Full Terms & Conditions of access and use can be found athttps://www.tandfonline.com/action/journalInformation?journalCode=rmmm20

Journal of Multilingual and Multicultural Development

ISSN: 0143-4632 (Print) 1747-7557 (Online) Journal homepage: https://www.tandfonline.com/loi/rmmm20

Cultural festivals as intergroup settings: a casestudy of Pacific Islander identification

Matt Giles, Howard Giles & Quinten Bernhold

To cite this article: Matt Giles, Howard Giles & Quinten Bernhold (2019): Cultural festivals asintergroup settings: a case study of Pacific Islander identification, Journal of Multilingual andMulticultural Development, DOI: 10.1080/01434632.2019.1569666

To link to this article: https://doi.org/10.1080/01434632.2019.1569666

Published online: 05 Feb 2019.

Submit your article to this journal

Article views: 44

View Crossmark data

Cultural festivals as intergroup settings: a case study of PacificIslander identificationMatt Giles, Howard Giles and Quinten Bernhold

Department of Communication, University of California, Santa Barbara, CA, USA

ABSTRACTAddressing the current gap in the literature regarding cultural festivals as aunique site of intergroup discourse, we invoke social identity and groupvitality theories to explore the effect of attending an internationalcultural festival on members of different groups. A total of 143participants at the 2016 Festival of Pacific Arts in Guam completedsurveys and interviews concerning identity salience. Measures of ethnicidentity and meta-identity salience both increased (and interacted) afterparticipation in the Festival, and the region of origin also hadmoderating effects. Standard paradigms regarding single identitysalience are discussed as well as future avenues for intercultural work.

ARTICLE HISTORYReceived 2 April 2018Accepted 2 January 2019

KEYWORDSFestival; intergroupcommunication; identitysalience; Pacific culture;functional antagonism

Performers from dozens of island nations in the Pacific gathered in Guam during the Summer of2016 for the Festival of Pacific Arts, a quadrennial two-week cultural celebration of their unique tra-ditions, cultural performances, and native artwork. These traditional performances take place along-side contemporary art and collaboration between groups. Participants enjoyed an impromptuMichael Jackson dance party that took place after the Festival’s closing ceremonies and the Tongan,Samoan, and Easter Island delegations collaborated in carving an elaborate ceremonial canoe prowfigurehead using chainsaws. During this synthesis of tradition and modernity, the different groups ofPacific Islanders perform and provide a fascinating opportunity to study group membership nego-tiation and group identity salience in an intergroup environment. The resulting data provides a richmodel for how identity is made salient and the complex way that multiple identities become salientsimultaneously. This has valuable theoretical implications, as the predominant paradigms in the fieldposit that only one single identity can be brought to salience at any one time, invoking the idea offunctional antagonism (Turner et al. 1987). To explore this idea, the festival context in which thesedisparate identities are being negotiated, namely, the 2016 Festival of Pacific Arts, deserves furtherdiscussion.

Literature review

Festivals and communication

Festivals warrant a significant place within culturally-based research because of their function as aconfluence of ethnic and cultural identities. Identity scholars argue that participants in large culturalevents go through a process of depersonalization (e.g. Hogg, Terry, and White 1995; Turner et al.1987). At this point, individuals see themselves as ‘an embodiment of the in-group prototype ratherthan a unique individual’ (Stets and Burke 2000). It is during this process that pertinent groupphenomena such as cooperation, group cohesion, ethnocentrism, and collective action occur (Turner

© 2019 Informa UK Limited, trading as Taylor & Francis Group

CONTACT Matt Giles matt@ucsb.edu

JOURNAL OF MULTILINGUAL AND MULTICULTURAL DEVELOPMENThttps://doi.org/10.1080/01434632.2019.1569666

et al. 1987). These phenomena reflexively create and reify identities which are communicatedbetween participants.

Festivals function as uniquely valuable sites for relationship development between members ofdisparate cultural groups, and the individual performances within a festival are reflexive demon-strations of group identities (Bauman 1992). These events provide a recurring opportunity to estab-lish social order at a local level (Mendoza 2000), and as a strategy to increase autonomy for thosedisenfranchised by a global cultural economy (Noyes 2003). Both the identities of festival partici-pants and those of local community members are enhanced and reinforced through the intergroupinteractions inherent to these large group events (Getz 1993). Group vitality (for a review, see Smith,Ehala, and Giles 2018) focuses on the resources that a group needs in order to survive and ‘behave asa distinctive and active collective identity in intergroup situations’ (Giles, Bourhis, and Taylor 2009).At these festivals (including the one studied herein), vitality is on public display via pride displays,public representations, and comparisons between groups.

In what follows, we introduce the festival setting and examine intergroup theory that addressesthe salience of one’s multiple identities in terms of whether only one identity can be triggered orwhether more than one is evident simultaneously. We argue that the festival under investigationpotentially allows for ethnic-level as well as meta-identities (herein Pacific Islander) to subjectivelyco-exist and explore the Festival as a unique site for doing so.

The cultural site

Given the significance of social identity in intergroup environments, we explore the effect of partici-pation in the 2016 Festival of Pacific Arts on the identity salience of performers who were involved.The Festival is coordinated by an international politico-economic organisation known as the PacificCommunity, a regional intergovernmental organisation comprised of the various island nations ofthe Pacific. The Pacific Community seeks to foster an overarching Pacific Islander identity that isachieved by reinforcing the traditional customary practices that are central to that identity, whilesimultaneously embracing emergent understandings of this identity. By performing in the Festival,‘participants confront profound existential contradictions in their attempts to reconcile tensionsbetween tradition and modernity’ (Glowczewski and Henry 2011). The 12th Festival of PacificArts took place in June 2016 and was held in Hagåtña, Guam (for anthropological and economicdiscussions of the Festival of Pacific Arts in previous years, see Bellwood et al. 2011; Chang et al.2015; Glowczewski and Henry 2011; Henry and Foana’ota 2015; Mirabal et al. 2013).

