Peoples and Spaces in a Multicultural Nation: cultural Group Segregation in Metropolitan Australia

Transcription

1 Espace populations sociétés Space populations societies 2006/ Diasporas et grandes métropoles Peoples and Spaces in a Multicultural Nation: cultural Group Segregation in Metropolitan Australia Peuples et espaces dans une nation multiculturelle : la ségrégation des groupes culturels dans l Australie des métropoles James Forrest, Michael Poulsen et Ron Johnston Édition électronique URL : ISSN : Éditeur Université des Sciences et Technologies de Lille Édition imprimée Date de publication : 1 avril 2006 Pagination : ISSN : Référence électronique James Forrest, Michael Poulsen et Ron Johnston, «Peoples and Spaces in a Multicultural Nation: cultural Group Segregation in Metropolitan Australia», Espace populations sociétés [En ligne], 2006/1 2006, mis en ligne le 01 avril 2008, consulté le 04 novembre URL : eps.revues.org/1049 ; DOI : /eps.1049 Ce document est un fac-similé de l'édition imprimée. Espace Populations Sociétés est mis à disposition selon les termes de la licence Creative Commons Attribution - Pas d'utilisation Commerciale - Pas de Modification 4.0 International.

2 151 ESPACE, POPULATIONS, SOCIETES, pp James FORREST Michael POULSEN Ron JOHNSTON Department of Human Geography Macquarie University NSW 2109 Australie School of Geographical Sciences University of Bristol Bristol BS8 1SS Grande-Bretagne Peoples and Spaces in a Multicultural Nation: Cultural Group Segregation in Metropolitan Australia Australia ranks today as the fourth largest immigrant receiving country in the last 150 years, after the USA, Canada and Brazil, followed closely by Argentina [Burnley, 2001, p. 4] From the first British settlement in 1788, and for the next 150 years, emigration to Australia overwhelmingly comprised people from England, Ireland and Scotland. Until the mid-1970s most indigenous Aborigines were excluded even from citizenship. Elements of this past situation continue to be seen in the degree of contemporary socioeconomic disadvantage experienced by Australia s Indigenous people, and which parallels the situation of contemporary immigrant streams. Meanwhile, nearly all non-caucasians were kept out under a White Australia policy. The past 50 years, however, have seen major changes to the makeup of Australia s population, accompanied by significant changes to official attitudes towards immigration and immigrants. After World War 2, issues of defence and economic development led to a demand for population growth which could not be met from traditional British sources. Australia took in eastern and central Europeans as war refugees, then southern Europeans in the 1950s and 1960s. However, improving national economies in Europe led to a decline of Europeans wanting to emigrate and from the mid-1960s, immigrants from the eastern Mediterranean were added to the population mix. World approbation attaching to racist immigration policies led to the abandonment of the White Australia policy in the early 1970s and the acceptance of people from anywhere in the world, subject only to policy settings relating to needed qualifications and skills; Asians dominated subsequent in-flows [Jupp, 2001]. Officially, Australia became a multicultural society comprising a culturally dominant Anglo (or Anglo-Irish) host so-

3 152 ciety and a large number of ethnic minority groups: the 2001 census recorded 191 ethnic (ancestry) groups. More than 23 per cent of today s population is born overseas. A major feature of the immigrant stream of the past 50 years is the degree to which it has concentrated in the major urban areas (Figure 1). Until the late 1960s, Melbourne was the favoured destination; more recently Sydney, with increasing status as a world city, has attracted a growing proportion of the new arrivals. Adelaide shared in the new ethnic mix in the 1950s and 1960s, as did Perth and the major industrial centres of Geelong, Newcastle and Wollongong [Birrell and Rapson, 2002, p. 10]. More recently, the major focus of new arrivals has been on Sydney, Melbourne and Perth. By the turn of the century, people from non-english speaking birthplaces comprised nearly a quarter of Sydney and Melbourne s populations, 15 per cent of Perth s, 13 per cent of Adelaide s and 10 per cent of Brisbane s [Birrell and Rapson, 2002, p. 11]. How have these immigrant groups fared in their new country of residence? Figure 1. Map of Australia showing cities studied Studies of the absorption of ethnic minority groups traditionally portray them as initially concentrated in lower status occupations and low-cost housing areas, more or less segregated from other minority groups and the host population. The development of ethnic enclaves is commonly seen to result from economic disadvantage and social discrimination, associated with recency of arrival and rates of in-flow, coupled with a view of residential propinquity as providing cultural, social, economic and, in some cases, physical security. At any one time, however, the presence and size of ethnic minority concentrations depends on a balance of social, economic and demographic circumstances, countered by shifts away from the enclaves as social and economic assimilation allows group members to choose whether or not to move out of such areas. Rarely, however, does assimilation occur evenly. Rather, Portes and Zhou (1993) introduced the concept of segmented assimilation to describe the range of possible outcomes as processes of reducing social and economic distance are worked through. To this has recently been added a specifically geographic notion of everywhere different

