Urban Studies, Vol. 44, No. 4, 699–721, April 2007

Constructing Racism in Sydney, Australia’s Largest EthniCity

James Forrest and Kevin Dunn

[Paper first received, March 2005; in final form, June 2006]

Summary. Contemporary Australia is in a contradictory situation as a nation where multiculturalism co-exists with various forms of what are collectively called racisms. Based on a survey of Sydney residents, this study uses a social constructivist approach to investigate the nature and sociospatial context of racist attitudes in Sydney, Australia’s largest EthniCity. Results show a mix of compositional (aspatial) and contextual (spatial) associations with racisms. The former indicate a general but inconsistent relationship between socioeconomic status and tolerance, and also between cultural diversity and tolerance. The latter, however, reveal place- based cultures of tolerance and intolerance cutting across compositional relationships. A geography of racism in Sydney therefore adds a level of understanding which cannot be obtained from aspatial analysis alone. This helps to understand the complexity of local political cultures and can assist with the formulation of anti-racism interventions.

Introduction Racism is an historical and complex societal of cultural pluralism prior to the 1960s with problem among settler societies such as two ‘charter groups’—English and French— Australia, Canada, Israel, the US and New and a more recent (post-1962 and 1967 Zealand. All are countries where immigration changes to immigration laws) multicultural has long been a significant factor in population approach to an increasingly diverse, post- growth and, importantly, where the wide range ‘White Canada’ ethnic mosaic (Bourque and of national origins of more recent immigration Duchastel, 1999). Canada went further than streams has resulted in increasingly ethnically the other immigrant-receiving countries men- diverse populations. Yet each country is differ- tioned here in enshrining multiculturalism ent. In the US, a dominant ethnic group phase legally and constitutionally during the 1980s of Americanism saw the emergence of a (Helmes-Hay and Curtis, 1998). In New ‘White nation’ until the mid 1960s (Kaufman, Zealand, the issue is largely one of bicultural- 2004) and a potential future divide between ism involving the indigenous Maori and ‘Blacks’ and ‘non-Blacks’ (Rose, 1997), but Pakeha (non-Maori, but usually seen as with the social position of Asians and Hispa- ‘White’ New Zealanders) with, as yet, little nics, admitted after changes to immigration regard for the growing cultural diversity of laws in 1965, not yet resolved (Kivisto, the country’s people since the ending of a 2002). In Canada, issues of a core culture and ‘White New Zealand’ immigration policy in ethnic minorities assumed a particular form 1986 (Hiebert et al., 2003).

James Forrest is in the Department of Human Geography, Macquarie University, New South Wales, Sydney 2109, Australia. Fax: þ61 2 9850 6052. E-mail: [email protected]. Kevin Dunn is in the School of Biological, Earth and Environmental Sciences, University of New South Wales, Sydney 2052, New South Wales, Australia. Fax: þ61 2 9385 2067. E-mail: [email protected]. 0042-0980 Print=1360-063X Online=07=040699–23 # 2007 The Editors of Urban Studies DOI: 10.1080=00420980601185676 700 JAMES FORREST AND KEVIN DUNN