Participants at the Festival claim ancestral ties to islands in the Pacific, in a region known as Ocea-nia. Many cultural groups attending the conference are national, but some are not; for example, thedifferent Guamanian, Hawaiian, and Micronesian delegations are all from the United States. Like-wise, many participants hail from a singular village within a larger nation composed of such hetero-genous groups, such as the Triobrand Islanders who are almost entirely disconnected from theSouthern Highlands Huli tribe, yet both hail from Papua New Guinea, and are referred to assuch. Within the broader scope of Oceania, individual groups are generally associated with one ofthe three semi-distinct cultural groups of Micronesia, Melanesia, and Polynesia (Figure 1), butthis division traces its roots back to non-scholarly imperialistic classifications by colonisers, and isa fraught method of distinction (Hermann 2011; Tcherkézoff 2003). As such, the dividing linesbetween the groups are crude and permeable – linguistically, geographically, culturally, and pheno-typically (Jolly, Tcherkézoff, and Tryon 2009). Despite their imperfect nature, these groupings reflecta reference point which has been used extensively and can be considered salient categories in inter-group contexts (Thomas 1989).1

As with any colonial-era division and classification, globalisation and indigenous scholarshipreveal the flaws in any oversimplified distinction between Melanesian, Micronesian, and Polynesianpeople groups. Migration, inter-group marriage, urbanisation, and other factors blur what may haveonce been considered rigid dividing lines, while simultaneously revealing the permeable and

2 M. GILES ET AL.

transitory nature of supposed cultural boundaries. Events such as cultural festivals serve to reify thesecultural boundaries while dialectically providing the participants with the opportunity to deconstructthem, reconstruct them, and synthesise their own. Pacific Islander identity cannot be understood as afixed target. Any analysis of this construct must be engaged with in a way that honours both its fluidnature and its ancestral traditions, a dialectic that is inherent to the diaspora (Mackley-Crump 2015).It is this individual-yet-reciprocal understanding of identity that provides focus and context for thisresearch, which investigates the salience of group level and meta-group level identities.

The Festival seeks to embody and reify the notion that the Pacific is not a location inhabited bydetached ‘islands in a far sea’ but instead is home to an interconnected ‘sea of islands’ (Haùofa 1994,152–153); this is seminal discourse in Pacific Studies.

Research on identities

This study tracks how meta-identity of Pacific Islanders interrelates with ethnic/national identitiesrepresented by Festival participants. We do not attempt to define Pacific Islander identity but,instead, seek to gain a sense of the level of identification with multiple identities. Tajfel and Turner(1979) argued that there are three mental processes associated with identity: Individuals learn to cat-egorize themselves and others, adopt social identifications, and compare relevant identity dimensionsbetween social groups. These identifications are significant because identity affects what decisions aremade by an individual. Social identity theory (SIT: e.g. Tajfel and Turner 1979) asserts that by affect-ing an individual’s values and ideologies, identity cannot be considered entirely separate from actions(Abdelal et al. 2006). Many scholars argue that SIT predicts ingroup cooperation and outgroup confl-ict (Bonnell 1980; Siegelbaum and Suny 1994). Identity has been shown to predict social behaviourssuch as participation in both violent and non-violent social movements (Leff 1997; Rouhana 1997;Stokes 1995) and to explain otherwise irrational business decisions (Ouchi 1981). Identity influences

Figure 1. Map of Oceania.

JOURNAL OF MULTILINGUAL AND MULTICULTURAL DEVELOPMENT 3

voters based on either ingroup/outgroup dynamics or due to shared values based on identity (Jep-person, Wendt, and Katzenstein 1996).

When an individual identifies with multiple groups, some identities are more significant in onesituation than another. This is referred to as identity salience and work arising from this constructhas explored when one identity takes precedence over another (Burke and Reitzes 1981). Accordingto Hogg, Terry, and White (1995), ‘Identity salience is conceptualised (and operationalised) as thelikelihood that the identity will be invoked in diverse situations’ (257). A salient identity is theone that functions psychologically to increase the influence of group membership on both interpret-ations of events and decisions made on its behalf (Oakes 1987).

Situated identity has received significant attention as a topic of study as it relates to language andidentity. Because language is intrinsic to identity (see Fishman 1977; Sapir and Mandelbaum 1968),the ability to successfully communicate in intergroup settings makes language behaviour and ethnicidentity especially salient at festivals. Noels and Clément (2015) demonstrates how exposure to inter-active intercultural situations changes both identity and behaviour. Noels (2014) explains these shiftsas part of the process of acculturation which penetrates into more intimate situations outside oflarge-scale public interactions, thus demonstrating the lasting significance of events such as festivals(Clément, Dornyei, and Noels 1994; Clément, Noels, and Denault 2001). Social psychologists arguethat contact with other cultural groups allows an individual to identify with multiple identities (e.g.Berry 1990; Noels, Clément, and Gaudet 2004). The possibility of assuming multiple ethnic identitieshas been studied extensively by researchers with respect to social identity complexity and bicultur-alism (Benet-Martínez and Haritatos 2005; Brewer 2001); however, it should be noted that thisresearch has rarely, if ever, attempted to demonstrate simultaneous or combined identity salience.

Problematising functional antagonism

Much of the current research on identity operates from the standpoint of functional antagonism,which asserts that only a single identity is operationally salient at any given point in time (Turneret al. 1987). This dominant perspective within SIT contends that ‘only one identity is psychologicallysalient to govern self-construal, social perception, and social conduct. As the situation… changes, sodoes the salient identity’ (Hogg 2006). From this perspective, a person makes decisions and reacts tosituations at any one moment based entirely on one single identity at a time. Similarly, Mullen (1991)posits an asymmetry of identity – a relationship in which one identity can only come to prominenceat the expense of another.

Thoits and Virshup (1997) state that a single ‘role-based identity becomes salient and supplantsthe previous identity when an individual takes on specific responsibilities that provide a frameworkfor negotiating identity issues’ (94). Similar perspectives can be seen throughout social psychologicalresearch (Burke and Reitzes 1981; Oakes 1987). Within this framework, when one identity becomessalient, other identities are diminished or even dismissed so that, in any one moment, a person ident-ifies with one distinct group (Tajfel and Turner 1979). This dynamic has been explored to identifywhat factors influence the primacy of individual identities, such as generational status (Garcia et al.2015), gender (Dion and Dion 2001), reference group (Clément and Noels 1992), and situation(Clément, Singh, and Gaudet 2006). In sum, single identity salience is the base understanding withinSIT theory.