4 153 to describe how assimilation is influenced by local as well as national circumstances. This expansion of the concept of segmented assimilation, in the context of Australia s absorption of cultural minorities and its immigration history, acts to produce a set of relationships among peoples and spaces in the nation s major urban areas which is uniquely Australian. This study analyses those relationships. TOWARDS A MULTICULTURAL SOCIETY Much of our understanding of the process of initial residential segregation and longer term desegregation of cultural minority groups in cities comes from the United States experience of a melting pot process leading to the assimilation or Americanisation of immigrants and Indigenous people over the past century or more (though not of Black immigrants from the country s rural south). This is very much a general social process in which the Government has played little direct part except, for example, by removing certain areas of discrimination through civil rights legislation in the 1960s, though with little effect. Countries like Australia, Canada and New Zealand, on the other hand, have, over the past four decades, adopted a multicultural approach, where the overall aim is acceptance of cultural pluralism and structural (economic and political) integration. Some see this as a contrasting multicultural model of the acceptance of cultural minorities, where the maintenance of residential enclaves is viewed as a positive feature supported by governments; there is economic but not necessarily social assimilation, accompanied by anti-discrimination legislation. Difference is both accepted and celebrated, with outcomes which are less pronounced than those between ethnic clusters and a white host society. This has been referred to in the Australian context as assimilation in slow motion [Jamrozik, Boland and Urquart, 1995, p. 110] or as a gentle form of assimilation and incorporation [Jupp, 2002, p. 121]. Multiculturalism has as its goal co-existence of different cultural groups in a context of equal opportunity for all, as opposed to assimilation into a new hybrid society where the norms of the host society nevertheless continue to predominate. In Australia, the official view of a multicultural society is one where national solidarity transcends ethnic identity and where, according to the federal Department of Immigration and Multicultural and Indigenous Affairs, social harmony among the different cultural groups in our society is promoted. Of course, reality is a little less perfect, with tensions between the host and ethnic minorities, and in terms of racist attitudes towards some cultural minorities [Dunn et al., 2004], reflecting degrees of social acceptance. Alternatively, some social scientists and commentators see multiculturalism as a recipe for spatial (and implicitly social) segregation, hence pluralism [Healy and Birrell, 2003]. Nevertheless, the reality seems to favour spatial assimilation. Thus Burnley s (1994) work on ancestry data from the 1986 Australian Census, comparing major ethnic groups who had arrived from southern Europe Greeks, Italians and the eastern Mediterranean Lebanese in the 1960s and 1970s found that: they were all segregated in poorer, inner parts of the city in the first generation; the second generation (their children) moved out to middle and sometimes outer suburbia; while for the third generation (their grandchildren) Burnley found no evidence of continuing ethnic concentrations. Economic disadvantage Australia experiences little overt discrimination, at least as to the occupational attainment of immigrant groups. Rather, immigrant labour market segmentation is dominated by economic disadvantage, in terms of time of arrival, competence in English, and skills gained prior to arrival. A study of occupational backgrounds of immigrants prior to arrival in Australia in [Bureau of Immigration Research 1990] suggests four labour market segments: 1. Immigrants from northwestern Europe, Britain and New Zealand show an emphasis on white collar (managerial-

5 154 professional-clerical/sales) backgrounds with smaller numbers in the blue collar (manual) occupations; 2. Dominantly blue collar manual skills among those from Italy, Greece and (the former) Yugoslavia, with a major focus on semi-skilled (processing and production workers) and unskilled (labourer) backgrounds; 3. Among later refugee streams from Lebanon, Vietnam and elements from China and Sri Lanka (the Tamils), where there was an even heavier concentration in the semi-skilled and unskilled sectors; 4. Immigrants from Asia, other than those who entered under the humanitarian (refugee) program, showed a strong white collar background; the exception is Philippinos, where the greater emphasis is on clerical/sales and semi-skilled work; Time of arrival adds to the complexity of potential immigrant assimilation patterns in that those who arrived during the 1950s and 1960s came during the post-world War 2 economic boom. More recent immigrant streams have been disadvantaged by three decades of economic restructuring, deregulation and structural job-shedding, with periods of growth interspersed with economic recessions in the mid-1970s, the early 1980s and the early- to mid-1990s; after each recession there was contraction in the labour market resulting from cost-cutting technological advances especially in manufacturing, to the particular disadvantage of immigrants who are concentrated in that sector. These factors impacting on immigrant labour force participation time of arrival and labour force segmentation are apparent in the present-day occupations of immigrant groups (Table 1). This is significant because income reflects occupation, and thereby influences differential access to the housing market. Unemployment is high among refugees from the Middle East and southeast Asia, and among those from the former Yugoslavia who, as refugees, arrived more recently; an exception is the Italian born, who are relatively concentrated in skilled and managerial occupations. Unemployment Table 1 : Occupations for host society and major immigrant groups for major immigrant receiving cities in 2001 (Sydney-Melbourne-Perth)