Contemporary Australian society is often elsewhere, is also unsettled by its dynamism characterised as increasingly multicultural, but and development. It remains an unfinished still struggling to disengage from a legacy of project, with gaps in its coverage and limits to Anglo privilege and cultural dominance its reach in some sections and demographics. (Forrest and Dunn, 2006a). Exclusion of non- It is a feature of each of these major immi- Europeans, embodied in an Immigration grant receiving countries, however, that the Restriction Act, 1901, was one of the earliest major focus of dominant culture–minority pieces of legislation passed by the federal ethnic groups tensions is in the cities and parliament of the new Commonwealth of usually in the largest cities. In New Zealand, Australia. What became known as the White Auckland stands out as the most culturally Australia policy lasted until the early 1970s, diverse city in the country (Johnston et al., to be replaced by policies promoting multi- 2002, 2003). In Canada, the major immigrant- culturalism. Even so, the experience of receiving cities are Toronto, Montreal and post-World-War-2 non-English-speaking back- Vancouver (Simmons and Bourne, 2003). In ground immigrants, especially from eastern and the US, the focus on urban areas is more wide- southern Europe during the 1950s and 1960s— spread, but is an urban focus nonetheless as opposed to the English-speaking background (Johnston et al.,2004).InAustralia,amajor (British) immigrants who had absolutely domi- feature of the immigrant stream of the past 50 nated migration flows for 150 years prior to the years has been the degree to which it has 1950s—was frequently marked by discrimi- become concentrated in the major urban areas. nation and socioeconomic disadvantage Sydney’s population comprises some 43 per (Vasta and Castles, 1996, p. 4). Subsequently, cent of immigrants from a non-English- post-White-Australia immigrants from Asia, speaking background by ancestry and especially those who were Muslims, along Melbourne’s 41 per cent. Over the period with Indigenous Australians, came to be 1996–2001, some 39 per cent of new immigrant especially identified as key Others in the arrivals came to Sydney, with 22 per cent going national imaginary (Hamilton, 1990; to Melbourne and 15 per cent to Perth (Forrest Rajkowski, 1987; Rizvi, 1996, pp. 176–177). et al., 2003). Australia is thus in a contradictory situation Sydney is, therefore, Australia’s major where multiculturalism co-exists with various immigrant receiving city, especially in terms forms of what are collectively called racisms of the large number of Asian immigrants (Vasta and Castles, 1996, p. 5). The latter coming into Australia since the early 1970s include Anglo-Celtic cultural dominance, intol- (Forrest et al., 2003; Poulsen et al., 2004). erance of diversity, antagonism towards some Here, racial tensions are most apparent cultural groups and xenophobia. This contradic- (Forrest and Dunn, 2006a). Based on a tion may not be unique, but rather common survey of Sydney residents on attitudes to among settler societies and among all those various aspects of racism—part of a wider that have experienced substantial immigration study of racist attitudes in eastern Australia in the past four decades, such as countries in (Dunn et al., 2004)—this study uses a social western Europe. These nations have what are constructivist approach to investigate the sometimes called ‘unsettled multiculturalisms’ nature and sociospatial context of racist atti- (Hesse, 2000). The unsettledness relates to com- tudes in Sydney. Several questions are peting ideas about nation, including earlier more posed: is there a culture of racism in exclusive encapsulations of nationhood. These Sydney; if so, how is it constructed and posi- legacies persist, often in a minority context, per- tioned across key social factors such as ethni- formed in nostalgic and tragic ways but with city, class and age; is there any geography to exclusionary impacts (Dunn, 2005), or explod- that culture which might be used to tailor ing sensationally in ‘race riots’ such as around approaches to anti-racism initiatives? Sydney’s Cronulla Beach on 11 December, There is a long-established tradition of 2005. But multiculturalism in Australia, as examining variations in attitudes to cultural CONSTRUCTING RACISM IN SYDNEY 701 diversity across social groups, which has been Another class-based explanation points to explored most fully in contemporary times by effects of affluence, and specifically resource social psychologists (for example, see Pedersen competition, among the working class. There et al., 2000). However, scholarship on spatial is a perception, that, because of a paucity of variations to these attitudes has been social capital—education, qualifications or decidedly sparse and largely non-existent in non-recognition of qualifications, and also recent decades (Pettigrew, 1959; Robinson, time of arrival—members of minority ethnic 1987; Schaefer, 1975). Contemporary communities form a large proportion of the geographical scholarship has excelled at iden- working class (Jupp, 1984, p. 11) where they tifying how racialisation operates within place are sometimes seen as an ‘industrial reserve and how racialisation is placed (Bonnett, army’ acting to depress all workers’ incomes 1996; Durrheim and Dixon, 2001), but the (Collins, 1984; Lever-Tracey and Quinlan, study of spatial variation in racist attitudes 1988; McAllister and Kelley, 1984, pp. 53– and experiences, and of racialisation, has 54). Such a competition-for-jobs basis to been neglected. Yet, as shown in the next racism is accentuated in the current climate section, contemporary geographical theory of economic restructuring, resulting in job dis- makes a compelling case for research on the placement and marginalisation, especially in spatially varied nature of community relations the manufacturing sector where most immi- and nationalism. Furthermore, geographers grants have traditionally been concentrated have clearly advocated the importance of (Vasta and Castles, 1996, pp. 38–40). such spatial variations to the formulation and Racist attitudes are also associated with a treatment of anti-racism. The clearest articula- form of national ethnocentrism, in which tion of this comes from Kobayashi and ‘Australianness’ is tightly linked to Anglo Peake’s (2000) programmatic statements (or Anglo-Celtic) culture (Dixson, 1999; on anti-racist geographical scholarship. Johnson, 2002). Ethnocentrism derives from Geographical variations in racism across a an assumption of a pre-existing culture and city mandate a geography of anti-racism. society to which newcomers are expected to This research on the geographies of racism conform, although it may not be overtly pro- across Sydney is an empirical study pointing claimed, nor even intended to be oppressive ultimately towards a geography of anti- (see Kobayashi and Peake, 2000, p. 393). racism for that city, using techniques that Nonetheless, it involves intolerance of cul- may be replicated for other, culturally tural difference and of minorities and, in its diverse cities in Western-settler societies. more extreme forms, the superiority of one’s own ethnicity: an ethnocultural or assimila- tionist viewpoint. Multiculturalism, or liberal Theorising Racism in Australia egalitarianism is, of course, an alternative to Among a range of views about racism, it has this viewpoint (Kelly, 2002; Kivisto, 2002). long been accepted that it is negatively associ- Other scholars of racism emphasise the ated with affluence and educational attainment effect of cultural mix in an area, or the lack (see Nunn et al., 1978; Smith, 1981). But the of it, as respectively inhibiting or promoting negative association between racist attitudes intolerance (Valenty and Sylvia, 2004). The and class (as measured through educational combined effect of these viewpoints is a achievement and affluence variables) is not a complex mix of possible outcomes, reflecting straightforward indicator of a link between community relations likely to operate differ- racism and interclass conflict. For example, ently in various parts of the city, even, educational achievement among the middle perhaps, cutting across social divides. class may have a much greater impact upon Contemporary logics impacting on the for- the expression of racist sentiment than it mation of racist attitudes (the older forms does upon oppressive actions by those indi- were largely sociobiological), operate viduals (see Yinger, 1986, pp. 36–37). through differentiation, which underpins 702 JAMES FORREST AND KEVIN DUNN separation and exclusion of racialised groups overall response rate was 70.3 per cent. As on the grounds of cultural difference (Jayasuriya, part of that survey, which was conducted 2002). This is often expressed in terms of by telephone, data from 1845 respondents social cohesion—failure to assimilate—and in the Sydney metropolitan region were gen- national identity (Dixson, 1999; Phillips, erated and used in this study. The sample 1998; for survey results of such attitudes in was area-stratified so as to draw from the Australian context, see Brian Sweeney & within every second postcode, with the aim Associates, 1996a, pp. 2–23; McAllister and of including at least one postcode from Moore, 1989, pp. 7–11; Pedersen and every Local Government Area (LGA). Walker, 1997; Pedersen, Clarke et al., 2005). Valid response sets were derived for 43 of There is, nevertheless, a body of research the 45 LGAs in the Sydney region. Such a examining the co-existence of both old- focus on LGAs is important, given that fashioned and modern prejudice. Most of they are one of two community-based these studies find that both forms are strongly vehicles providing access to the benefits of correlated and that the correlations are rising government services for all the people of as the years go by (see McConahay, 1986). New South Wales and notably to NESB Recently, Pedersen et al. (2004) found that groups (the other is the network of Hospital old-fashioned and modern prejudice formed Board Community Service areas, but these a single meaningful factor. cover very much larger areas than the However, because of the variety of view- LGAs). They also act as a vehicle for points, touched on above, what constitutes locally based multicultural and citizenship racism for one person may be quite different initiatives (as exemplified in the state of for another and may vary not only from Queensland’s Local Area Multicultural person to person, but also among people of Programs). similar social backgrounds and from place to Much of the methodological means of scho- place (Dunn and McDonald, 2001). Thus larship on racialisation and ethnic relations in Bonnett (1996, p. 872) argued for a combined the past three decades has been qualitative, social and spatial perspective on racism and using ethnographic and often deeply self- suggested the value of social constructivism reflective techniques (for example, Twine as an analytical approach to understanding and Warren, 2000). In Australia, the field the processes involved. Constructivism, has been dominated by some excellent dis- according to Jackson and Penrose (1993, course analysis, especially of media commen- p. 3), works by identifying the components tary, and usually of a very qualitative form. and processes of category construction, and For example, Lamont et al. (2002, p. 395) notions of spatial identity or culture as well admitted that their extensive field interviews as what constitutes racism itself. Recent with working-class folk in France and North work in cultural geography, for example, has America were not as qualitative as most scho- witnessed a proliferation of studies of ‘race’ larship in the field. This same period in human within the larger discourse of social construc- geography has seen a reorientation mostly tion (Kobayashi, 2004, p. 239). towards qualitative approaches, including field interviewing and discourse analysis (Hay, 2005). There has been a clear qualitat- Survey and Data ive emphasis in racism research and, relatedly, The purpose of the University of New South somewhat of a paucity of more quantitative Wales/Macquarie University (UNSW/ approaches (the exception being the work of MQU) Racism Survey, conducted in late social psychologists). In a less apologetic 2001, was to collect data on racist attitudes mode, Modood (2000, p. 180) asserted that in the states of New South Wales and the Fourth National Survey of Ethnic Min- Queensland, which together approximate 50 orities in Britain was providing empirical per cent of Australia’s population. The material that had strong policy and public CONSTRUCTING RACISM IN SYDNEY 703 impact. We agree that survey data on racist Sydney Attitudes to Cultural Diversity and attitudes and experiences can have robust pol- Racism itical effects. As we indicate later, empirical Cultural Diversity and Nation material on racist attitudes has utility for framing anti-racism. Researchers have long pointed to links between Traditional questions used in racist attitudes racism and narrow constructions of national surveys are somewhat limited, although we identity or ideology of nation (Gilroy, 1987; have, for example, retained some Bogardus- Hage, 1998, pp. 2–55; Goodall et al., 1994, style indicators of social distance (Table 1). pp. 16, 188). Yet contradictions are evident in Some of the survey questions are adapted public opinion on national identity, cultural from existing research in this area. These diversity and multiculturalism (cultural diver- include aspects of the ‘old racism’—belief in sity is used throughout this study to denote a racial hierarchy, racial separatism and belief the presence of a wide range of birthplace in racial categories (racialism)—as well as the origins). Public opinion surveys in Australia extent of racism at both the general and individ- in the mid 1990s showed that 60 per cent of ual levels. Nonetheless, in order to deploy some respondents were against immigrant groups of the current theoretical work on racism, new maintaining their own cultural traditions. questions were developed. These operation- However, only 20 per cent agreed that multi- alised aspects of ‘new racism’: degree of accep- cultural policies should be abolished (Dunn tance of or opposition to cultural diversity and and McDonald, 2001, pp. 34–35). Yet cultural multicultural values, how narrow are construc- maintenance is a core principle of multicultur- tions of national identity (or cultural norms), alism (Commonwealth of Australia, 1999, protection of cultural privilege, judgements p. 19; Office of Multicultural Affairs, 1989, about the presence of ‘out-groups’, ideology p. vii). Similarly, the UNSW/MQU survey of nation and perceptions of Anglo-Celtic found that, while 85 per cent of respondents (‘host’ society or dominant culture) privilege were favourably disposed to cultural diversity, (Dunn et al., 2004). 45 per cent were of a view that cultural diver- While surveys of this magnitude have a sity and multiculturalism were a threat to Aus- certain quantitative punch, there are always tralian nationhood (Dunn et al., 2004). questions regarding the fidelity of responses. Contradictory views on multicultural values Do respondents truthfully answer questions and notions of nationhood reflect the presence regarding their attitudes? An even broader of two competing discourses. The first, a pro- question is whether these expressions of cultural diversity discourse, is based on opinion are reflected in behaviour (in racist liberal values of cultural equality, as presented acts). For example, more highly educated in the official rhetoric about multiculturalism respondents might more easily recognise a since the mid 1970s. The second relates to question as a test of their ‘racism’ and ‘new racism’ perspectives of culture and answer so as to conceal their own intolerances nation which act to mitigate the wider sense which they may intuitively perceive as reflect- of citizenship and belonging that multicultural- ing tainted or less progressive attitudes. ism should facilitate. As to the first, multicul- Equally, expressions of tolerance might bear turalism is often interpreted as an invitation little relation to continuing behaviour in to cultural pluralism. A recent survey of 3501 everyday life, which may be marked by state- Australians found that three-quarters of those ments of intolerance and discriminatory acts who had been in the country for a generation (Yinger, 1986). Whatever the potential cause and more (principally Anglo and Indigenous of such potential infidelity, it is likely that Australians) identified as ‘Australian’. Only survey results on racism are, if anything, 10 per cent of this group from non-English- underestimations of the phenomenon. A judi- speaking backgrounds so identified themselves cious way to approach such data is to remem- (Ang et al., 2002, p. 40). The survey also found ber that they are indictors of societal attitudes. that “mainstream definitions of Australian 704 JAMES FORREST AND KEVIN DUNN