Despite the prominence of single identity salience (SIS) approaches, there are researchers whosuggest that the SIS paradigm needs additional nuance. These perspectives advocate for identitiesthat are combined, transient, or layered, depending on the situation (Mullen, Migdal, and Rozell2003). Swann et al. (2009) assert the notion of identity fusion in which individuals feel that their per-sonal and social identities overlap. Research into multiple group memberships recognises ‘crossedcategorizations’ in which individuals identify with multiple groups in certain situations (Crispand Hewstone 2001), as well as larger meta-groups as manifest in superordinate attributes such asgender or nationality (Park and Rothbart 1982). The commodification and festivalization

4 M. GILES ET AL.

(Mackley-Crump 2015) of identity in the tourism industry and at international events deserves aspecific focus of study within identity research, although traditional theories of identity do not alwaysmap well onto the terrain. Complementing this work, this paper seeks to identify potential situationsunder which an individual experiences multiple identities as salient simultaneously.

The current study

Within this paradigm of multiple identity salience – which we find more convincing than the func-tional antagonist position – the exact nature of the interaction between the multiple identities as theystruggle for prominence is currently unknown, and this study seeks to clarify that interaction. Toexplore the relationship between ethnic-level identities and the international meta-identity ofPacific Islander, we propose two research questions:

RQ1:Will festival-goers report increased salience in either or both their ethnic-level (e.g. Tongan, Papuan, Cha-morro) and meta-identities (variously named Pacific Islander or Oceanic) after having participated in culturalevents at the Festival?

RQ2: How will the experiences of festival-goers from various ethnic-level identities differ relative to each othergroup in terms of ethnic and meta-identity salience?

Method

Participants

A consecutive sampling technique was adopted to survey a total of 143 performers at the 2016 Fes-tival of Pacific Arts in Hagåtña, Guam. Approximately 1200 stage performers from 27 islandnations participated at the Festival, which attracted over 3500 attendees. Participants wereapproached and asked to complete a short survey about their experience at the Festival after per-forming at the main stage venue. Surveys were completed on hand-held tablets after performerscompleted their presentations. Because this form of sampling is non-random, where some nation-alities were underrepresented in the responses, a quota-based stratified sample was gathered. Par-ticipants performed as members of delegations representing the 23 nations of the Secretariat of thePacific and were sampled in a stratified design so that individuals from each national/ethnic del-egation attending the Festival was represented. Participants were informed that all responses wouldremain confidential. Nineteen participants used some level of translation assistance in filling outthe survey. After excluding incomplete responses, a total of 117 surveys were utilised (65 Male,49 Female, 3 specified as Other). All performers were between ages 18 and 75, with the majority(58%) ranging between ages 18–35. Regarding home region, 46 were Polynesian, 32 Melanesian,and 39 Micronesian.

Measures

A 24-item questionnaire (adapted from Clément and Noels 1992) addressed the expressed level ofsalience of both performers’ ethnic-level and meta-level identities (as Pacific Islanders) after havingperformed at the Festival as well as their retrospectively-rated salience of these identities prior to pre-paring to attend. 12 questions in the survey were posed, as below, regarding the pre- and post- Fes-tival cognitions and feelings:

To determine the reliability of the six pre- and six post- ethnic-level and metal-level identity sal-iences, Cronbach alphas were computed and ranged between .82 and .91, and the composite scoresfor these identities were thus used. All questions were framed in terms of 7-point rating scales fromstrongly disagree to strongly agree. Additional questions were posed about the performer’s gender,age, and experience with the Festival.

JOURNAL OF MULTILINGUAL AND MULTICULTURAL DEVELOPMENT 5

Results

Tests of RQ1

RQ1 asked whether participants experienced increases in their ethnic-level and meta-level identitysalience after having participated in the Festival relative to a point prior to having prepared forthe festival. To answer RQ1, we computed a two-way repeated-measures MANOVA (see Table 1)using the change variable between pre- and post- Festival measures to identify changes in how par-ticipants viewed their own identification with both meta- and specific ethnic-level identities. Toexplore this further, a two-way repeated-measures ANOVA was done. The first within-subject factorconsisted of participants’ ethnic-level identity salience (pre- versus post-Festival ethnic identity sal-ience) and the second within-subject factor consisted of participants’ meta-identity salience (pre-versus post-Festival meta-identity salience). A main effect was observed for ethnic-level identifi-cation, Λ = .78, F(1, 107) = 30.23, p < .001, h2

p .22; this increased from before preparing for the Fes-tival (M = 5.38, SD = 1.02) to after performing on-stage (M = 6.13, SD = 0.94). Similarly, a main effectwas observed for meta-identity, Λ = .39, F(1, 107) = 170.38, p < .001, h2

p .61. Meta-identity saliencealso increased from before the Festival (M = 4.42, SD = 1.11) to the time after performing at the Fes-tival (M = 5.97, SD = 1.21). The model also revealed a significant interaction between both types ofidentity, Λ = .61, F(1, 107) = 69.22, p < .001, h2

p .39. As displayed in Figure 2, the interaction involveda steeper increase in meta-identity salience and a shallower increase in ethnic-level identification.Although participants began with lower levels of recalled meta-identity salience compared to eth-nic-level identification before the Festival, this gap was reduced during the event due to the moremarked increases in meta-identity salience.

Tests of RQ2

RQ2 inquired how the experiences of festival-goers would vary based on ethnic-level identity. Toanswer this question, we conducted latent profile analyses in Mplus 7.3 (Muthén and Muthén

Table 1. Results of repeated measures MANOVA test for culture- andmeta-level identity variables; pre vs. post as repeated variable.

Item Df F p Item Df F p

I feel like I am [ethnic-level identity]… I feel like I am a Pacific Islander…Pre vs. Post 109 38.11 <.001 Pre vs. Post 109 38.11 <.001(Pre vs. Post) × Region 2.35 .10 (Pre vs. Post) × Region 2.35 .10I think I am [ethnic-level identity]… I think I am a Pacific Islander…Pre vs. Post 107 44.51 <.001 Pre vs. Post 107 44.51 <.001(Pre vs. Post) × Region 1.95 .15 (Pre vs. Post) × Region 1.95 .15I would describe myself as [ethnic-levelidentity]…

I would describe myself as PacificIslander…

Pre vs. Post 109 37.00 <.001 Pre vs. Post 109 37.00 <.001(Pre vs. Post) × Region 4.73 .01 (Pre vs. Post) × Region 4.73 .01I talk like I am [ethnic-level identity]… I talk like I am a Pacific Islander…Pre vs. Post 110 6.98 <.01 Pre vs. Post 110 6.98 <.01(Pre vs. Post) × Region .63 .54 (Pre vs. Post) × Region .63 .54The people I talk to would think I am[ethnic-level identity]…

The people I talk to think I am aPacific Islander…

Pre vs. Post 110 17.92 <.001 Pre vs. Post 110 17.92 <.001(Pre vs. Post) × Region 3.77 .03 (Pre vs. Post) × Region 3.77 .03I want others to describe me as [ethnic-level identity]…

I want others to describe me as aPacific Islander…

Pre vs. Post 107 32.84 <.001 Pre vs. Post 107 32.84 <.001(Pre vs. Post) × Region .62 .54 (Pre vs. Post) × Region .62 .54Composite Variable, means of 6 questions…

Composite Variable, means of 6questions…

Pre vs. Post 98 68.57 <.001 Pre vs. Post 98 68.57 <.001(Pre vs. Post) × Region 3.98 .02 (Pre vs. Post) × Region 3.98 .02

Note: Significant results in bold.