6 155 also remains high among those from Vietnam, even though most of them arrived up to two decades or so ago. On the other hand, most other immigrants from Asia, who arrived relatively recently and mainly under the skilled immigration program, have maintained the structure of their pre-emigration labour force experience; where unemployment rates are higher than for the Australian born; this largely reflects refugee components. Aboriginal groups Among disadvantaged groups in Australia s major cities, the situation of Aborigines, in that they are involved in rural-to-urban migration, closely parallels that of overseas immigrants, which is why their situation is best considered in the context of immigrant groups. Many who migrate into the larger, metropolitan centres gravitate towards areas of low socio-economic status. In Sydney, many Aborigines experience social problems associated with high unemployment and concentration in either relatively run-down inner city enclaves or in western Sydney on public housing estates [Kohen, 2000]. Nevertheless, only a tiny proportion, some 0.5 per cent, of Sydney s Aborigines live in ghetto-like concentrations, where they comprise 70 per cent or more of total local populations [Johnston et al., 2001]. Social disadvantage and racism In spite of the predominance of economic disadvantage impacting on the absorption of immigrants from non-english speaking backgrounds, there is some evidence of racist attitudes towards, and certainly of social distancing from, some cultural minority groups. Among the overseas-born in Queensland, Timms (1969) found close relationships between group marriage patterns and perceived social distance on the one hand, and indexes of dissimilarity spatial segregation on the other. Those from English-speaking backgrounds were least socially distant, while those from central and eastern Europe, and then southern Europe, were perceived as increasingly distant and residentially segregated. More recently, from a national survey also using social distance scales, McAllister and Moore (1991) found that among Australians as a whole, Asians and Muslims faced the highest levels of prejudice, while towards the other end of the scale were, successively, central and eastern Europeans, southern Europeans (though the difference between these two groups was marginal) and, least socially distant, the British-born; apart from Muslims and Asians, these results confirm Timms earlier findings. Aborigines as a whole occupied an intermediate position, as did Jewish immigrants among four religious groups. Most recent work on racism in Australia by Dunn et al. (2004) indicates the lasting nature of the previous findings. Based on degree of concern about inter-group marriage from a survey of respondents in New South Wales and Queensland, Dunn et al. found considerable unevenness of esteem for others. Muslims were the major out-group, i.e. those who are seen as not belonging or fitting into the Australian national culture, with 54 per cent of all respondents expressing concern. This was more than twice the levels experienced by other groups. Aboriginal, Asian and Jewish Australians remain as significant out-groups, but in the intermediate range of acceptance (25-30 per cent expressed concern). There was much more acceptance of Italian-born Australians (less than 12 per cent expressed concern), while few (8 per cent) were concerned about the British-born. The degree of social distancing suggested above, however, is not borne out in studies of urban residential segregation. Johnston et al. (2001) conclude that Sydney, Australia s largest EthniCity, is not, as the largely United States based literature on world cities suggests, a fragmented city. Rather, multicultural Sydney is best characterised by ethnic group and host society mixing. What is not known, and what this study investigates, is whether this finding for one major EthniCity is generally true of Australia s major urban areas and if so, to what extent. Based on the potential for everywhere different, we might anticipate significant differences among Australia s metropolitan and major regional cities linked to ethnic group mix and main period of arrival in terms of occupational characteristics.