Table 1. Defining the variables used in the entropy analysis Variable number Question wordinga Indicator 1 It is a good thing for a society to be Strongly disagree þdisagree: made up of people from different opposition to cultural diversity cultures 2 You feel secure when with people of Strongly disagree þdisagree: different ethnic backgrounds concern/opposition to cultural difference 3 There is racial prejudice in Australia Strongly agree þagree: recognition of racism in society 4 You are prejudiced against other Strongly agree þagree: self-identified cultures racism 5 It is not a good idea for people of Strongly agree þagree: belief in a different races to marry one another need to keep ‘races’ separate 6 Australians from a British Strongly agree þagree: cultural background enjoy a privileged privilege enjoyed by Anglo- position in our society Australians 7 Australia is weakened by people of Strongly agree þagree: concern/ different ethnic origins sticking to opposition to cultural difference their old ways and multicultural values 8 All races of people are equal Strongly disagree þdisagree: belief in a racial hierarchy 9 Humankind is made up of separate Strongly agree þagree: belief in races ‘natural’ racial groups 10 Do you believe that there are any Yes: suggests a right to make cultural or ethnic groups that do not judgements about in-groups and fit into Australian society out-groups 11 Age of respondents Aged 18–34: acculturated since end of White Australia policy (in early 1970s) 12 Age of respondents Aged 35–64: acculturated during post-WW2 period of European origin of migrants 13 Age of respondents Aged 65 þ: acculturated during pre- WW2 period of dominant British origin of migrants 14 Education of respondents Tertiary level qualifications 15 Education of respondents Higher School Certificate level (completed senior high school) 16 Education of respondents School Certificate (completed high school to school-leaving age) 17 Birthplace of respondents Born in Australia: Anglo homogeneity or cultural diversity. 18 Birthplace of respondents Born overseas, from a non-English- speaking background: cultural diversity aResponse options for Questions 1–9 used a 1–5 point Likert scale. cultural identity still tend to ignore or The White Australia policy has had a overlook the social diversity of the overall lasting impact on the national social devel- population”, hence adoption of assimilationist opment of Australia. It allowed the con- attitudes. Thus a Human Rights and Equal struction of a populist national identity Opportunity Commission investigation which excludes and marginalises groups found that … This has led to popular ideas of the CONSTRUCTING RACISM IN SYDNEY 705

need for people to conform to a set of per- to ‘assimilate’ (Brian Sweeney and Associates ceived cultural and social norms if they 1996a, 1996b; Pedersen et al., 2000). Anti- are to be truly ‘Australian’ (HREDC, Asian and anti-Muslim sentiment (the latter 2001, p. 19). often manifest as anti-asylum-seeker opinion) has been found in more recent These two contradictory discourses, of attitude polling in Australia (Klocker, 2004; Anglo-centrism (pro-assimilationist) and Pedersen, Clarke et al., 2005; see also pro-diversity (pro-multiculturalism), are McAllister and Moore, 1989). In the UNSW/ perceptible within the attitudes of Sydney MQU survey, 45 per cent of respondents ident- respondents on the topic of cultural diversity. ified a cultural group or groups that they felt did not fit into Australian society (for an Normalcy and Privilege expanded discussion of this issue, see Forrest and Dunn, 2006a, pp. 179–183) Critical social theorists have referred to the normalcy of racism: a context of White privi- lege associated with a way of life and thinking Belief in the ‘Old Racisms’ where racism is not consciously seen, or is con- Arguments that ‘racial groups’ should be separ- sidered an exceptional aberration (Bonnett, ated from each other (that intermarriage is not a 1996; Kobayashi and Peake, 2000, pp. 393– good idea) or that some ‘racial groups’ are not 397). Except, of course, that it is not a privilege equal to others (the notion of a racial hierarchy) of Whiteness (although see Hage, 1998) but of and the related notion that there are ‘natural’ Anglo or Anglo-Celtic privilege in the Austra- (and different) racial groups, are variously lian context (Johnson, 2002). Pedersen and referred to as the ‘old racisms’ or as ‘blatant’ Walker (1997, p. 565) have observed of con- or ‘old-fashioned’ racisms (Pettigrew and temporary Australian society that, alongside Meertens, 1995). Support for including such an “apparent egalitarianism”, there is a strong questions in the UNSW/MQU survey comes strain of new racism aimed at “defend[ing] from Jayasuriya’s (2002, p. 41) coupling of the privileges of the dominant culture”. Most issues of inferiority and inequality (the ‘old’ respondents to the UNSW/MQU survey (83 racisms) with differentiation (the ‘new’ per cent) recognised that there is racial preju- racisms) as the two basic logics of racism in dice in Australia and this compares with 79 the contemporary Australian context. per cent from an earlier survey who were con- Results from the UNSW/MQU survey cerned that racism was ‘rife’ (Brian Sweeney confirm that old racist sentiment remains and Associates, 1996a, p. 23; 1996b, pp. 11– part of contemporary thinking. More than 12). Recognition of Anglo privilege was less one-in-eight Australians believes in some apparent, but still a majority (57 per cent) form of racial supremacy. Some 13 per cent agreed that it existed. believe that these ‘races’ should be kept sexu- ally separate in terms of the undesirability of interracial marriage. Nevertheless, the pro- Out-groups portions involved are small and are mainly Previous discussion suggests that there is older people with lower education achieve- much about contemporary racism in Australia ment levels. On the other hand, the belief which is linked to historical constructions of that there are natural ‘racial’ categories, the country’s national identity, to questions defined as ‘racialism’ (Hannaford, 1997; of acceptance or otherwise of cultural diver- Miles, 1989) is widespread. sity and of who does or does not ‘belong’. Intolerance of Indigenous Australians, for Prejudice—Symbolic Racism example, is an enduring form of racism that is linked to stereotyping based on supposed Respondents were asked two questions about welfare dependency, drunkenness and failure prejudice and a related question about cultural 706 JAMES FORREST AND KEVIN DUNN hegemony. One focused on recognition of Outwardly at least, Sydney, Australia’s pre- racism in society generally (‘There is racial eminent EthniCity, displays evidence of the prejudice in Australia?’) and the other on success of multicultural policies in absorbing self-identification as a racist (‘You are preju- the diversity of post-World-War-II immigrant diced against other cultures?’). Some 12 per streams. Analysis of ancestry data from the cent of respondents agreed with the pro- 2001 census supports a transitory-nature position that they were prejudiced against view of ethnic migrant enclaves (Forrest other cultures according to the UNSW/ et al., 2003). This indicates a strong tendency MQU survey, compared with 83 per cent towards spatial mixing or assimilation which who recognised that there is a general places Sydney (and other Australian cities) problem with racism in Australia. This among the least segregated of cities of devel- suggests that otherwise endemic racism is oped nations in the English-speaking world seen mainly as a problem affecting other (Poulsen et al., 2001, 2004). In substantial people which, apart from those who self- part, this relates to the great diversity of identify as racists, tends to support Kobayashi national and ethnic origins among immigrants and Peake’s (2000, pp. 393–397) argument to Australia which has acted to prevent any that racism is most often seen as an aberration build up of a smaller number of particular associated with a relatively small minority. groups leading to widespread segregation The third aspect, the privileged cultural pos- such as found in many US cities. ition of Anglo-Australians, bears out notions of contemporary Australian society as, on the one hand, increasingly multicultural or at Spatial Analysis least culturally diverse, but on the other, still seen to be struggling to disengage from a To test for the culture of racism in Sydney and legacy of Anglo privilege and cultural its geography, we adopt an approach bringing dominance. together the social constructivist perspective set in a quantitative analytical approach. The latter is an entropy procedure based on infor- Sydney as EthniCity mation theory which groups urban sub-areas Sydney provides an ideal context in which to (Local Government Areas—LGAs—in this test ideas about the nature and construction case) based on commonality of profiles of racism in Australia. Since the early 1980s, across the range of attitudinal and socio- post-White-Australia settlement patterns demographic variables (Table 1). The vari- embodying an increasingly Middle Eastern ables comprise respondents’ answers to the and Asian immigrant stream have focused 10 attitudinal questions asked in the UNSW/ on Sydney. In the decade to 1996, Sydney’s MQU Racism Survey, aggregated to the share of Australia’s total population born in LGA level, along with aspects of ethnicity, Asia and the Middle East increased from 6 age and education from the 2001 Census for per cent to 13 per cent (Birrell and Rapson, each LGA. The advantage of this approach 2002, p. 11). Since then, Sydney’s share of is that variable loadings on each group (of new immigrants has continued to increase, LGAs) profile can be related to major with 39 per cent of total new arrivals locating aspects of old, new and symbolic racisms pre- there between 1996 and 2001 (pp. 11, 15). viously discussed. The answers to the attitudi- Sydney continues to dominate among arrivals nal questions were recorded on a 1-5 scale as of the post-White-Australia period, mainly indicated in Table 1. Among the 2001 independent (skilled) and business migrants Census data used for the socio-demographic from China, the Philippines and Hong Kong, (compositional) information for each LGA, along with smaller streams of refugee groups education was preferred to occupation or from Lebanon, Vietnam and now from east income, otherwise a large number of women Africa (Forrest et al., 2003, pp. 503–504). who work part-time or are in domestic duties CONSTRUCTING RACISM IN SYDNEY 707 would be marginalised in the assessment of Middle Eastern immigrants are among the socioeconomic status. most tolerant in terms of admitted personal A major attribute of the entropy procedure prejudice, rather more tolerant in fact than is that it is not constrained by issues of the Australian-born on this issue. normal distribution. Its ability to characterise Among the ‘new’ racisms, feelings of inse- and group observation areas with a curity with different cultural groups and a minimum of information loss is reviewed in desire to avoid social assimilation are Johnston and Semple (1983). In summary, it highest among Asians and Middle Eastern groups LGAs with similar responses across birthplace groups; again, north-east Asians the attitudinal and compositional variables. have the strongest degree of intolerance on Unlike other grouping procedures, the both of the ‘new’ racism questions. This is amount of within-group variance for (1 … n) potentially significant, in that those opposed groups at each iteration is minimised by to multiculturalism (those who agree that retesting all possible groupings of obser- Australia is weakened by immigrant groups vations. The number of groups selected is retaining their old ways)—47 per cent of the determined subjectively based on a decreasing Australian-born—are mainly concerned amount of variation accounted for by further about cultural pluralism or cultural segre- increasing the number of groups. In the gation. Finally, southern Europeans stand out Sydney case, 14 groups of LGAs accounted as being personally prejudiced against other for 73 per cent of variation across all 18 attitu- cultures; all other birthplace groups, including dinal and compositional variables (Table 2)— immigrants from the Middle East, are below the groups are ranked here on the socioeco- Australian-born levels of intolerance in this nomic status/education variables. Four of attitude. Findings for this birthplace–attitude these are single LGAs; another four comprise mix may well, therefore, lead to exacerbation just two or three LGAs. The other six groups of levels of intolerance in some culturally are made up of larger numbers of LGAs. diverse regions of the city, as well as being present in areas dominated by people of Anglo backgrounds. Towards a Geography of Racism in Sydney The approach used here is, first, to examine Patterns of intolerance in Sydney are constructions of attitudes and their compo- complex, in terms of both attitude mix and sitional correlates across the 14 entropy associated socio-demographic profiles among groups. Then, secondly, to aggregate the the 14 groups of LGAs brought out by the entropy results presented in Tables 2 and 3 entropy analysis (see Tables 2 and 3). Nor is into a comprehensive construction of the inci- intolerance the preserve of the Australian- dence of racist attitudes in Sydney. In the dis- born alone, as highlighted by the attitudes of cussion which follows, the Local Government different birthplace groups to aspects of intol- Areas referred to are those identified in erance (Table 4). Compared with the Austra- Figure 1. lian-born, the most intolerant groups are generally Asians, especially those from north-east Asia (principally China and Hong Cultural Diversity and Nation Kong); southern Europeans have quite high Contradictory views about national identity, levels of intolerance, surprisingly, perhaps, cultural diversity and multiculturalism, noted exceeding that of Middle Eastern immigrants. earlier, are a lasting legacy of the former More specifically, southern Europeans White Australia policy. The relevant ques- (mainly from Italy, Greece and the former tions are Yugoslavia) have the highest level of intoler- ance towards racial intermarriage, an aspect of —opposition to cultural diversity (Qu. 1); the ‘old’ racism, with those from north-east —concern or opposition to cultural difference Asia not much more tolerant. Interestingly, (Qu. 2); and 708

