which people determine their identity-based beliefs and ideologies. [12] Importantly, researchers study crossed categorisations that include the salience of national identities as part of a blended identification. [13]
If people have many constituent aspects of self, it should not be surprising that sometimes two or more of these identities are simultaneously salient and responsible for shaping our perspective on a social issue. Perhaps these more complex, blended identities are in fact more normal for some people including migrants than is the simpler case of the salience of only one (national) identity. The burgeoning interest in the social psychology of crossed categorisations is a valuable addition to citizenship research. It is useful to consult research on crossed categorisation before citizenship law reform championing identification with only one single national identity is thought superior to celebrating blended identity.
Measurement of identification with a blended identity
Verkuyten studied the relative strength of religious and national identification in a sample of 206 Turkish-Dutch citizens who identified as Sunni Muslims. Attitudes, though not real behaviours, towards other religious groups in the Netherlands were also measured. The study was conducted in the context of what Verkuyten describes as Dutch citizenship policies promoting assimilation via strong national identification. [14]
Verkuyten found that half of the sample rated their Islamic identity as strongly as they could on all measurement dimensions, choosing seven on a seven-point rating scale on all measures of the importance of the Islamic identity. This ceiling effect or ‘total’ identification as a Muslim within these blended identities was interpreted as supporting studies showing the very high importance of Islamic identity for Muslims living blended identities in the West. [15]
For these ‘total’ Muslim identifiers, the religious component of identity appeared a highly central way of defining self in the social judgment contexts examined. Importantly, even though half of the participants had a ‘Muslim identity [that] was an integral or inextricable part of how they saw themselves’, [16] their Dutch national identification was not eliminated. Their Dutch national identification was significantly lower than the national identification of the other half of participants whose Islamic identity was strong though not ‘total’. Those with ‘total’ identification rated their in-group most positively and felt strongly about Islamic group rights. However, Verkuyten could still classify 19 per cent of the ‘total’ Muslim identifiers as high Dutch identifiers (who rated the importance of that identity above the midpoint of the response scale). Also, 31 per cent of high Muslim identifiers simultaneously identified as Dutch nationals and had high levels of national identification.
For some, the impact of strong religious identification upon national identification could appear alarming. However, the observable psychological benefits of these blended identities are an important part of evaluating any alleged risk of blended identification. For these ‘totally identified’ Muslim participants, the most negative feelings towards other religious groups in the Netherlands were towards non-believers. When total Muslim identification and high national identification co-occurred, hostility towards other religious groups (Jews, Christians, Hindus) decreased. Most negative beliefs about other Dutch religious groups were held by total Muslim identifiers with low national identification, suggesting that hostile beliefs were caused more by a single rather than a blended identification.
These data suggest the importance of relationships between each aspect of a blended identity for resulting social attitudes (that is, beliefs about nationality and about religious groups to which the perceiver does or does not belong). It does not suggest, however, that high or total identification as a Muslim always eliminates or suppresses national identification. This detail of this study is highlighted as an example of the type of research that should be conducted and consulted in order to understand the subtle dynamism of blended identity expression in political contexts. The reasons why a blended identifier does or does not strongly identify as a national of their country of residence is an important social-psychological and political question. Not only do these data suggest the reality of blended identities and the possibility of identification with a blended identity and high national identification, these data suggest that detailed measurement of the impact of citizenship regimes is needed so we can better understand exactly how and when citizenship regimes may stifle blended identities or allow them to exist.
Verkuyten notes too that crossing religious identity with national identity is an example of a complex blended identity, perhaps even more complex than combining two national identities or, say, language groups. [17] It should be remembered that each example of crossed categorisation, in political and historical context, will produce its own unique identity management challenges. Possible social tensions that result are not simply a function of whether blended rather than single identity salience occurs but are influenced by relationships between the constituent identities in a crossed categorisation and social treatment of these identities.