7 156 ANALYSING ETHNIC SEGREGATION Studies of ethnic group segregation in cities have most commonly used relative measures of spatial concentrations, in particular the indices of dissimilarity and segregation. However, such measures (which only report the average situation in each city) take into account only one aspect difference or unevenness of one population group compared with another of what Massey and Denton (1988) argue is a multi-dimensional situation including isolation (hence potential for interaction), clustering (of ethnic groups relative to each other), concentration (the physical space occupied by each group) and centralisation (relative to the city centre). But of greater concern is that these indices are all relative measures, relative, that is, to the particular study area at a point in time or space; change the time of study (e.g. one city at different census dates) or the spatial context (e.g. comparing a number of cities) and the base line varies. This makes comparative studies extremely difficult. Peach (1996) introduced threshold analysis as a way forward by focusing on the absolute percentages of any sub-area s ethnic population present at a single threshold level. Peach set this at 30 per cent or higher, that is, the percentage of an ethnic group who live in sub-areas where they comprise 30 per cent or more of local populations. Developing this approach, Poulsen et al. (2001) introduced a range of absolute thresholds to link Peach s methodological innovation to Boal s (1999) conceptualisation of the processes involved in ethnic segregation into four scenarios representing a heterogeneity-homogeneity continuum from: assimilation where (p. 588) difference reduces and social and spatial boundaries dissolve, through pluralism, with individual ethnic groups encouraging group diversity and the maintenance of group boundaries and segmentation, or much sharper divides among ethnic groups and between them and the host society such that (p. 590) deteriorating inter-ethnic relations [characterised by] insecurity and mistrust contrast with the mild separations of pluralism, to polarisation, an extreme case of segmentation where some cultural minority group members are virtually excluded from many areas into ghetto situations. To these four scenarios, Marcuse (1997) has added another, called citadels, the exclusive residential areas occupied by the wealthiest members of the host society. Poulsen et al. (2001) combined these two main elements host and cultural minority groups and added an additional, non-isolated host community group to produce a composite homogeneity (ghetto)-heterogeneity (mixed areas)-homogeneity ( white citadels) continuum. Poulsen et al. (2001) thus provide a standardised measuring tool designed to facilitate comparative and temporal studies. Cut-offs defining membership types were determined on the basis of comparative research and remain constant to allow analysis of differences across time and space. The methodology first divides urban sub-areas into two main types: minority enclaves, where the host society is in a minority, and host communities, where the host society group forms a majority. Each type is then divided into sub-types which form the continuum from extreme polarized enclaves (ghettos) at one end to isolated host communities (citadels) at the other. There are four minority enclave types, comprising: 1. Extreme polarised enclaves (EPE or ghettos), with a high concentration of a minority group (at least 60 per cent) and at least 30 per cent of that group s total (citywide) population living in such areas; 2. Polarised enclaves (PE) with one minority group substantially concentrated (at least 60 per cent, as with the ghettos), and not sharing the area with substantial numbers of other minorities, but without the 30 per cent of total population requirement of the ghetto situation; 3. Mixed-minority enclaves (ME), shared by two or more minority groups but with few (<30 per cent) members of the host society; 4. Associated assimilation-pluralism enclaves (PluE), where the host population is a significant minority (30-50 per cent) and one or more cultural minority groups are at least 20 but less than 60 per cent of the total.

8 157 There are two host community sub-types: 1. Non-isolated host communities (NIHC) where per cent of the total population comprise the host population and minority group(s) make up a substantial minority; and 2. Isolated host communities (IHC or citadels) with 80 per cent or more of their populations from the host society and minority groups largely absent. Urban sub-areas are categorised according to this typology. The more polarised an urban society is, the greater the concentration towards each end of the continuum. But in a more multicultural (as opposed to plural) society, mixed minority enclaves should be more common. Where spatial integration is the norm, as is suggested for most community types in Australian cities, then most residents should be living in mixed areas, in non-isolated host situations and, to a lesser extent, in associated assimilation-pluralism enclaves. ETHNIC RESIDENTIAL SEGREGATION BY ANCESTRY, 2001 Because the Australian census does not ask any questions about ethnic identity, the analyses presented here rely on ancestry data. Such data have only been collected once before, in 1986, but because of problems with the data, little use was made of them. In 2001, people were asked to nominate more than one ancestry if applicable. Only the first two ancestry groups nominated were recorded, accounting for 94 per cent of the total (i.e. the other 6 per cent gave more than two ancestries and this information was lost). Among those who nominated their ancestry as Australian, however, there is no way of distinguishing between originally English-speaking and non-english speaking backgrounds, but this is less of a problem because the major period of non-anglo-irish immigration commenced after World War 2. For this study, data on the five mainland state capitals (Hobart does not have a significant cultural minority component) and three major regional industrial cities (Figure 1) are from Collection Districts (CDs), the smallest unit of enumeration with populations of some people. The urban areas used are Urban Centre Localities (UCLs), or continuously built up areas. The host population is defined as those who nominated themselves as of Australian, English, Scottish, Irish, Welsh or New Zealand ancestry, that is, those from English-speaking, Anglo or Anglo- Irish cultural backgrounds (North Americans were not included because the census data do not differentiate between them and non-english speaking background Central and South Americans). A white-nonwhite dichotomy or white nation conceptualisation, often used in studies of other major immigrant receiving nations, is inappropriate in the Australian context [Forrest and Dunn, 2006]. Table 2 : Major Urban Areas: Total population, percentage non-host and segregation of total population