Table 2. Entropy analysis of racist attitudes and social contextual attributes among Sydney LGAs

Recognise Identify Cultural diversity and Self as racism and Old racisms: separation, out-groups nation racist privilege hierarchy and racialism Ethnicity Age Education

Question number (10) (1) (2) (7) (4) (3) (6) (5) (8) (9) Australian-born NESB 18–34 35–64 65 þ Year 10 Year 12 Tertiary AE ORS N EI DUNN KEVIN AND FORREST JAMES

Average 45.96 8.75 10.97 44.34 14.61 11.07 40.84 12.55 11.39 76.98 71.96 18.39 37.37 47.26 15.37 31.27 46.03 22.70 Standard deviation 11.81 6.23 8.59 12.29 9.33 7.87 12.98 6.14 6.14 7.69 10.29 11.99 4.70 3.02 3.76 13.32 6.47 7.22

Group 1 223.74 28.74 0.14 233.23 7.61 0.04 36.94 21.44 20.28 210.31 221.50 16.04 12.90 26.50 26.40 220.17 9.98 10.19 2 2.54 28.74 22.87 6.58 24.66 27.37 1.39 20.75 6.67 24.99 2.25 29.27 22.41 0.43 1.98 220.15 8.75 11.40 3 214.92 23.24 29.97 213.76 27.84 22.16 8.02 23.53 24.50 21.86 21.50 23.04 3.21 22.13 21.09 214.91 6.57 8.34 4 24.04 1.25 210.96 5.66 214.60 8.93 19.14 212.54 8.61 3.02 7.26 27.28 26.09 20.19 6.28 210.86 3.43 7.43 5 20.50 0.34 210.96 226.16 25.52 211.06 222.66 23.46 6.80 24.25 2.68 211.72 21.37 20.95 2.33 212.33 5.32 7.01 6 28.46 28.74 1.53 219.34 214.60 13.93 215.86 212.54 211.38 21.98 211.53 12.84 20.92 21.61 2.53 26.40 2.20 4.20 7 2.14 3.83 0.15 0.65 0.71 23.99 3.58 21.20 0.10 2.87 4.84 25.88 23.67 3.12 0.55 23.28 1.16 2.02 8 23.77 26.01 26.02 2.56 26.62 16.92 9.54 2.00 25.90 2.64 28.92 10.57 1.32 22.32 0.99 22.19 2.49 20.30 9 12.22 13.76 2.09 4.47 12.83 22.78 28.60 1.75 4.06 10.04 28.67 12.52 20.26 20.88 1.15 20.48 1.24 20.76 10 1.60 4.53 15.49 6.42 5.43 8.82 20.89 7.07 3.99 29.51 28.93 13.48 0.82 0.31 21.13 4.04 20.06 23.97 11 1.50 22.02 0.63 0.44 4.14 22.58 21.12 5.21 22.03 24.03 27.58 11.62 21.10 21.24 2.34 5.96 21.40 24.56 12 10.52 21.79 0.68 13.22 22.06 22.97 0.86 22.15 2.58 7.96 15.98 215.66 25.55 1.90 3.64 14.72 29.00 25.72 13 11.84 0.50 13.87 13.66 21.98 24.53 223.71 5.74 0.57 7.01 15.82 214.27 0.10 4.39 24.48 21.04 212.17 28.87 14 21.91 2.95 1.06 5.95 22.26 21.49 211.01 21.59 0.52 20.90 8.94 27.95 4.30 1.46 25.77 17.99 28.58 29.41

Note: All values in the table are percentages and expressed above or below (2) Sydney means. Question numbers are shown in parenthesis. CONSTRUCTING RACISM IN SYDNEY 709

Table 3. LGAs in each entropy group Group 1 Group 6 Group 10 Group 14 Sydney Ashfield Concord Blacktown Fairfield Camden Group 2 Group 7 Liverpool Campbelltown Mosman Baulkham Hills Strathfield Penrith Woolahra Drummoyne Hornsby Group 11 Group 3 Pittwater Bankstown Ku-ring-gai Warringah Canterbury Leichhardt Hurstville Marrickville Group 8 Parramatta North Sydney Botany Rockdale South Sydney Burwood Waverley Randwick Group 12 Willoughby Blue Mountains Group 9 Gosford Group 4 Auburn Sutherland Hunters Hill Kogarah Wyong Ryde Group 5 Group 13 Manly Hawkesbury Wollongdilly

—concern or opposition to multicultural reminded scholars of Allport’s (1954) four values (Qu. 7). conditions necessary for contact to have a beneficial, rather than negative impact, on Inner western and south-western districts of community relations. These comprise equality Sydney (groups 9 and 10), strongly support among groups, a sense that wider goals are all three of these viewpoints. LGAs in these being pursued, an absence of intergroup com- groups have high levels of cultural diversity. petition (such as competition for jobs) and They also span a wide range of socio-demo- some sense of official sanction promoting graphic characteristics, but with an above- contact and the wider endeavour. average proportion of people with lower edu- Sydney’s higher SES areas (completed high cational achievements—low to low-middle school and tertiary education) largely reject the socioeconomic status (SES)—as well as viewpoints enshrined in each of the three ques- above-average proportions of immigrants tions. Areas involved include northern parts of from non-English-speaking backgrounds. the city (group 3 in particular) and the gentrify- This is a region where working-class Anglo- ing inner city (group 1). There is very strong Celtic-Australians and non-English-speaking support for multicultural values (Qu. 7) here, background immigrant community groups although some disagreement on the other two are intermixed, but where cross-cultural questions; the northern beachside suburb of contact and developing familiarity have not Manly (group 5) is more ambivalent about yet generated acceptance of cultural diversity. support for cultural diversity. Yet in terms of Chicago School suggestions of ‘contact’-gen- cultural diversity, the two main entropy erated tolerance are clearly tempered by local groups here (groups 1 and 3) are very different. and social circumstances, and the relatively Levels of cultural diversity in the inner city are recent—only over the past two to three markedly above-average, but in the northern decades—presence of large numbers of cultu- city areas they are below average; younger rally diverse groups. Both Paradies (2005) and working-age (18–34) populations are a Pedersen, Walker et al. (2005) have recently common factor here. 710

Table 4. Aspects of tolerance and intolerance in Sydney Birthplace groups New Zealand/UK/North Australia America South and south-east Asia North-east Asia Southern Europe Middle East Total DUNN KEVIN AND FORREST JAMES Q2 Feeling secure with different cultural groups Not secure 10.3 11.4 16.4 19.6 12.5 17.2 11.0 Ambivalent 13.1 13.9 16.4 13.0 12.5 6.9 13.2 Q7 Australia is weakened by groups retaining old ways Disagree 35.9 37.3 39.3 50.0 41.7 43.1 37.4 Agree 47.3 18.1 16.4 34.8 43.8 39.7 45.6 Q5 Not good for people of different races to intermarry Agree 11.9 10.9 13.9 17.4 20.8 12.1 12.7 Disagree 76.7 80.3 76.2 71.7 79.2 77.6 76.8 Q4 Personal prejudice against other cultures Agree 14.8 12.4 12.3 15.2 18.7 10.3 14.3 Disagree 75.9 80.3 77.0 69.6 75.0 79.3 76.8

Note: All values are percentages of each birthplace group. Source: UNSW/MQU Racism Survey (2001). CONSTRUCTING RACISM IN SYDNEY 711

Figure 1. Local Government Areas in Sydney. Scale: 1 cm ¼ 6.5 km.