6 M. GILES ET AL.

1998–2014). Latent profile analysis (LPA) examines how participants respond to a series of continu-ous items. LPA is an inductive technique: profiles emerge based on participants’ actual response pat-terns rather than on predetermined response patterns specified by the researcher. In this study, twosets of LPA models were run. The first set of models used the 12 items asking about participants’identifications before the Festival (i.e. six pre-Festival specific identity items and six pre-Festivalmeta-identity items). Using these 12 items, five LPAs were run specifying between one and five latentprofiles. The second set of LPA models used the 12 items asking about participants’ identity afterhaving performed at the Festival (i.e. six items pertaining to specific ethnic identity after having per-formed at the Festival and six items pertaining to meta-identity after having performed at theFestival).

For questions asking about identity prior to attending the Festival, a two-profile solution wasdeemed most appropriate.2 Figure 3 depicts the latent profile plot for this two-profile solution.The first profile (comprising 59.9% of the sample) was labelled lower meta-identity. Participantsin this profile felt a mild identification with their specific ethnic identity but scored lower foridentification with their meta-identity as Pacific Islanders. The second profile (comprising 40.1%of the sample) was labelled highly identified on both identities. Participants in this profile indicatedhigh levels of identification with both their specific ethnic group and the meta-group before theFestival.

Figure 2. Net change over the timespan of festival for ethnic level (top line) and meta-identity (bottom line) salience.

Figure 3. Latent Profile Plot of Specific (i.e. Ethnic)- and Meta-Identification Before Attending the Festival.Note: SI = Specific Ethnic Identity. MI = Meta-Identity.

JOURNAL OF MULTILINGUAL AND MULTICULTURAL DEVELOPMENT 7

Regarding post-performance identity, a two-profile solution was judged most appropriate for thetime after having performed. Figure 4 depicts the latent profile plot for this two-profile solution. Thetwo profiles in this solution resembled the two profiles described previously. One profile (comprising87.0% of the sample) was highly identified on both identities; the other profile (comprising 13.0% ofthe sample) reported lower meta-identity compared to their specific ethnic identity. Thus, themajority (87%) found themselves experiencing a high level of salience in both their specific ethnicand meta-identity. This led to important questions about who specifically was identifying highlywith which identities before and after the Festival, which is explored further below.

A majority of participants (59.9%) fell into the lower meta-identity profile before the Festival, butonly a minority of participants (13.0%) fell into the lower meta-identity profile after having per-formed at the Festival. Conversely, a minority of participants (40.1%) could be characterised ashighly identified on both identities before the Festival, but this percentage increased to 87.0%when considering post-performance scores. Taken together, this suggests that many participantsmoved from the lower meta-identity profile to the highly identified on both identities profile overtime. To more systematically examine the extent to which participants remained in the sameprofile versus moved to the other profile, a latent transition analysis was conducted in Mplus 7.3.Latent transition analysis (LTA) allows researchers to examine the degree to which participantsmove (or transition) from one profile to another over time, versus the extent to which participantsremain in the same profile over time.

The LTA revealed that lower meta-identity participants before the Festival had a .20 probability ofremaining lower meta-identity participants after having performed at the Festival and a .80 prob-ability of transitioning to highly identified on both identities participants after having performed atthe Festival. On the other hand, highly identified on both identities participants before the Festivaldemonstrated virtually no likelihood of transitioning to lower-meta identity participants after havingperformed at the Festival (.02 probability). Rather, participants who were highly identified acrossboth specific and meta-identities before the Festival were almost certain to remain in the sameprofile over time (.98 probability of remaining in the same profile). This demonstrates not onlythe low likelihood of deidentification after having participated in the Festival, but also the dramaticeffect that the Festival had on raising metagroup identity salience for the majority of participants (butnot all participants).

This latter finding raised another question about whether or not participants with specific ethnicidentities differed from one another in terms of their profile memberships from the time before theFestival to the time after having performed at the Festival. No specific identity group demonstratedan increase in the number of participants in the lower meta-identity group from the time before the

Figure 4. Latent Profile Plot of Specific (i.e. Ethnic)- and Meta-Identification After Having Performed at the Festival.Note: SI = Specific Ethnic Identity. MI = Meta-Identity.

8 M. GILES ET AL.

Festival to the time after having performed at the Festival. Rather, all specific ethnic identity groupseither saw no change in their participants’ profile membership over time or an increase in the per-centage of participants in the highly identified on both identities profile over time. For example,French Polynesians were almost evenly split between the two profiles before the Festival (n = 7 inthe lower meta-identity profile, n = 6 in the highly identified across both identities profile) but wereuniformly in the highly identified across both identities profile after having performed at the Festival(n = 13). Perhaps even more notable, New Zealanders were almost all in the lower meta-identityprofile before the Festival (n = 11, with n = 1 in the highly identified across both identities profile),but all New Zealanders were in the highly identified across both identities profile after having per-formed at the Festival. Similarly, all Samoans in the sample transitioned from the lower meta-identityprofile before the Festival to the highly identified across both identities profile after having performedat the Festival. These results suggest that participants with certain specific ethnic identities (mostnotably, French Polynesians, New Zealanders, and Samoans) demonstrated marked shifts to thehighly identified on both identities profile.