9 158 Two features of segregation patterns in Australia s largest cities stand out. First, a sizeranking based on their proportion of nonhost groups (Table 2, 2 nd column) shows three groupings: a high proportion of nonhost ancestral groups more than 40 per cent in Sydney and Melbourne; smaller proportions around 30 per cent in the other mainland State capitals; and lower figures among the three industrial cities. Second, a salient feature of urban residential patterns is a relatively non-segregated residential landscape (Table 2, right hand columns). There are no ghettos (extreme polarised enclaves), or polarised enclaves either apart from small traces in Sydney and Melbourne. Rather, the majority of ethnic groups in Newcastle all of them are living in types 5 and 6 areas where those of Australian ancestry predominate. Relatively small proportions are to be found in mixed ethnic (types 3 and 4) areas. We turn now to urban residential patterns of host and individual ethnic groups and particular cities. URBAN ETHNIC SEGREGATION PATTERNS BY CITIES Table 3 : Ethnic (ancestry) composition of urban populations as per cent of non-host population The ethnic composition of the eight urban areas, shown in Table 3, falls broadly into three main clusters: those that are relatively evenly distributed across most of the urban areas; those which tend to be concentrated in about half of the cities; and a few which are more concentrated. Northwestern Europeans dominate those who are widely distributed, are socially least distant from the host community and who arrived in Australia mainly in the first two decades after World War 2. This group is largely spatially assimilated.

10 159 Surprisingly, there is a major difference between two ethnic groups from southern Europe who arrived during the same period: Italians, who are to be found in most urban centres, and Greeks, who are largely concentrated in three or four cities. Also among the most widely distributed are the Chinese, who have been present in Australia since the gold rushes of the 1860s, though in larger numbers only in the past decade or so; Indians, principally those who have entered Australia recently as skilled immigrants, as professionals or skilled trades people; and Filipinos, including a proportion who are women married to Australian men. Ethnic groups from the former Yugoslavia Croatians, Macedonians and Serbians along with those from other parts of eastern Europe, are also widely distributed, apart, perhaps, from in Brisbane. Less widely distributed are those of Greek ancestry, who went into semi-skilled and unskilled (manufacturing-related) occupations. Italians, on the other hand, gravitated more into skilled occupations, which may help to explain their wider distribution. Among the least widely distributed are refugee groups, including people from the Middle East generally, Pacific nation peoples the New Zealand Maori, people from Papua-New Guinea, the Melanesian and Polynesian islands, and Australia s indigenous Aborigines. Among the latter, there is little concentration in the main metropolitan or industrial centres; instead most have gravitated to regional cities in northern Australia. Among post-1960s refugee and Other Middle Eastern groups, the major focus has been on Sydney, Melbourne. These variations to the multicultural nature of the various cities highlight the importance of a segmented assimilation approach to any explanation of the detail of segregation patterns. Sydney and Melbourne Until the 1960s, Melbourne was the most favoured immigrant destination. It was the manufacturing capital of Australia in the 1950s and 1960s, but with global economic restructuring in the 1970s it subsequently became associated with old economy activities. Thus the largely semi-skilled and unskilled (manufacturing) workforce of the 1950s to 1970s from Greece and Italy settled in Melbourne over Sydney at a ratio of nearly 2:1, although in Sydney ethnic Italians have since become more economically assimilated. From the 1980s, however, Sydney, as an emerging world city and increasingly a focus of new economy (multinational business and IT) activities, has been the increasingly favoured destination of Middle Eastern and Asian immigrant streams. Sydney s share of new immigrant arrivals is still increasing, with 39 per cent locating there, compared with 22 per cent in Melbourne [Birrell and Rapson, 2002]. (See TABLE 4a on next page) Despite the fact that they are Australia s major receptors of immigrant groups over the past half century, the great majority of the host population in Sydney and Melbourne show a substantial mixing of ethnic minority and host society groups in non-isolated host society (type 5) areas; less than 10 per cent live in citadel areas (Table 4a). Northwest Europeans follow the same pattern. Conversely, very few ethnic minority members are located in polarised enclaves: this is the case, however, with some Aborigines in Sydney but not in Melbourne, along with small proportions of ethnic Middle Easterners, Southeast Asians and Chinese in both cities. On the other hand, there are also substantial numbers of many ethnic ancestral groups in the mixed (type 3 and 4) enclaves. Thus ancestral groups from northwestern Europe are largely spatially assimilated, with small numbers in the mixed enclaves. The situation of southern Europeans largely reflects period of arrival, with Italians and Greeks having higher proportions living in areas where those of Australian ancestry predominate, whereas ancestral groups from the former Yugoslavia, some of whom arrived in the 1960s and 1970s, and some more recently as refugees from the fighting in the later 1980s and 1990s, are more segregated. Mainly refugee groups from Lebanon, Vietnam and other southeast Asian origins in particular are the most segregated, in spite of being among the earlier arrivals of the 1970s and 1980s, while Asians who arrived under the skilled immigrant program, though more recently, are more spatially assimilated. Perth, Adelaide and Brisbane Among the second group of cities, Perth currently ranks third and not very far behind