In between these two attitudinal positions is populations. These outer-suburban areas are a group of areas where respondents have where local conflicts over non-Christian mixed views, some agreeing, some disagree- places of worship have been prominent in ing on all three questions, and with noticeable recent times (Dunn, 2004). Respondents variations to the strength of views held. Two from Sydney’s north-western and northern outer, rural–urban fringe districts (group 13) beach suburbs (group 7), while opposed to are strongly opposed to cultural difference cultural diversity, are only mildly concerned and to multicultural values, but only mildly about cultural difference and multicultural against cultural diversity. A group of outer values. Like the other two groups, these western and south-western suburbs (group areas are mainly Australian-born, but of 14) is opposed to both multicultural values middle to higher SES; while this may and to cultural diversity, but has few concerns help to account for greater ambivalence on about cultural difference. Both groups of areas two of the questions, respondents from the are dominated by the Australian-born and by areas in this group are clearly closer to those lower SES (mainly year 10 schooling) from the working-class suburbs in their 712 JAMES FORREST AND KEVIN DUNN concern over—disagreement with—cultural ‘others are not racist and neither am I’ diversity. persuasion. Most are above-average in the presence of Australian-born residents and include a wide range of age-groups. But Normalcy of Racism and Anglo Privilege: while rejecting racist attitudes, they agree Symbolic Racisms with the existence of privilege. This latter The survey results for Sydney confirm response suggests a strong recognition of a Kobayashi and Peake’s (2000) view that form of the ‘new racism’ whereby ethnic min- racism is seen by most people as an aberration orities are culturally disadvantaged by the of a relatively small minority. The questions dominant cultural group’s understanding of are national culture and identity (see Forrest and Dunn, 2006a). —recognition of racism in society (Qu. 3); The notion that ‘others are racist but I am —self-identification as racist (Qu. 4); and not’, accompanied by a strong sense of privi- —that cultural privilege is enjoyed by Anglo- lege among Anglo-Australians, is prevalent Australians (Qu. 6). in a small number of LGAs mainly in Socioeconomic status is the major differen- Sydney’s eastern and inner western districts tiating element in area responses to these (groups 6, 8). Most here have above-average questions; the impact of age and cultural levels of cultural diversity, are mainly diversity is less apparent. middle-aged to older and of generally One of the major issues here is the distinc- middle to higher SES. The exception in com- tion between a potentially endemic form of positional terms is the inner city, which is gen- racism in Australia, based on recognition of erally younger to middle-aged, associated racial prejudice in society generally (Qu. with extensive gentrification. It is as though 3)—others are racially prejudicial—and self- there is a culture of the inner-city region identification as prejudiced against other cul- which transcends compositional character- tures (Qu. 4)—I am racially prejudiced—on istics—such a feature is also found in the part of a relatively small minority. aspects of political behaviour (Forrest et al., Results can be discussed in terms of a four- 1984). way classification from ‘others are not racist Strong levels of self-identification as and neither am I’ through ‘others are racist racists—LGAs where ‘others are not racist but I am not’, ‘others are not racist but I am’ but I am’—are found in two outer, rural– to ‘others are racist and so am I’. Interestingly, urban fringe areas (group 13), privilege, associated as it is with an assimila- strongly working-class, middle-aged and tionist or ethnocultural perspective on the dominantly Australian-born with lower treatment of non-Anglo ethnic immigrant levels of cultural diversity. There is a low groups, is less obviously linked to overtly awareness of racist attitudes in society gener- racist attitudes. And whereas with the two ally or of Anglo privilege, which suggests that questions on endemic and personal racism they are comfortable in their particular attitu- there is a general (although not always con- dinal niche. sistent) negative relationship with Highest levels of racism by Sydney stan- socioeconomic status, this does not apply to dards and recognition of Anglo privilege recognition of Anglo privilege, where both occur in several inner western and south- lower and higher SES respondents recognise western LGAs of lower SES but with high that Australians of a British background levels of cultural diversity (group 10); an enjoy a privileged social position in this above-average number of respondents say country. that ‘others are racist and so am I’. Signi- High SES LGAs on Sydney’s northern dis- ficantly, perhaps, these are areas with the tricts, inner Sydney and the largely gentrified highest numbers of recently arrived immi- inner city (groups 1, 3, 7) are strongly of the grants from Asian and Middle Eastern CONSTRUCTING RACISM IN SYDNEY 713 countries where contact has not yet led to and 7). No common socio-demographic intergroup social acceptance and where, on characteristics stand out and there are no par- the evidence presented in Table 4, ethnic ticular spatial patterns. groups are as intolerant as the Australian- A much smaller number who believe in the born. separation of the ‘races’ (no intermarriage) stand out in a few LGAs (groups 10, 13 and 11). All are lower SES areas in western Belief in the ‘Old Racisms’ Sydney. This is a little surprising in terms of the- Variation in belief in the perceived inferiority ories about frequency of contact associated with of some groups and associated notions of social acceptance because, in many of these dis- ‘racial’ inequality (‘old racisms’) are associ- tricts—although not all—cultural diversity is ated most strongly with lower levels of edu- strong. However, it is consistent with findings cation, but not generally with age or levels noted above for these areas in terms of symbolic of cultural diversity. The relevant questions racism and attitudes brought out among the were Australian-born and other immigrant groups in Table 4. Rather more believe in the notion —belief in a need to keep races separate of a hierarchy of ‘races’, again mainly in (intermarriage) (Qu. 5); lower SES LGAs in outer, inner western and —belief in a hierarchy of races (Qu. 8); and south-western districts (groups 9, 12 and 10), —belief in ‘natural’ racial groups (Qu. 9). but also in some higher SES areas in the The ‘old racisms’ are largely a feature of areas mainly inner city, culturally diverse region of with higher proportions of lower SES groups. LGAs. Some areas of higher relative diversity, However, ‘old racisms’ are also a feature of such as in groups 8 and 11, have stronger some more affluent, inner northern LGAs, levels of disagreement with the notion of with older age profiles, such as Manly, ‘racial’ hierarchy, yet there is an above- Mosman and Hunters Hill. One explanation average agreement with the need for ‘racial’ of the inclusion of this affluent group of separation. It is of some concern that residents LGAs may be their older age structure, of some of these culturally diverse areas, cogni- which contrasts with the younger age structure sant of the fallacy of ‘racial hierarchy’, nonethe- of the outer, lower SES areas. Conversely, less express some preference for ‘racial’ areas with negative—agree to strongly agree separation. Other culturally diverse areas with—responses to all three ‘old racism’ show higher levels of opposition to both forms questions (group 3) are also higher SES of ‘old racism’, such as those in groups 3 and 6. LGAs, in northern and north-western Sydney. The most tolerant—disagree to Cultural Diversity and Out-groups strongly disagree with—groups (groups 6 and 1) are in parts of inner western Sydney Attitudes to cultural diversity and to the exist- and inner Sydney respectively. The latter ence of out-groups in Sydney as a whole can can be described as places of long-established be summarised as moderately tolerant. This cultural and social (in terms of both age and discussion focuses on two questions class) diversity. Long-standing cross-cultural —opposition to cultural diversity (Qu. 1); and contact, in cosmopolitan inner Sydney —judgements about in-groups and out-groups especially, has resulted in stronger levels of (Qu. 10). cultural acceptance. A widespread acceptance of ‘racialism’, of In general, higher levels of education and natural racial categories, is reflected in the inner-city location combine to typify areas high average incidence and relatively low which are least antagonistic towards other cul- standard deviation on this question (Qu. 9). tural groups. Thus acceptance of cultural This attitude was prevalent across 6 of the diversity and absence of any recognition of 14 groups of LGAs (groups 9, 12, 8, 4, 13 out-groups is a feature of four groups of 714 JAMES FORREST AND KEVIN DUNN