Chamorros, Micronesians, and Guamanians demonstrated a different pattern. Some participantsin these three specific ethnic groups started out in the lower meta-identity profile and transitioned tothe highly identified on both identities profile over time. However, these three ethnic groups areunique in that at least half of all participants who began in lower meta-identity profile before the Fes-tival remained in the lower meta-identity profile after having performed at the Festival. In otherwords, participants in these three ethnic groups did not demonstrate as seismic of shifts to the highlyidentified on both identities profile over time. Rather, many participants who started in the lowermeta-identity profile remained stagnant in this profile over time, thereby contributing to the overall.20 probability of remaining in the lower meta-identity profile over time.3

Comparisons of identity salience change

Given that some groups within the Pacific experienced similar changes across their different iden-tities at the Festival, the way in which these groups relate to one another in terms of identity sal-ience change merits further inquiry. A 2×3 chi-square was therefore run with participants’ mostlikely profile membership after having performed at the Festival (two categories: lower meta-iden-tity and highly identified on both identities) and broader geographical region (three categories:Polynesian, Micronesian, and Melanesian) as the two nominal variables. The test was significant,χ2 (2) = 11.84, p = .003, meaning that participants in the three broader geographic regions system-atically differed in the frequency with which they fell under the two profiles. As Table 2 illustrates,Polynesians almost all fell in the highly identified on both identities profile after having performedat the Festival (98%), whereas only one Polynesian (2% of all Polynesians) fell in the lower meta-identity profile. Polynesians contrasted most sharply with Micronesians: 29% of Micronesiansremained in the lower meta-identity profile after having performed at the Festival, and only71% of Micronesians fell in the highly identified on both identities profile after having performedat the Festival.

Table 2. Mean levels of meta-identity salience for Festival participants across regions and time.

Participantgroup

Identity salience

F η2Prior to preparing for the

festivalAfter performing at the

festivalTotal change between pre-and

post-

Polynesian 3.29(1.43)

5.26(0.87)

1.97 13.96 .247

Micronesian 3.51(1.03)

4.30(1.55)

0.79 13.96 .247

Melanesian 3.27(1.12)

4.79(1.36)

1.52 13.96 .247

Note: Standard deviations appear in parentheses below means. For all groups, p < .05.

JOURNAL OF MULTILINGUAL AND MULTICULTURAL DEVELOPMENT 9

Given these systematic differences, two mixed-model analyses were run to further explore therelationship between Melanesian, Micronesian, and Polynesian salience of ethnic-level and meta-identities. The first MANOVA used region as the between-subjects factor and ethnic-level identityas the repeated within-subjects factor. The model revealed a significant interaction effect between thetwo factors, Λ = .94, F(2, 105) = 3.23, p = .04, h2

p .07. Inspection of 95% confidence intervals revealedthat participants from Polynesia experienced stronger identification with their ethnic-level groupduring the festival (M = 6.10, SE = 0.14, 95% CI: 5.82–6.37) than they did before the festival (M =5.13, SE = 0.15, 95% CI: 4.83–5.42). Participants from Melanesia also experienced stronger identifi-cation with their specific ethnic-level group during the Festival (M = 6.27, SE = 0.17, 95% CI: 5.93–6.61) than they did before the Festival (M = 5.51, SE = 0.18, 95% CI: 5.15–5.87). However, partici-pants from Micronesia did not experience changes in their ethnic identity from before the festival(M = 5.61, SE = 0.18, 95% CI: 5.26–5.97) to the time in which they were at the festival (M = 6.06,SE = 0.17, 95% CI: 5.72–6.40).

The second mixed-model ANOVA consisted of geographic origin as the between-subjects factorand meta-identity as the repeated within-subjects factor. A significant two-way interaction emergedin this model, Λ = .88, F(2, 105) = 7.20, p = .001, h2

p .12. Festival-goers from Polynesia experiencedgreater meta-identity salience during the Festival (M = 6.29, SE = 0.17, 95% CI: 5.94–6.63) thanbefore the Festival (M = 4.36, SE = 0.16, 95% CI: 4.03–4.69). Festival-goers from Melanesia alsoexperienced greater identification with their ethnic-level group during the Festival (M = 5.88, SE= 0.21, 95% CI: 5.46–6.30) compared to before the Festival (M = 4.32, SE = 0.20, 95% CI: 3.92–4.72). Finally, festival-goers from Micronesia also experienced increases in their meta-identity sal-ience during the Festival (M = 5.58, SE = 0.21, 95% CI: 5.16–6.00) compared to before the Festival(M = 4.62, SE = 0.20, 95% CI: 4.22–5.02). Hence, people from all three geographic regions benefittedfrom the Festival in terms of enhanced meta-identity salience, but participants from Polynesiaexperienced the greatest increases (an increase of 1.93 points on the 7-point scale), followed by par-ticipants from Melanesia (a 1.56-point increase), and ending with participants from Micronesia (a0.96-point increase) (Tables 3 and 4).

Hence, participants from Polynesia and Melanesia experienced relatively large increases in theirethnic-level and meta-identity salience during the Festival compared to the time before the Festival.Conversely, participants from Micronesia experienced no change in their ethnic-level identity;

Table 3. Profile breakdown of participants according to broader geographical region.

Time before the festival Time after having performed at the festival

Geographicalregion

Lowermeta-identity

% Lowermeta-identity

Highlyidentified on

bothidentities

% Highlyidentified on

bothidentities

Lowermeta-identity

% Lowermeta-identity

Highlyidentified on

bothidentities

% Highlyidentified on

bothidentities

Polynesia n = 32 70% n = 14 30% n = 1 2% n = 45 98%Micronesia n = 14 45% n = 17 55% n = 9 29% n = 22 71%Melanesia n = 19 61% n = 12 39% n = 4 13% n = 27 87%

Table 4. Mean levels of ethnic-level identity salience for Festival participants across regions and time.

Participantgroup

Identity salience

F η2Prior to preparing for the

festivalAfter performing at the

festivalTotal change between pre-and

post-

Polynesian 4.18(1.05)

5.14(1.83)

0.96 2.97 .065

Micronesian 4.55(1.02)

4.90(0.42)

0.35 2.97 .065

Melanesian 4.56(0.93)

5.27(0.86)

0.71 2.97 .065

Note: Standard deviations appear in parentheses below means. For all groups, p < .05.

10 M. GILES ET AL.

although Micronesians’meta-identity increased, this increase was not as large as the increase for par-ticipants from Polynesia and Melanesia. These results are discussed below (Figure 5).

Discussion

From an intergroup communication perspective (see Giles and Harwood 2018; Giles and Maass2016), the Festival of Pacific Arts takes a noteworthy approach to building a meta-identity amongstdisparate groups. Because all participants presented their own unique cultural arts, the Festival runsthe risk of raising ethnic-level identity at the expense of meta-identity. However, the documentedincrease in identity salience for both ethnic and meta-identities demonstrates the effectiveness ofthis strategy for mutually increasing group vitality as a result of intergroup interaction. This increasein both identities demonstrates that the concept of functional antagonism needs to be reconsideredin light of complementary and nested identities. Social identity complexity theory (Roccas andBrewer 2002) could provide a valuable framework for this phenomenon.