Common in-group identity research
Some social psychologists argue that adopting one superordinate, common in-group identity (for example, Australian) within diverse societies can reduce tension and hostile beliefs between groups. However, some researchers have found this occurs when valued cultural identities (for example, Arab identity) can remain strong following identification with the common in-group identity (for example, Australian). [18] This suggests the benefit of blended identification that prevents mere recategorisation of all subordinate identities into a higher-order national identity. Rather than replacing this diversity with a unitary national identity, there is benefit in promoting a superordinate Australian citizenship that allows explicit and continuing identification with other elements of a blended identity.
Similarly, important recent empirical work reveals the impact on in-group bias when making a common in-group identity salient whilst measuring how people perceive the relative importance of maintaining a crossed categorisation. [19] In this work, the perceived importance of the crossed categorisations determined whether use of a common in-group identity (for example, Australian) resulted in harmony and perceived inclusiveness instead of continuing tension and continuing perceptions of inter-group difference. Results showed that when the elements of a blended identity remained subjectively important for a perceiver, then identifying only in terms of a common in-group identity did not reduce perceived inter-group tension between subordinate identities.
[7] Cashmore, E. 2003, ‘The impure strikes back’, British Journal of Sociology, vol. 54, no. 3, pp. 407–14, 411, citing Kumar, K. 2003, The Making of English National Identity, Cambridge University Press, Cambridge, p. 258.
[8] Ibid., p. 411, citing Anthias, F. 2001, ‘New blendedities, old concepts: the limits of “culture”’, Ethnic and Racial Studies, vol. 24, no. 4, pp. 619–41, 637.
[9] Korac, M. 1996, ‘Understanding ethnic-national identity and its meaning: questions from women’s experience’, Women’s Studies International Forum, vol. 19, no. 1–2, pp. 133–43; Verkuyten, M. 2005, The Social Psychology of Ethnic Identity, Psychology Press, London.
[10] Duany, J. 2003, ‘Nation, migration, identity: the case of Puerto Ricans’, Latino Studies, vol. 1, no. 3, pp. 424–44; Glazer, N. 2004, ‘Assimilation today: is one identity enough?’, in T. Jacoby (ed.), Reinventing the Melting Pot: The new immigrants and what it means to be American, Basic Books, New York, pp. 61–73; Mahtani, M. 2002, ‘Interrogating the hyphen-nation: Canadian multicultural policy and “mixed-race” identities’, Social Identities, vol. 8, no. 1, pp. 67–90.
[11] Mullen, M., Migdal, M. J. and Hewstone, M. 2001, ‘Crossed categorization versus simple categorization and intergroup evaluations: a meta-analysis’, European Journal of Social Psychology, vol. 31, no. 6, pp. 721–36.
[12] For an early example of experimental research with crossed categorisations, see Brown, R. J. and Turner, J. C. 1979, ‘The criss-cross categorization effect in intergroup discrimination’, British Journal of Social & Clinical Psychology, vol. 18, no. 4, pp. 371–83.
[13] Verkuyten, M. 2007, ‘Religious group identification and inter-religious relations: a study among Turkish-Dutch Muslims’, Group Processes and Intergroup Relations, vol. 10, no. 3, pp. 341–57.
[14] Ibid.
[15] Ibid., p. 343: supporting observations from H. C. Triandis (1992, ‘Comments on Hinkle, Brown and Ely’, Revisita de Psicologia Social, vol. 3, pp. 113–23) that for some identifications in some cultures, one is clearly a (strongly identified) member of a group or one is not. This could mean that group identification is a binary choice rather than a form of belonging that can be measured on a continuum in terms of low and high identification. Being a low identifier with these identities is not an option.
[16] Ibid., p. 347.
[17] Ibid., p. 351.
[18] Hornsey, M. and Hogg, M. 2000, ‘Subgroup relations: a comparison of mutual intergroup differentiation and common ingroup identity models of prejudice reduction’, Personality and Social Psychology Bulletin, vol. 26, pp. 242–56.
[19] Crisp, R. J., Walsh, J. and Hewstone, M. 2006, ‘Crossed categorization in common ingroup identity contexts’, Personality and Social Psychology Bulletin, vol. 32, no. 9, pp. 1204–18.