LGAs: in the inner city (group 1); in the gen- Anglo privilege, arguably among the ‘new erally middle-status inner west and eastern racisms’, but lesser identification of out- suburbs (groups 6 and 8); and in Sydney’s groups. In this region, these attitudes are higher-status northern suburbs (group 3). common to all age-groups, in spite of Three of these groups have above-average marked variation in levels of cultural diver- levels of cultural diversity, but the fourth— sity. Respondents from upper-middle status, the northern suburbs—lacks cultural diversity largely middle-aged LGAs on Sydney’s but has high educational levels. Again, we upper north shore (Region II), on the other point to a contextual effect of inner-city hand, are generally opposed to cultural diver- cosmopolitanism. At the other end of the sity, although this varies across the region, and scale, opposition to cultural diversity and people there are more prepared to identify out- identification of out-groups are strong charac- groups. However, like the higher-status LGAs teristics of inner western and outer south- in Region I, rejection of racism in society gen- western LGAs (groups 9 and 10), all with erally is accompanied by recognition of Anglo high levels of cultural diversity and of privilege. There is, however, greater acknowl- middle to lower social status. More weakly edgement of personal racism in this region, held opposition is a feature of the northern although people are, on the whole, neutral beaches and north-western Sydney LGAs on ‘old racist’ attitudes. Both of these (groups 4 and 7); these have low levels of cul- regions have relatively low levels of cultural tural diversity (a dominance of Australian- diversity. born) and are mainly of higher social status. The inner city (Region III) has previously In between, only a few groups of LGAs been identified as having an interesting mix stand out, while some others are close to the of cultures, classes and age-groups alongside city average on both questions. particularly tolerant and pro-diversity atti- tudes. It stands out on its own in this regard. Other parts of inner to middle suburbia, on Constructing a Geography of Racism the other hand, exhibit a wide range of atti- in Sydney tudes, some very tolerant, some not. Thus From the previous discussion, there is evi- respondents in Region IV, a combination of dence of a distinctive geography of racial eastern and inner-middle LGAs, are generally intolerance in Sydney that transcends aspatial accepting of cultural diversity, recognise that (compositional) relationships. Such com- others are racist and that Anglo privilege positional relationships do occur, but are exists, yet support ‘old racist’ attitudes, but strongest, in the case of socioeconomic do not identify any out-groups. Closer in to status, only at the higher and lower ends of the inner city, Region V shares many of the the range and then with variations, especially characteristics of Region IV, but respondents among the ‘new racism’ dimensions. Rather, are more likely to deny that racism exists in in various groups of LGAs, there is evidence any form and are generally tolerant. That of a particular combination of attitudes such this is an area long identified with Italian that a geography of racist attitudes can be con- and Greek immigrants who began settling structed using all the attitude variables from there in the 1950s and 1960s may suggest the entropy analysis (Figure 2) acceptance of diversity based on a long Respondents from higher SES, generally period of mutual contact, as opposed to the older LGAs on Sydney’s north side and more recent immigrant streams in major eastern suburbs (Region I) are more accepting parts of Region IV, which includes many of cultural diversity at the national level and non-Christian immigrants. of multiculturalism; there is lesser acknowl- To the south and west of these regions, edgement of personal racism and greater respondents in Region VI hold neutral views denial that racism is general in Australian on issues of cultural diversity at the national society. There is, however, recognition of level and on multiculturalism; they are mildly CONSTRUCTING RACISM IN SYDNEY 715

diversity or multiculturalism to mildly opposed; they are mildly aware of racism generally and of Anglo privilege; they reject racism in themselves and in others while sharing some aspects of ‘old racisms’. Region IX, covering south-western Sydney, on the other hand, is the least tolerant part of Sydney. Here, there is opposition to cultural diversity and to multiculturalism; personal and general racism is admitted and recog- nised; cultural privilege is seen to be strongly present; however, respondents are less likely to identify out-groups. There is a younger to middle-aged complexion to this region and it is home to a very wide range of immigrant groups and nationalities, European, Middle Eastern and Asian, including a number of refugee groups. It is also an area of high unemployment and strong competition for employment opportunities. Finally, Sydney’s outer commuter and rural–urban fringe districts combine into Figure 2. Geography of racist attitudes in Sydney. Region X. Younger to middle-aged, with Scale: 1 cm ¼ 3.5 km. pockets of older people, this region is more culturally homogeneous, dominated by concerned about racism generally and Anglo Australian-born residents. Attitudes on issues privilege. There is, however, an above- of cultural diversity, and on multiculturalism, average presence (albeit the overall average is range from mildly to strongly opposed. low) of those who are personally racist, while However, there is little recognition of there is a strong rejection of values associated endemic racism, or of Anglo privilege, but with the ‘old racisms’. Respondents in there is strong support for ‘old racism’ atti- Region VII, still part of the inner western tudes. This region is like Region IX, therefore, LGAs, oppose cultural diversity and multicul- in being among the least tolerant areas in turalism, but exhibit little evidence of racism Sydney, although the dimensions that are in either its general or personal forms. relevant in each of these regions vary. However, there was a strong attachment to There are parallels in the above patterns of the ‘old racisms’ and agreement that there are tolerance and intolerance with local political cultural groups that do not belong in Australia. cultures, exemplified by recently formed Parts of Region VII have Muslim immigrant minor political parties (Smith, 2005). Thus populations and this is the location of one of the Unity Party was formed in 1998 to Sydney’s major mosques; other parts have a mobilise support for multiculturalism strong Asian (Chinese and Korean) pre- (Healy, 1999; Money, 1999; State Electoral sence—both groups are among those com- Office, 2003c). Its main support base has monly identified as out-groups (Dunn et al., been in inner-city to middle-ring suburban 2004; Pedersen et al., 2005). districts. The party has also attracted signifi- Sydney’s western and south-western LGAs cant support in the south-western suburbs divide into two main regions. In Region VIII, (Region IX), the least tolerant of the Sydney attitudes to cultural diversity and to multi- regions. Among two right-wing minor culturalism vary, but around a middle pos- parties—Australians Against Further Immi- ition, from generally unconcerned about gration (Newman, 1995), formed in 1989, 716 JAMES FORREST AND KEVIN DUNN and the One Nation Party (Johnson, 1998), Conclusions formed in 1997—on the other hand, Austra- This construction of racism in Australia’s lians Against Further Immigration presents largest EthniCity indicates a mix of compo- an anti-immigration, anti-multiculturalism sitional (aspatial) and spatial (contextual) platform, arguing that “the people have associations, but with the latter standing out never been consulted on, or given their sufficiently to emphasise the importance of consent to, the interwoven policies of immi- an ‘everywhere different’ conclusion. The gration and multiculturalism” (State Electoral social construction approach we have been Office, 2003a). Their support is strongest in guided by eschews any fixation upon a outer suburban (lower to middle class) and single social variable (such as class, or some middle suburban regions, such as cross-cultural contact/diversity) and has Region X and western Sydney generally. allowed us to weave through the analysis, Relatedly, One Nation argues that “the cul- our interest in contextual variables. Evidence tural, ethnic and racist makeup of Australia from discussion of the major dimensions of must not be radically altered through immi- prejudice suggests a relationship between gration without the express consent of the socioeconomic status and tolerance, such Australian people” (State Electoral Office, that class-based cultures underpin racist atti- 2003b). Its support base focuses on outer tudes—although this is most clear only at suburban areas, especially in western and the upper-middle SES (non-racist) and lower south-western Sydney (Regions VI, VIII, IX SES (racist) ends of the scale, and even here and X). not consistently so. In some cases, attitudinal Such a geography of political cultures variation can be linked with a stronger or across Sydney, like the geography of attitudes, lesser level of LGA cultural diversity, but in reflects historical demographic and urban other cases it cannot. trends. For example, the inner city is the Age structure emerges as a relatively unim- home of Hage’s (1998) cosmo-multicultural- portant variable. There are glimpses but no ists (younger, sometimes affluent, often body of evidence in the geography of racial although not necessarily tertiary educated, prejudice in Sydney to suggest any uniformity and culturally diverse) and an emergent politi- of generational differences in attitudes, not cal culture that Young (1990) might have even in the prevalence of ‘old’ versus ‘new called a politics of difference. Some recent racisms’. This interpretation, based on the scholarship has suggested that the inner city geography of attitudes, also contrasts with holds significant emancipatory potential non-geographical analyses, where older (Lees, 2004). Others have gone so far as to people, taken as a whole, have been found to suggest that inner cities are places where be more inclined to ‘old racist’ attitudes and poorer communities retain a grasp on their identification of out-groups (Dunn et al., neighbourhoods and where intolerance and 2004). A wider reconstruction of all the attitu- ‘race’ are not as significant as previously dinal dispositions discussed here indicates that asserted (see Maginn, 2004). Within Sydney, a geography of racism adds a level of under- the areas adjacent to and within the inner- standing of racism in Sydney and is a basis city gentrification belt are areas of greater tol- for future work; findings presented here erance. It may also be that inner-city living in could not have been obtained from aspatial Australia is driven in part by a cultural prefer- analysis alone. From the evidence of vari- ence for cosmopolitanism (including the ations in attitudes across the city, even experience of cultural diversity), such that among people of otherwise similar social anti-racists self-select into these areas of toler- backgrounds, it is apparent that spatial ance. Research into the geography of attitudes context is important: place matters. can add substantial value to the emergent Consistent with results from earlier scholarship on the cultures of urbanism research, the range and mix of attitudes (Zukin, 1986). CONSTRUCTING RACISM IN SYDNEY 717 revealed here, and the contradictions, within that are already culturally diverse and where and among the regions, indicate a diversity people are currently making decisions about of racist attitudes in Sydney. Variations racism and about community relations, other among LGAs with the same level of cultural cultural groups they have contact with and on diversity bear out these contradictions, as do national identity (Regions VI and VII). Anti- variations among the dimensions discussed racism initiatives in those areas should be during the earlier deconstruction—the five different from those where diversity is a dimensions—phase of this study. These are much more recent phenomenon and where consistent with contradictions between social the ‘issues’ are becoming more palpable and contact and social conflict theories of accep- perhaps problematic (Regions VIII and IX). tance and friction respectively (see Valenty In other regions, cultural diversity is low and and Sylvia, 2004), but with the added sugges- the issues are still being considered, or an a tion that local attitudes may relate to the pre- priori stance of intolerance has become appar- sence of certain immigrant groups rather ent (Regions X and II respectively). In still than others—those that are still regarded as other areas, there is both a strong diversity ‘Other’ (see Dunn et al., 2004, p. 414; and a robust basis for harmonious community Forrest and Dunn, 2006a, pp. 179–183). relations. The geographically informed analy- That a fuller understanding of racist attitudes sis outlined in this study provides anti-racism is also linked to understanding cultural or campaigners with the means to prepare region- dominant group (Anglo) privilege is another ally sensitive programmes that would address area which remains to be analysed and one ‘unsettled multiculturalisms’ across a major which leads ultimately into issues of national immigrant-receiving metropolis like Sydney. identity, citizenship and related issues of the Part of this situation of ‘unsettled multicul- supposed social compatibility of immigrant turalisms’ is the recent emergence of minority groups, of those who may be seen to ‘fit in’ political parties of the left and right respect- and those who may not. One thing in particu- ively and avowedly pro- and anti-multicultur- lar which does come out of this study is the alism, where previously the structural importance of incorporating both compo- assimilation of NESB immigrant groups sitional (social and demographic) and contex- occurred through membership of one or tual (place-specific) characteristics into future other of the major political parties, the left- analysis of attitudes to racism. leaning Labor or right-leaning Liberal In terms of future applied work, findings parties. While far from representing any fun- from this study are potentially important in damental change to the nature of politics in policy terms, specifically regarding community New South Wales brought about by the new relations (anti-racism) initiatives. There is a politics of immigration and multiculturalism, range of means by which anti-racism can be as claimed for English politics by the British advanced (see categories and reviews in Para- National Party after recent (May 2006) dies, 2005; Pedersen, Walker et al., 2005). Council elections there, the sentiments Many such initiatives are delivered at the com- echoed by each of the Australian minor munity (LGA) level, rather than through one- parties do suggest a geographical base to on-one workshops with individuals. Group- increasing polarisation of political responses level initiatives need to be sensitive to the con- which relates to the geography of tolerance texts of racism within a specific community and intolerance found in this study that also (Dunn and McDonald, 2001, pp. 38–41; deserves further attention. Paradies, 2005, pp. 14, 23; Pedersen, Walker et al., 2005, pp. 26–28). Thus our discussion References of the geography of racism in Sydney brought ALLPORT, G. (1954) The Nature of Prejudice. out in Figure 2, for example, identifies the Reading, MA: Addison-Wesley. need for at least four types of regionalised ANG, I., BRAND, J. E.., NOBLE, G. and WILDING, response to racism. There are those localities D. (2002) Living Diversity: Australia’s 718 JAMES FORREST AND KEVIN DUNN