11 160 Table 4a : Percentage distribution of ethnic (ancestry) groups across the six categories of the CD classification for selected urban areas, 2001 Melbourne as a major receiver of immigrants, with higher proportions of more recent arrivals than Adelaide or Brisbane (Table 4b). In terms of segmented assimilation, this is reflected in the lower proportions of northwest Europeans resident in type 6 (host majority) areas (Table 4b). More recently arrived refugee groups also stand out, not, however, as enclave dwellers, but intermixed with substantial host (type 5) populations: Lebanese and Vietnamese in particular are much more concentrated in such areas than in Adelaide or Brisbane, but also have high proportions in minority ethnic (type 3) areas. Among southern Europeans, those of Italian and Greek ancestry are less segregated, compared with Adelaide, though this is not true of the Asian (non-refugee) segment. (See TABLE 4b on next page) According to Burnley (2001, p. 321), Perth is a city of considerable ethno-cultural diversity where: The typical residential pattern... is of residual communities of European settlers in low to middle income neighbourhoods but with increasing admixture of new arrivals from South Africa, the USA, the UK, Hong Kong, Singapore and Malaysia... Refugees from Vietnam, Iraq and the Balkans [are largely concentrated] in the new public [government rental] housing estates... Meanwhile, residual European communities such as Croats, Serbs, Dutch and Italians, survive in suburban areas... that were formerly market garden lands. The spatial assimilation of cultural minorities in Perth largely parallels the experience of Adelaide. A higher proportion of Aborigines, people from Oceania and from northwestern Europe are to be found in nonisolated host community areas than in Adelaide, but for most other groups the proportions are very similar. On the other hand, the more recent and continuing role of Perth as a receiver of immigrants is apparent in the higher proportions of various cultural groups in mixed (dominantly non-host) enclaves than is the case in Adelaide, where the experience of receiving new immigrant groups is less recent.

12 161 Table 4b : Percentage distribution of ethnic (ancestry) groups across the six categories of the CD classification for selected urban areas, 2001 Germans were the major non-anglo immigrant group in South Australia in the 19 th century. They made a major contribution to the rural settlement of the hill country and farming areas north and east of Adelaide [Burnley 2001, pp ]. Since World War 2, the major source of migration flows was from Britain and continental Europe, plus, as in Brisbane, a strong German component from rural to urban migration flows of the earlier generations of German settlers. Italians and Greeks were also attracted by an expanding building and construction industry in the 1950s and 1960s. However, reflecting the decline of the city s industrial base with economic restructuring, Adelaide s share of immigration from Asia and the Middle East was rather less than Perth s (Table 4b). Spatial integration in Adelaide reflects its earlier history as an immigrant destination (Table 4b). There are higher proportions of host ancestry groups living in enclave types 3 and 4 mixed minority ethnic group enclave areas, and more of the northwestern Europeans living in type 6 areas among host group majorities. However, differences between Adelaide and Perth are less obvious among most other ancestral groups southern Europeans and (non-refugee) Asians. Refugee groups, on the other hand, are noticeably more segregated in Adelaide than they are in Perth. After World War 2, Brisbane was subject to Queensland State government policy which was anti-diversity with respect to receiving non-british immigrants directly. Many ethnic Italians and Greeks and, later, people of Asian descent moved to Brisbane s middle income areas after first becoming established elsewhere in Australia. During the late 19 th and early 20 th centuries too, there was a strong presence of German and Dutch settlers in rural areas of southern Queensland, many of whom over time migrated into Brisbane, which largely accounts for their strong, but largely spatially assimilated ethnic presence in Reflecting these historical factors, Brisbane has a much larger proportion of the host society living in isolated (type 6) areas, more so than any other city with a high non-host population (Table 4b). But there is also a high proportion of all ethnic groups living in type 5 areas of Australian ancestral majorities, in higher proportions than in Adelaide or Perth. Where Brisbane is different from the other two state capitals is in the generally smaller proportions living in mixed enclave situa-

