ETHNIC DIVERSITY AND THE CITY Ceri Peach COBISS 1.01 In the period 1945 to 1996, there has been a net movement of about 33 million migrants and refugees across international borders in western Europe. By way of comparison, in the period 1820 and 1985, there was a net migration of about 56 million people into the USA. The population of Europe is larger than that of the USA in the period in which most of the US migration was taking place, but the impact in demographic and psychological terms on Europe, a continent that was previously thought of as an exporter of population has been considerable. The post-war period in Europe has seen four main waves of inward immigration: • Reflux of Europeans from their colonial territories overseas or from their areas of historic settlement in Eastern Europe. 1945 - 1974 about 16 million of such people exist or existed. • Influx of worker and family immigration 1950 - 1999 (Gastarbeiter) and dependants, of whom there were in 1993 about 12 million in western Europe and by the late 1990s closer to 20 millions. • Refuge. 1989 - 2000 The new wave of asylum seekers from the former Socialist bloc : the former USSR, Yugoslavia as well as the disposessed from earlier Islamic upheavals in Iran, Afghanistan, Iraq, Kurdistan and Algeria, civil wars in Sri Lanka, Somalia Rwanda and Burundi and elsewhere. • Fourthly, we have seen the growth of the two newish movements. There are the white-collar internationalists. These are the chess pieces of transnationalism. They include Japanese and Americans parachuted into corporate city slots (de Lannoy, 1975; Glebe, 1986; Kesteloot and van der Haegen, 1997). Alongside them are Europeans on the move, settling for a while in other people’s countries. Paul White, for example, writes how the French have produced a distinctive flavour to a part of Kensington in London, playfully dubbed ‘frog valley’ alongside the young Australian’s‘kangaroo valley’ (White, 1998a, 1998b) • Finally, there have been lifestyle developments, particularly in the display of sexuality and the formation of gay villages, which have added to the diversity of the city (Adler and Brenner, 1992; Knopp,1995; Binnie, 1995). However, space does not allow us to develop this theme. Post war Western Europe has experienced a demographic shock. The continent which considered itself to be an exporter of population has found itself to be the destination of substantial waves of immigration. The causes and consequences of these Dve domovini • Two Homelands 15 • 2002, 13-32 movements are complex and contested, so that this paper attempts an overview of the major contributory factors. The essence of the account is that the first phase was one of reflux : retreat from Empire (typified by Germans from Slavic lands, French from North Africa, Dutch from Indonesia and British from their colonies); the second was influx : worker immigration caused by expanding economies and ageing populations (typified by Turks to Germany, Algerians to France, West Indians and Indians to Britain); the third is eflux : the flight of refugees and asylum seekers into Europe (typified by Iranians and North African fleeing the Islamic revolutions and by East European nationals fleeing the breakup of the former socialist bloc). Whereas the political response to reflux was sympathy and the response to influx was neutral acceptance, the reaction to asylum seekers and refugees has been closer to panic. Western Europe is battening down the hatches and trying to repel potential refugee seekers. At the same time, there are profound differences in the strategies adopted by different countries in their acceptance or denial of their new minority ethnic populations. There are also powerful political conjunctions in the movements: the populations of the reflux form the hard core of the right wing opposition to influx and efflux. Immigration into across European borders falls into three main (but not completely distinct) types. (1)Returnees and expellees from former colonial territories or territories where national boundaries were pushed eastwards at the end of the Second World War. About 16 million of such people exist or existed. (2) Worker migration (Gastarbeiter) and dependants, of whom there were in 1993 about 12 million in western Europe. (3)The new wave of asylum seekers, fleeing upheavals mainly in the former socialist countries, but also crises elsewhere in the world. These numbered about 2 million from 1980 to 1991 (SOPEMI, 1992, 132). (1) REFLUX The largest number of returnees and expellees came to Germany. The Potsdam post war settlement removed East Prussia and severed Silesia from Germany, allocating the territory to the USSR and Poland. Ethnic Germans were expelled from these areas and those such as Sudetenland, which had been annexed by Germany during the war. The Soviet Zone of Germany was eventually created into the new state of the German Democratic Republic. It is thought that between 1945 and 1960, 9 million Germans fled to West Germany from the annexed territories. Between 1945 and the building of the Berlin Wall in 1961, it is thought that a further 3 million East Germans fled west. Altogether, about 