Multicultural Future. Artarmon: Special Broad- FORREST, J. and DUNN, K. (2006a) Racism and casting Services Corporation. intolerance in Eastern Australia: a geographic BIRRELL, B. and RAPSON, V. (2002) Two perspective, Australian Geographer, 37, pp. Australias: migrant settlement at the end of 167–186. the 20th century, People and Place, 10, pp. FORREST,J.andDUNN, K. (2006b) ‘Core’ culture 11–25. hegemony and multiculturalism: perceptions of BONNETT, A. (1996) Constructions of ‘race’, place the privileged position of Australians and discipline: geographies of ‘racial’ identity with British backgrounds, Ethnicities,6,pp. and racism, Ethnic and Racial Studies, 19, pp. 237–264. 864–883. FORREST, J., HAY, A. M. and JOHNSTON,R.J. BOURQUE, G. and DUCHASTEL, J. (1999) Erosion (1984) In what ways are the suburbs different?, of the nation-state and the transformation of Politics, 19, pp. 97–101. national identities in Canada, in: J. L. ABU- FORREST,J.,POULSEN,M.andJOHNSTON,R. LUGHOD (Ed.) Sociology for the Tewnty-first (2003) Everywhere different? Globalisation and Century: Continuities and Cutting Edges, the impact of international migration on Sydney pp. 183–198 Chicago, IL: University of and Melbourne, Geoforum, 34, pp. 499–510. Chicago Press. FORREST,J.,POULSEN,M.andJOHNSTON, R. (2006) BRIAN SWEENEY AND ASSOCIATES (1996a) A Peoples and spaces in a multicultural nation: cul- report on a national quantitative survey of tural group segregation in metropolitan Australia, community attitudes towards Aboriginal recon- Espace, Populations et Socie´te´s, pp. 151–164. ciliation. Brian Sweeney and Associates, South GILROY, P. (1987) There Ain’t No Black in the Melbourne. Union Jack: The Cultural Politics of Race and BRIAN SWEENEY AND ASSOCIATES (1996b) Report Nation. London: Hutchinson. on supplementary qualitative research on GOODALL, H., JAKUBOWICZ, A., MARTIN,J.ET AL. community attitudes to indigenous citizenship (1994) Racism, Ethnicity and the Media. and a document of reconciliation. Brian St Leonards, Sydney: Allen & Unwin. Sweeney and Associates, South Melbourne. HAGE, G. (1998) White Nation: Fantasies of COLLINS, J. (1984) Marx’s reserve army of White Supremacy in a Multicultural Society. labour: still relevant 100 years on, Journal of Annandale, Sydney: Pluto Press. Australian Political Economy, 16(March), pp. HAMILTON, A. (1990) Fear and desire: Aborigines, 51–65. Asians and the national imaginary, Australian COMMONWEALTH OF AUSTRALIA (1999) Living in Cultural History, 9, pp. 14–35. harmony. Department of Immigration and HANNAFORD, I. (1997) Race: The History of an Multicultural Affairs, Canberra. Idea in the West. Baltimore, MD: Johns DIXSON, M. (1999) The Imaginary Australian: Hopkins University Press. Anglo-Celts and Identity—1788 to the Present. HAY, I. (Ed.) (2005) Qualitative Research Methods Sydney: University of New South Wales Press. in Geography, 2nd edn. Sydney: Oxford DUNN, K. (2004) Islam in Sydney: contesting the University Press. discourse of absence, Australian Geographer, HEALY, E. (1999) The Unity Party and the attempt 35, pp. 333–353. to mobilize Australian electoral support DUNN, K. M. (2005) Repetitive and troubling for multiculturalism, People and Place, 7, pp. discourses of nationalism in the local politics 51–62. of mosque development in Sydney, Australia, HELMES-HAY, R. and CURTIS, J. (Eds) (1998) The Environment and Planning D, 23, pp. 29–50. Vertical Mosaic Revisited. Toronto: University DUNN, K. M. and MCDONALD, A. (2001) The of Toronto Press. geography of racisms in NSW: a theoretical HESSE, B. (Ed.) (2000) Un/settled Multicultural- exploration and some preliminary findings, isms. London: Zed Books. Australian Geographer, 32, pp. 29–44. HIEBERT, D., COLLINS, J. and SPOONLEY, P. (2003) DUNN, K., FORREST, J., BURNLEY, I. and Uneven globalization: neoliberal regimes, immi- MCDONALD, A. (2004) Constructing racism in gration and multiculturalism in Australia, Australia, Australian Journal of Social Issues, Canada and New Zealand. Working Paper No. 39, pp. 409–430. 03-05, Centre of Excellence for Research on EADE, J. (Ed.) (1997) Living in the Global City. Immigration and Integration in the Metropolis, London: Routledge. Vancouver. DURRHEIM, K. and DIXON, J. (2001) The role of HREOC (HUMAN RIGHTS AND EQUAL OPPORTU- place and metaphor in racial exclusion: South NITIES COMMISSION) (2001) I want respect and Africa’s beaches as sites of shifting racializa- equality: a summary of consultations with civil tion, Ethnic and Racial Studies, 24, pp. 433– society on racism in Australia. HREOC, 450. Canberra. CONSTRUCTING RACISM IN SYDNEY 719