13 162 tions, though certain groups, such as Asians, notably Vietnamese refugees, stand out as important exceptions with a strong presence in mixed (types 3 and 4) ethnic minority enclaves. There is also a significant New Zealand Maori presence in Brisbane, but most of these live in host community areas. The industrial cities: Newcastle, Wollongong, Geelong Australia s three main industrial cities demonstrate the importance of local economic histories to the reception experience of immigrant groups. Among the post-world War 2 immigrant streams, a significant number sought out these cities. Of equal importance, the ending of the White Australia policy coincided with the first impacts of economic restructuring and the decline of manufacturing employment in Australia. As a consequence, post-white Australia immigrant groups are generally underrepresented in the industrial cities. Newcastle pre-dated Wollongong and Geelong as a centre of manufacturing and heavy industry: its main period of growth was in the 1920s and 1930s. Post-1945 economic development was diminished compared with the other two cities, in major part because of national policy left over from the threat of Japanese invasion to diversify the locations of Australia s heavy industry concentrations. Nevertheless, there was considerable population growth after World War 2. Among non-host groups, many war refugees from northwest Europe were attracted to Newcastle. The second major stream, in the early 1970s, comprised Macedonians. Further ethnic group in-flows were largely halted by recession in the manufacturing and heavy industry sector from the mid-1970s immigrants from Asian countries, for instance, represent a noticeably lower proportion of the non-host society than in the state capitals (Table 3). Spatially, the contrast between Newcastle and the other two industrial cities is considerable, with the great majority of ethnic groups in Newcastle spatially absorbed into type 5 and 6 areas among host society groups (Table 4c). Table 4c : Percentage distribution of ethnic (ancestry) groups across the six categories of the CD classification for selected urban areas, 2001

14 163 In Wollongong, the iron and steel industry expanded rapidly after 1945, attracting large numbers of war refugees, especially from Germany and Italy, then later skilled and unskilled immigrants, notably Croatians, Serbians and Macedonians, along with Turks and Greeks. But here too, recession in the heavy industry and manufacturing sector from the mid-1970s produced high unemployment rates, to the continuing economic disadvantage of ethnic minorities. This impacted on segregation patterns in two main ways. First, it reduced employment opportunities for earlier immigrant streams. This is reflected in relatively high proportions of most non-host ancestral groups (though less so generally than in Geelong) still concentrated in mixed enclave situations where ethnic minorities form a majority. Second, it made Wollongong a much less attractive destination for post-white Australia policy immigrants than it had to European immigrants in the 1950s to early 1970s; those that did come are largely to be found today living in type 5, non-isolated host society areas. Ethnic segregation is generally greater in Geelong than in Wollongong. This reflects unemployment rates as a result of economic restructuring which have historically been higher in Victoria than in New South Wales. This is especially true for those of Croatian, Macedonian, Serbian and other Balkan ancestries, as well as with Vietnamese and southeast Asian refugee groups generally. However, the reverse is also true: there is greater segregation in Wollongong among ethnic groups from other parts of Asia Chinese and Indians, for instance. On the other hand, other European ethnic groups are more spatially assimilated in both cities, while another feature to Geelong s advantage is the higher proportion of nearly all ethnic groups living in type 6 areas where members of the host society are in an absolute majority. CONCLUSIONS A number of empirical findings emerge from this study. The first is the degree of spatial assimilation or desegregation of non-host ethnic ancestral groups and Aboriginal people, with evidence of the importance of the segmented nature of assimilation in accounting for differences to the incorporation of the different ethnic groups. Because most spatial distancing derives from economic rather than social disadvantage, time of arrival is the major factor accounting for segregation in that it is closely linked to opportunities for economic assimilation. Thus greatest de-segregation is apparent among those who arrived from Europe in the first decades after World War 2 Germans, Dutch, other northwestern Europeans, and Italians and least among more recent immigrant streams from Asia, the Middle East and the Balkans. That there are also variations among these latter groups largely relates to their status at time of entry, which is especially true of the refugee segment. Aborigines, New Zealand Maori and people from Oceania, though small in number, generally tend to fall into an intermediate position on degree of segregation, which is consistent with their ranking on social distance scales. The second is the amount of variation among the cities studied here, bound up as it is with the immigration and economic histories of individual cities or groups of cities. Australia s two major EthniCities, Sydney and Melbourne, both with claims to world city status, exhibit both a high level of host society integration with non-host groups and the presence of mixed enclaves. Prior history of the multicultural model of ethnic minority group desegregation based primarily on economic considerations suggests the ultimately transitory nature of these enclaves. This is especially important in view of Sydney s predominance as a receptor of more recent, post-white Australia immigrant flows. Among the other State capitals, ethno-cultural diversity is the norm, except in Brisbane, which was not a first destination for many cultural minority groups, and where in-migration of ethnic groups previously established elsewhere in Australia has long been more important. The only exception is Adelaide, where higher proportions of ethnic groups live in mixed enclaves, in spite of the relative slowing down of in-flows to that city in recent times.