Because participants were asked about both ethnic-level identity and meta-identity salience sim-ultaneously, an increase in both identities indicates that the participant had experiences which con-tributed to the salience of multiple identities. Given that all groups of participants evinced anincrease in level of identity salience for both ethnic and meta-identity, the Festival was successfulat its stated goal of increasing the sense of unity for all participants.

It is important to note that even though identity salience was increased for both ethnic and meta-identities in the sample, the effect of festival participation was more effective for some groups thanothers. To understand the way that different groups within the Pacific were building identities rela-tive to the Festival, we found different profiles of how identity salience changed between the two timepoints. Although these profiles did not precisely match the designations of Polynesian, Melanesian,and Micronesian, there were striking similarities within these classifications across their respectivegroups contained therein, prompting analysis within that framework. The similarities in changebetween the Polynesian participants from Samoa and New Zealand suggested that regional classifi-cation could shape how participant identity was experienced.

Both Melanesians and Polynesians indicated a significant increase in the salience of their ethnicand meta-identities, which was not as pronounced for Micronesians. This could be due to a proto-typicality bias in the context of Pacific performances (Bettencourt et al. 2001), which favours Poly-nesians among the various groups of Pacific Islanders. It is similarly possible that the experience oftravelling across the Pacific to another nation for the Festival has a net-positive mediating effect, orthat hosting the Festival in one’s home nation has a deleterious mediating effect. Although partici-pation in the Festival resulted in a statistically significant net change for all groups, these differencesbetween regional groups merit further inquiry and explanation.

Figure 5. (a) Net change over the timespan of festival for ethnic-level salience for different groups within the Pacific. (b) Net changeover the timespan of festival for meta-identity salience for different groups within the Pacific.

JOURNAL OF MULTILINGUAL AND MULTICULTURAL DEVELOPMENT 11

The usage of the terms Melanesian, Micronesian, and Polynesian to signify individual culturaltypes remains controversial due to the outdated and counterfactual logics that helped establishthe distinctions among Melanesians, Polynesians, and Micronesians. However, the prominence ofthese labels as culturally-meaningful regional types has persisted despite various substantial changesin interpretation and intellectual value judgements. These discredited ethnological typifications con-tinue to live on in political, academic, and social realms. Outside of academic circles, the terms havebeen integrated into the everyday life for the people from these regions, and any well-intentionedattempt to decrease their usage is disconnected from the lived reality of Melanesians as self-identify-ing Melanesians (Lawson 2013). The results from this study suggest that at the Festival of PacificArts, these identifiers are linked to a common experience which is felt similarly by group memberswithin their group, but differently from other groups. One of the most important findings from thisstudy is that Micronesians are experiencing the Festival differently than Melanesians or Polynesians,and further efforts should be made to understand why these differences occur. If the goal of the Fes-tival is to build a unifying Pacific Islander identity, then the fact that this goal is being achieved atdisparate levels of efficacy by different culture groups provides information about the nature ofsocially identifying distinctions and the way that the Festival operates.

Limitations of this study suggest the benefit of longitudinal research. Because participants wereonly able to be contacted at the Festival, participants completed the surveys by reflecting on their pre-vious experiences from ‘before preparing for the Festival’. Clearly, it would be advantageous for par-ticipants to be contacted prior to, during, and multiple times after an event to gauge long-termeffectiveness of change in identity salience. Interviews with, and focus-group discussions among, par-ticipants about their feelings towards and practices about enacting ethnic-level and meta-identitieswould yield important discursive data for in-depth analysis. Participants were given the opportunityfor follow-up interviews, but due to the drop-out rate, an appropriate sample size could not beachieved for the post-study. Due to the regional focus of the Festival, findings about the efficacy ofcultural festivals in spurring identity salience may not translate across regional or cultural boundaries.It should be noted as well that within the different classifications, there were a range of responseswithin the group. More specific work could be done to investigate these variances on a micro level.

This study supports the possibility of multiple identity salience and demonstrates that ethnic andmeta-identities can both become more salient without having a negative effect on the other. In thecontext of politically-divided nations, or environments where minority cultures can be seen as athreat to the larger national meta-identity, this indicates the possibility for social interventionsthat strengthen and make salient multiple distinct identities without negative interactions betweenthe two. Future research attending to demographic variables such as gender and status as performeror audience member needs to attend to different kinds of festivals globally as well as examine morefocally how participants enact their meta-identities contrasted against other salient identities. Suchresearch, when appropriately framed, will provide valuable information on how festivals are rep-resented across different media and create and contribute to a group’s relative ethnic-level as wellas meta-identities and, thereby, enhance vitality theory development with regard to the inclusionof superordinate identities (Abrams, Eveland, and Giles 2003; Harwood, Giles, and Bourhis 1994;Smith, Ehala, and Giles 2018).

Data availability

The data that support the findings of this study are available from the corresponding author uponrequest.

Notes

1. The way in which these identities relate in broader Pacific context, and how participants labelled themselves atthe Festival is discussed in-depth in the Methods section. Participants were asked to provide the name of their

12 M. GILES ET AL.

own ethnic-level ethnic category for analytical purposes. Throughout this manuscript, we refer to these categ-orizations as ethnic identity groups, with a recognition that ethnic, cultural, and national identity are not synon-ymous. However, throughout the Festival the participants are divided up into delegations based on thesegroupings. When participant responses were analyzed, some key trends emerged which correlate to regionalgroupings within Oceania that are constantly made salient during the Festival.

2. A table of fit indices for all ten LPAmodels are available from the authors upon request. The two-profile modelswere chosen after examining changes in a range of fit indices and weighing conceptual considerations. Extensivework was completed to decide on these specific profiles as well as the LTAs. Both the dataset and the analysesrun are available upon request from the first author.

3. For the sake of space, we discussed three examples of how participants transitioned profiles over time (e.g., theresults for New Zealanders) and three examples of how participants stayed in the same profiles over time (e.g.,the results for Guamanians). A full table detailing how participants in all other ethnic groups transitionedbetween profiles across time is available upon request.