JACKSON, P. and PENROSE, J. (Eds) (1993) to Cultural Geography, pp. 238–249 Oxford: Constructions of Race, Place and Nation. Blackwell Publishing. London: University College London Press. KOBAYASHI, A. and PEAKE, L. (1994) Unnatural JAYASURIYA, L. (2002) Understanding Australian discourse: ‘race’ and gender in racism, Australian Universities Review, 45, pp. geography, Gender, Place and Culture, 1(2), 40–44. pp. 225–243. JOHNSON, C. (1998) Pauline Hanson and One KOBAYASHI, A. and PEAKE, L. (2000) Racism out Nation, in: H. G. BETZ and S. IMMERFALL of place: thoughts on whiteness and an antiracist (Eds) The New Politics of the Right: Neo- geography in the new millennium, Annals of the populist Parties and Movements in Established Association of American Geographers, 90, pp. Democracies, pp. 211–218. New York: St 392–403. Martins Press. LAMONT, M., MORNING, A. and MOONEY,M. JOHNSON, C. (2002) The dilemmas of ethnic (2002) Particular universalisms: North African privilege: a comparison of constructions of immigrants respond to French racism, Ethnic ‘British’, ‘English’ and ‘Anglo-Celtic’ identity and Racial Studies, 25, pp. 390–414. in contemporary British and Australian political LEES, L. (2004) Emancipatory City? Paradoxes discourse, Ethnicities, 2, pp. 163–188. and Possibilities. Thousand Oaks, CA: Sage JOHNSTON, R. J. and SEMPLE, R. K. (1983) Publications. Classification Using Information Statistics. LEVER-TRACEY, C. and QUINLAN, M. (1988) A Catmog No. 39. Norwich: Geo Books. Divided Working Class. London: Routledge & JOHNSTON, R., POULSEN, M. and FORREST,J. Kegan Paul. (2002) Rethinking the analysis of ethnic resi- MAGGIN, P. J. (2004) Urban Regeneration, Com- dential patterns: segregation, isolation or con- munity Power and the (In)Significance of centration thresholds in Auckland, New Power. Newport, NSW: Ashgate-Gower Asia Zealand, Geographical Analysis, 34, pp. Pacific. 245–261. MARKUS, A. (2001) Race, John Howard and the JOHNSTON, R., POULSEN, M. and FORREST,J. Remaking of Australia. Crows Nest, Sydney: (2003) The ethnic geography of New Zealand: Allen & Unwin. a decade of growth and change, 1991–2001, MCALLISTER, I. and KELLEY, J. (1984) Asia Pacific Viewpoint, 44, pp. 109–130. Immigrants’ status, earnings and politics, JOHNSTON, R., POULSEN, M. and FORREST,J. in: J. JUPP (Ed.) Ethnic Politics in Australia, (2004) The comparative study of ethnic residen- pp. 53–68. St Leonards, Sydney: Allen & tial segregation in the USA, 1980–2000, Unwin. Tijdschrift voor Economische en Sociale MCALLISTER, I. and MOORE, R. (1989) Ethnic Geografie, 95, pp. 550–569. prejudice in Australian society: patterns, inten- JOHNSTON, R., POULSEN, M. and FORREST,J. sity and explanations. Office of Multicultural (2005) Ethnic residential segregation across an Affairs, Canberra. urban system: the Maori in New Zealand, MCCONAHAY, J. B. (1986) Modern racism, 1991–2001, Professional Geographer, 57, pp. ambivalence, and the modern racism scale, in: 115–129. J. F. DOVIDIO and S. L. GAERTNER (Eds) JUPP, J. (1984) Ethnic politics in Australia: a research Prejudice, Discrimination and Racism, pp. 91– agenda, in: J. JUPP (Ed.) Ethnic Politics in Austra- 125. New York: Academic Press. lia, pp. 4–13. Sydney: Allen & Unwin. MILES, R. (1989) Racism. London: Routledge. KAUFMAN, E. (2004) The Rise and Fall of Anglo- MODOOD, T. (2000) Anti-essentialism, multi- America: The Decline of Dominant Ethnicity in culturalism, and the ‘recognition’ of the US. Cambridge, MA: Harvard University religious groups, in: W. KYMLICKA and Press. W. NORMAN (Eds) Citizenship in Diverse KELLY, P. (Ed.) (2002) Multiculturalism Reconsid- Societies, pp. 175–195. Oxford: Oxford ered: Culture and Equality and its Critics. University Press. Cambridge: Polity Press. MONEY, J. (1999) Xenophobia and xenophilia: KIVISTO, P. (2002) Multiculturalism in a Global Pauline Hanson and the counterbalancing of Society. Oxford: Blackwell. electoral incentives in Australia, People and KLOCKER, N. (2004) Community opposition Place, 7, pp. 7–19. towards asylum seekers in Port Augusta, South NEWMAN, S. (1995) Population policies of Australia, Australian Geographical Studies, 42, Australian political parties: explicit and implicit, pp. 1–17. People and Place, pp. 46–47. KOBAYASHI, A. (2004) Cultural ‘race’ approaches NUNN, C. A., CROCKETT, H. J. and WILLIAMS,J. to cultural geography, in: J. S. DUNCAN,N.C. A. (1978) Tolerance for Nonconformity: A JOHNSON and R. H. SCHEIN (Eds) A Companion National Survey of Americans’ Changing 720 JAMES FORREST AND KEVIN DUNN

Commitment to Civil Liberties. San Francisco, Multicultural Australia, pp. 173–188. CA: Jossey-Bass. St Leonards, Sydney: Allen & Unwin. OFFICE OF MULTICULTURAL AFFAIRS (1989) ROBINSON, V. (1987) Spatial variability in atti- National Agenda for a Multicultural Austra- tudes towards ‘race’ in the UK, in: P. JACKSON lia...Sharing Our Future. Canberra: Australian (Ed.) Race and Racism: Essays in Social Government Printing Service. Geography, pp. 161–189. London: Allen and PARADIES, Y. (2005) Anti-racism and indigenous Unwin. Australians, Analyses of Social Issues and ROSE, P. (1997) Tempest-Tost: Race, Immigration Public Policy, 5, pp. 1–28. and the Dilemmas of Diversity. New York: PEDERSEN, A. and WALKER, I. (1997) Prejudice Oxford University Press. against Australian Aborigines: old-fashioned SCHAEFER, R. T. (1975) Regional differences in and modern forms, European Journal of Social prejudice, Regional Studies, 9, pp. 1–14. Psychology, 27, pp. 561–587. SIMMONS, J. and BOURNE, L. S. (2003) The PEDERSEN, A., BEVEN, J., GRIFFITHS, B. and Canadian urban system, 1971–2001: responses WALKER, I. (2004) Attitudes toward indigenous to a changing world. Research Paper No. 200, Australians: the role of empathy and guilt, Centre for Urban and Community Studies, Journal of Community and Applied Social Psy- University of Toronto. chology, 14, pp. 233–249. SMITH, A. W. (1981) Racial tolerance as a function PEDERSEN, A., CLARKE, S., DUDGEON, P. and of group position, American Sociological GRIFFITHS, B. (2005) Attitudes toward Review, 46, pp. 525–541. indigenous Australians and asylum seekers: SMITH, S. (2003) Electoral support for Australians the role of false beliefs and other social- against further immigration, one nation and psychological variables, Australian Psycholo- unity at the 1999 and 2003 New South Wales gist, 40, pp. 170–178. state elections. Honours thesis, Geography PEDERSEN, A., CONTOS, N., GRIFFITHS,B.ET AL. Program, University of New South Wales, (2000) Attitudes toward Aboriginal Australians Sydney. in city and country settings, Australian Psychol- SMITH, S. (2005) Chasing the ‘race’ and ‘ethnic’ ogist, 35, pp. 109–117. votes: profiles of minor parties and their pos- PEDERSEN A., WALKER, I. and WISE, M. (2005) itions within the Australian political landscapes ‘Talk does not cook rice’: beyond anti-racism of ‘racism’, 1999 and 2003 New South Wales rhetoric to strategies for social action, Australian state elections, in: T. KHOO (Ed.) The Body Psychologist, 40, pp. 20–30. Politic: Racialised Political Cultures in Austra- PETTIGREW, T. F. (1959) Regional differences in lia, pp. 1–9. Proceedings of the Australian anti-Negro prejudice, Journal of Abnormal and Studies Centre Conference, University of Social Psychology, 59, pp. 28–36. Queensland Australian Studies Centre, PETTIGREW, T. F. and MEERTENS, R. W. (1995) Brisbane, and Monash University National Subtle and blatant prejudice in western Europe, Centre for Australian Studies (http://www.asc. European Journal of Social Psychology, 25, uq.edu.au/bodypolitic/; accessed 20 April pp. 57–75. 2006). PHILLIPS, T. (1998) Popular views about Austra- STATE ELECTORAL OFFICE OF NEW SOUTH WALES lian identity: research and analysis, Journal of (2003a) Australians against further immigra- Sociology, 34, pp. 281–302. tion. Registered party information sheet, p. 1. POULSEN, M., JOHNSTON, R. J. and FORREST,J. State Electoral Office, Sydney. (2001) Measuring intra-urban residential segre- STATE ELECTORAL OFFICE OF NEW SOUTH WALES gation: a knowledge-based method for classify- (2003b) One nation NSW political party. ing intra-urban enclaves, Environment and Registered party information sheet, pp. 1–12. Planning A, 33, pp. 2071–2082. State Electoral Office, Sydney. POULSEN, M., JOHNSTON, R. and FORREST,J. STATE ELECTORAL OFFICE OF NEW SOUTH WALES (2004) Is Sydney a divided city ethnically?, (2003c) Unity Party. Registered Party Infor- Australian Geographical Studies, 42, pp. 356– mation Sheet, p. 1. State Electoral Office, 377. Sydney. RAJKOWSKI, P. (1987) In the Tracks of the TWINE, F. W. and WARREN, J. W. (Eds) (2000) Camelmen. North Ryde, Sydney: Angus & Racing Research, Researching Race: Robertson. Methodological Dilemmas in Critical RIZVI, F. (1996) Racism, reorientation and the Race Studies. New York: New York University cultural politics of Asia–Australia relations, in: Press. E. VASTA and S. CASTLES (Eds) The Teeth are VALENTY, L. O. and SYLVIA, R. D. (2004) Smiling: The Persistence of Racism in Thresholds for tolerance: the impact of racial CONSTRUCTING RACISM IN SYDNEY 721

and ethnic population composition on the vote J. REX and D. MASON (Eds) Theories of Race for California propositions 197 and 209, The and Ethnic Relations, pp. 20–41. Cambridge: Social Science Journal, 41, pp. 433–446. Cambridge University Press. VASTA, E. and CASTLES, S. (Eds) (1996) The Teeth YOUNG, I. M. (1990) Justice and the Politics of are Smiling: The Persistence of Racism in Multi- Difference. Princeton, NJ: Princeton University cultural Australia. St Leonards, Sydney: Allen Press. & Unwin. ZUKIN, S. (1986) Gentrification: culture and capital YINGER, J. M. (1986) Intersecting strands in the in the urban core, Annual Review of Sociology, theorisation of race and ethnic relations, in: 13, pp. 129–147.