15 164 Among the three main industrial cities, two Wollongong and Geelong share a combination of relatively high proportions living in mixed, non-isolated host society areas and a higher-than-average proportion in mixed enclaves. This is attributed to economic disadvantage associated with significant economic restructuring and job losses linked to the decline of heavy industrial and manufacturing. Newcastle, on the other hand, where the main period of industrial growth was earlier, with diminished economic growth and fewer immigrants after 1945, is today a relatively spatially integrated urban society. On the evidence presented here, patterns of ethnic group segregation and intermixing with the dominant host (Anglo) society overwhelmingly reflect the importance of economic issues, of economically segmented assimilation, largely time of arrival and prevailing economic circumstances applying nationally over social discrimination in their impact on the spatial assimilation of Australia s peoples. The outcome, at least so far, is the essentially transitory nature of enclave developments consistent with a multicultural model of the absorption of highly diverse ethnic ancestral populations. This is not to suggest that there are not tensions affecting cultural groups who are seen as socially most distant to the point of being racially discriminated against, or that issues of national identity and cultural homogeneity are not present. What it does suggest is that the spatial, and by implication social separation of host and non-host cultural groups is more consistent with dominant multicultural aspirations than it is in many other economically developed nations. BIBLIOGRAPHIE BIRRELL B. and RAPSON V. (2002), Two Australias: Migrant settlement at the end of the 20 th century, People and Place, 10, pp BOAL F.W. (1999), From undivided cities to undivided cities: assimilation to ethnic clensing, Housing Studies, 14, pp BURNLEY I.H. (1994), Immigration, ancestry and residency in Sydney, Australian Geographical Studies, 32, pp BURNLEY I. H. (2001), The Impact of Immigration on Australia, Oxford University Press, South Melbourne. Bureau of Immigration Research (1990), Settler Arrivals , Statistical Report No. 2, Australian Government Publishing Service, Canberra. DUNN K.M., FORREST J., BURNLEY I. and McDonald A. (2004), Constructing racism in Australia, Australian Journal of Social Issues, 34, pp HEALY E. and BIRRELL B. (2003), Metropolis divided: the political dynamic of spatial inequality and migrant settlement in Sydney, People and Place, 11, pp JAMROZIK A., BOLAND C. and URQUART R. (1995), Social Change and Cultural Transformation in Australia, Cambridge University Press, Cambridge. JOHNSTON R., FORREST J. POULSEN M. (2001), The geography of an EthniCity: residential segregation of birthplace and language groups in Sydney, 1996, Housing Studies, 16, pp JUPP J. (ed.) (2001), The Australian People: An Encyclopedia of the Nation, its People and their Origins. Cambridge University Press, Oakleigh. JUPP J. (2002), From White Australia to Woomera. The Story of Australian Immigration, Cambridge University Press, Port Melbourne. KOHEN J., First and Last People: Aboriginal Sydney, in J. Connell (ed), Sydney The Emergence of a World City, Oxford University Press, South Melbourne, pp MASSEY D.S. and DENTON N.A. (1988), The dimensions of residential segregation, Social Forces, 67, pp McALLISTER I. and MOORE R. (1991), Social distance among Australian ethnic groups, Sociology and Social Research, 75, pp MARCUSE P. (1997), The enclave, the citadel and the ghetto: what has changed in the post-fordist city? Urban Affairs Review, 33, pp PEACH C.(1996), Does Britain have ghettoes?, Transactions, Institute of British Geographers, NS, 22, pp PORTES A. and ZHOU M. (1993), The new second generation: segmented assimilation and its variants, Annals, Academy of Political and Social Science, 530, pp POULSEN M., JOHNSTON R. and FORREST J. (2001), Intraurban ethnic enclaves: introducing a knowledgebased classification method, Environment and Planning A, 33, pp TIMMS D.W.G. (1969), The Urban Mosaic: Towards a Theory of Residential Differentiation, Cambridge University Press, Cambridge. WILLIAMS L. and BATROUNEY T. (1998), Immigrants and poverty, in R. Fincher and J. Nieuwenhuysen (eds), Australian Poverty: Then and Now, Melbourne University Press, Melbourne, pp