Acknowledgements

Overwhelming thanks to the Festival organisers for their efforts in organising and shepherding the Festival, withespecial gratitude to both Elise Huffer who encouraged the delegations to participate in this research and to LilnabethSomera, who provided very helpful local support during the Festival. Thanks to Karen Myers and Daniel Linz forassistance in crafting the survey instrumentsWe appreciate greatly the careful feedback and assistance from two anon-ymous reviewers, as well as very influential support from the editor.

Disclosure statement

No potential conflict of interest was reported by the authors.

References

Abdelal, R., Y. M. Herrera, A. I. Johnston, and R. McDermont. 2006. “Identity as a Variable.” Perspectives on Politics 4:695–711.

Abrams, J., W. P. Eveland Jr., and H. Giles. 2003. “The Effects of Television on Group Vitality: Can TelevisionEmpower Nondominant Groups?” In Annals of the International Communication Association, 27, 193–219.Mahwah, NJ: Erlbaum.

Bauman, R. 1992. “Performance.” In Folklore, Cultural Performances and Popular Entertainments, edited by R.Bauman, 133–151. New York, NY: Oxford University Press.

Bellwood, P., G. Chambers, M. Ross, and H. C. Hung. 2011. “Are ‘Cultures’ Inherited? Multidisciplinary Perspectiveson the Origins and Migrations of Austronesian-speaking Peoples Prior to 1000 BC.” In Investigating ArchaeologicalCultures, edited by B. W. Roberts and M. Vander Linden, 321–354. New York, NY: Springer.

Benet-Martínez, V., and J. Haritatos. 2005. “Bicultural Identity Integration (BII): Components and PsychosocialAntecedents.” Journal of Personality 73: 1015–1050.

Berry, J. W. 1990. “Psychology of Acculturation.” In The Culture and Psychology Reader, edited by N. R. Goldbergerand J. B. Veroff, 457–488. New York, NY: New York University Press.

Bettencourt, B. A., K. Charlton, N. Dorr, and D. L. Hume. 2001. “Status Differences and in-Group Bias: A Meta-ana-lytic Examination of the Effects of Status Stability, Status Legitimacy, and Group Permeability.” PsychologicalBulletin 127: 520–542.

Bonnell, V. 1980. Roots of Rebellion: Workers’ Politics and Organizations in St. Petersburg and Moscow 1900–1914.Cambridge, MA: Cambridge University Press.

Brewer, M. B. 2001. “The Many Faces of Social Identity: Implications for Political Psychology.” Political Psychology 22:115–125.

Burke, P. J., and D. C. Reitzes. 1981. “The Link between Identity and Role Performance.” Social Psychology Quarterly44: 83–92.

Chang, C. S., H. L. Liu, X. Moncada, A. Seelenfreund, D. Seelendfreund, and K. F. Chung. 2015. “A Holistic Picture ofAustronesian Migrations Revealed by Phylogeography of Pacific Paper Mulberry.” Proceedings of the NationalAcademy of Sciences 112: 13537–13542.

Clément, R., Z. Dornyei, and K. A. Noels. 1994. “Motivation, Self-confidence, and Group Cohesion in the ForeignLanguage Classroom.” Language Learning 44: 417–448.

Clément, R., and K. A. Noels. 1992. “Towards a Situated Approach to Ethnolinguistic Identity: The Effects of Status onIndividuals and Groups.” Journal of Language and Social Psychology 11: 203–232.

JOURNAL OF MULTILINGUAL AND MULTICULTURAL DEVELOPMENT 13

Clément, R., K. A. Noels, and B. Denault. 2001. “Inter-ethnic Contact, Identity, and Psychological Adjustment: TheMediating and Moderating Roles of Communication.” Journal of Social Issues 57: 559–577.

Clément, R., S. Singh, and S. Gaudet. 2006. “Identity and Adaptation Among Minority Indo-Guyanese: Influence ofGenerational Status, Gender, Reference Group and Situation.” Group Processes and Intergroup Relations 9: 289–304.

Crisp, R. J., and M. Hewstone. 2001. “Multiple Categorization and Implicit Intergroup Bias: Differential CategoryDominance and the Positive-negative Asymmetry Effect.” European Journal of Social Psychology 31: 45–62.

Dion, K. K., and K. L. Dion. 2001. “Gender and Cultural Adaptation in Immigrant Families.” Journal of Social Issues 57:511–521.

Fishman, J. A. 1977. “Language and Ethnicity.” In Language, Ethnicity and Intergroup Relations, edited by H. Giles, 15–157. London, UK: Academic Press.

Garcia, J. L., J. I. Kosberg, W. P. Mangum, N. Henderson, and C. C. Henderson. 2015. “Caregiving for and by HispanicElders: Perceptions of Four Generations of Women.” The Journal of Sociology and Social Welfare 12: 169–188.

Getz, D. 1993. “Corporate Culture in Not-for-profit Festival Organizations.” Festival Management and Event Tourism1: 11–17.

Giles, H., R. Y. Bourhis, and D. M. Taylor. 2009. “Towards a Theory of Language in Ethnic Group Relations.” InSociolinguistics: Critical Concepts in Linguistics, edited by N. Coupland, 97–132. New York, NY: Routledge.

Giles, H., and J. Harwood. 2018. The Oxford Encyclopedia of Intergroup Communication, Vol. 1. New York, NY: OxfordUniversity Press.

Giles, H., and A. Maass. 2016. Advances in Intergroup Communication. New York, NY: Peter Lang.Glowczewski, B., and R. Henry. 2011. “Dancing with the Flow: Political Undercurrents at the 9th Festival of Pacific

Arts, Palau 2004.” In Challenge of Indigenous Peoples. Spectacle or Politics?, edited by B. Glowczewski and R.Henry, p. 159. Oxford: Bardwell Press.

Harwood, J., H. Giles, and R. Y. Bourhis. 1994. “The Genesis of Vitality Theory: Historical Patterns and DiscoursalDimensions.” International Journal of the Sociology of Language 108: 168–206.

Haùofa, E. 1994. “Our Sea of Islands.” The Contemporary Pacific 6: 147–161.Henry, R., and L. Foana’ota. 2015. “Heritage Transactions at the Festival of Pacific Arts.” International Journal of

Heritage Studies 21 (2): 133–152.Hermann, E., ed. 2011. Changing Contexts, Shifting Meanings. Transformations of Cultural Traditions in Oceania.

Honolulu: University of Hawai’i Press.Hogg, M. A. 2006. “Social Identity Theory.” In Contemporary Social Psychological Theories, edited by P. J. Burke, p.

144. Stanford, CA: Stanford University Press.Hogg, M. A., D. Terry, and K. White. 1995. “A Tale of Two Theories: A Critical Comparison of Identity Theory with

Social Identity Theory.” Social Psychology Quarterly 58: 255–269.Jepperson, R. L., A. Wendt, and P. J. Katzenstein. 1996. “Norms, Identity, and Culture in National Security.” In The

Culture of National Security, edited by P. J. Katzenstein, 33–75. New York, NY: Columbia University Press.Jolly, M., S. Tcherkézoff, and D. Tryon. 2009. “Oceanic Encounters: A Prelude.” In Oceanic Encounters: Exchange,

Desire, Violence, edited by M. Jolly, S. Tcherkezoff, and D. Tryon, 1–36. Canberra: ANU Press.Lawson, S. 2013. “‘Melanesia’: The History and Politics of an Idea.” The Journal of Pacific History 48: 1–22.Leff, C. S. 1997. The Czech and Slovak Republics: Nation vs. State. Boulder, CO: Westview Press.Mackley-Crump, J. 2015. The Pacific Festivals of Aotearoa New Zealand: Negotiating Place and Identity in a New

Homeland. Honolulu, HI: University of Hawai’i Press.Mendoza, Z. S. 1993. Shaping society through dance: Mestizo ritual performance in the Peruvian Andes. Chicago, IL:

University of Chicago Press.Mirabal, S., A. M. Cadenas, R. Garcia-Bertrand, and R. J. Herrera. 2013. “Ascertaining the Role of Taiwan as a Source

for the Austronesian Expansion.” American Journal of Physical Anthropology 150: 551–564.Mullen, B. 1991. “Group Composition, Salience, and Cognitive Representations: The Phenomenology of Being in a

Group.” Journal of Experimental Social Psychology 27: 297–323.Mullen, B., M. J. Migdal, and D. Rozell. 2003. “Self-awareness, Deindividuation, and Social Identity: Unraveling

Theoretical Paradoxes by Filling Empirical Lacunae.” Personality and Social Psychology Bulletin 29: 1071–1081.Muthén, B., and L. Muthén. 2014. Mplus User’s Guide. 7th ed. Los Angeles, CA: Muthén & Muthén.Noels, K. A. 2014. “Language Variation and Ethnic Identity.” Language and Communication 35: 88–96.Noels, K. A., and R. Clément. 2015. “Situational Variations in Ethnic Identity Across Immigration Generations:

Implications for Acculturative Change and Cross-cultural Adaptation.” International Journal of Psychology 50:451–462.

Noels, K. A., R. Clément, and S. Gaudet. 2004. “Language and the Situated Nature of Ethnic Identity.” In LanguageMatters: Culture, Identity, and Communication, edited by S. H. Ng, C. N. Candlin, and C. Y. Chiu, 245–266.Hong Kong: Hong Kong City University Press.

Noyes, D. 2003. Catalan Festival Politics After Franco. Philadelphia, PA: University of Pennsylvania Press.Oakes, P. J. 1987. “The Salience of Social Categories.” In Rediscovering the Social Group: A Self-categorization Theory,

edited by J. C. Turner, M. A. Hogg, P. J. Oakes, S. D. Reicher, and M. S. Wetherell, 117–141. Oxford, UK: Blackwell.Ouchi, W. 1981. Theory Z: How American Business Can Meet the Japanese Challenge. Reading, MA: Addison-Wesley.

14 M. GILES ET AL.

Park, B., and M. Rothbart. 1982. “Perception of Out-group Homogeneity and Levels of Social Categorization: Memoryfor the Subordinate Attributes of in-group and Out-group Members.” Journal of Personality and Social Psychology42: 1051–1068.

Roccas, S., and M. B. Brewer. 2002. “Social Identity Complexity.” Personality and Social Psychology Review 6: 88–106.Rouhana, N. N. 1997. Palestinian Citizens in an Ethnic Jewish State: Identities in Conflict. New Haven, CT: Yale

University Press.Sapir, E., and D. G. Mandelbaum. 1968. Selected Writings of Edward Sapir in Language, Culture and Personality.

Berkeley, CA: University of California Press.Siegelbaum, L. H., and R. G. Suny. 1994. Making Workers Soviet: Power, Class and Identity. Ithaca, NY: Cornell

University Press.Smith, B., M. Ehala, and H. Giles. 2018. “Vitality Theory.” In The Oxford Encyclopedia of Intergroup Communication,

Vol. 2, edited by H. Giles, and J. Harwood, 485–500. New York, NY: Oxford University Press.Stets, J. E., and P. J. Burke. 2000. “Identity Theory and Social Identity Theory.” Social Psychology Quarterly 63: 224–

237.Stokes, S. 1995. Cultures in Conflict: Social Movements and the State in Peru. Berkeley, CA: University of California

Press.Swann Jr., W. B., A. Gamez, C. D. Seyle, F. J. Morales, and C. Huici. 2009. “Identity Fusion: The Interplay of Personal

and Social Identities in Extreme Group Behavior.” Journal of Personality and Social Psychology 96: 995–1011.Tajfel, H., and J. C. Turner. 1979. “An Integrative Theory of Intergroup Conflict.” In The Social Psychology of

Intergroup Relations, edited by W. G. Austin, and S. Worchel, 33–47. Monterey, CA: Brooks-Cole.Tcherkézoff, S. 2003. “A Long and Unfortunate Voyage Towards the ‘Invention’ of the Melanesia/Polynesia

Distinction 1595–1832 Translated From French by Isabel Ollivier.” The Journal of Pacific History 38: 175–196.Thoits Jr., P. A., and L. K. Virshup. 1997. “Me’s and We’s: Forms and Functions of Social Identities.” In Self and

Identity: Fundamental Issues 1, edited by R. Ashmore and L. Jussim, 106–133. New York, NY: OxfordUniversity Press.

Thomas, N. 1989. “The Force of Ethnology: Origins and Significance of the Melanesia/Polynesia Division.” CurrentAnthropology 30 (1): 27–41.

Turner, J. C., M. A. Hogg, P. J. Oakes, S. D. Reicher, and M. S. Wetherell. 1987. Rediscovering the Social Group: A Self-categorization Theory. Oxford, UK: Blackwell.

JOURNAL OF MULTILINGUAL AND MULTICULTURAL DEVELOPMENT 15