12 million Germans fled to the west. In addition to these directly displaced or fleeing people, there was a continuous stream of ethnic Germans (Volksdeutsch) whom the Federal Republic virtually bought from the East European states, where they had been settled for hundreds of years. Anyone with German ancestry was entitled to come. Thus groups like the Silesian Poles, who had relatives in the German army in the past, for example, could claim German citizenship. Poland, Rumania and the USSR had considerable populations of ethnic Germans. As conditions in the Socialist countries deteriorated, the numbers moving to Germany increased substantially. Between 1968 and 1984, 652,897 ethnic Germans moved to the Federal Republic. From 1985 to 1991, a further 1,332,829 settled. Thus, from 1968 to 1991, nearly 2 million ethnic Germans returned to Germany, in addition to the 12 million discussed above in the earlier phase. Table l-.Aussiedler in Germany according to countries of origin, 1968 to 1991 Poland ex USSR Bulgaria Yugoslav Rumania Czechos Hungary Other Total 1968-1984 365234 72664 195 9982 147528 47011 7065 3218 652897 1985 22075 460 7 191 14924 757 485 69 38968 1986 27188 753 5 182 13130 882 584 64 42788 1987 48419 14488 12 156 13990 835 579 44 78523 1988 140226 47572 9 223 12902 949 763 29 202673 1989 250340 98134 46 1469 23387 2027 1618 34 377055 1990 113253 147455 27 530 107189 1324 1038 11 370827 1991 40129 147320 12 450 32178 927 952 27 22199 1006864 528846 313 13183 365228 54712 13084 3496 1985726 Source : Statistisches Jahrbuch, 1992, fur die Bundesrepublik Deutschland, table 3.39, p 91 Return movement to other European countries has been on a totally different scale from that of Germany. In 1962, after Algerian independence, there was a mass return of about 1 million pieds noirs. French settlers in that country (Guillon, 1974 ). The Netherlands experienced a return flow of about 250,000 to 300,000 Dutch and Indonesians in 1953, when Indonesia became independent. In a few years up to 1975, when Surinam became a separate state, about 104,000 Surinamese left for the Netherlands in order to claim Dutch rather than Surinamese citizenship (Peach 1991 20 -21). However, this latter movement is not an ethnic return movement in the sense of the German and French movements. In 1975 after Angolan independence from Portugal, about 400,000 Portuguese settlers fled that country, a substantial proportion returning to their native country. Settler return to Britain has undoubtedly occurred, but not in the dramatic way of the cases mentioned so far. The 1991 census shows that of the 1,653,002 born in the New Commonwealth, that is to say former colonial territories, living in Great Britain, 328,080 (or 20 per cent) were white plus 250,000 whites from Australia, New Zealand, Canada and South Africa. So there were over half a million whites from the Commonwealth. In western Europe, these reflux migrants have generally been quietly absorbed, although they may have had important political effects in some cases in their support for right wing parties. WORKER MIGRATION 1945 - 1990 However, when one refers to immigration in Europe, it is the worker movement which is assumed to be the focus of attention. Although earlier movements had taken place, such as the Polish worker migration to the northern coal fields of France in the 1920s and 1930s, the post war movement of labour into Europe was a revolutionary change of direction from the previous patterns of migration affecting the continent. Between 1950 and 1975, worker migration to western Europe, together with dependants, grew from practically nothing to about 12 million. Table 2 shows the position for some of the largest groups in 1990 for selected west European countries This table should be read in two directions : (1) along the bottom rows and (2) down the final column. Reading along the bottom rows, Belgium has nearly 1 million Immigration into western Europe, late 1990s Number of immigrants originating from specified countries c.1997 (but see note below) Number of people (millions) India Total immigrant population of destination country 3.8m 1.5m .3m J^0.9m Caribbean POPULATION Yugoslavia Turkey Algeria Morocco 3.5m 1.3(h (^> Pakistan Bangladesh India Caribbean OEC (J^> Other Yugoslavia Spain Italy Turkey Note: France c.1990 Tunisia Morocco Algeria Selected West European Countries of Destination for non-EC Sources c.1997 Source: Peach,1997 foreigners, about 40 per cent of which come from outside the EC; France has 3.6 million of whom two thirds are non EC. Germany has the largest number of foreigners of whom two thirds originate from outside the EC. Three-quarters of the 700,000 foreigners in the Netherlands are non EC and two thirds of the British total of 2.5 million overseas origin population are non EC. Reading down the final column, the largest individual source of migrants into Europe is Turkey with 2.2 million (three quarters of whom are in Germany); Italy is the next largest supplier with 1.5 million (over a third of whom are in Germany, but with substantial numbers in Switzerland, Table 2: Major Sources and Destinations of International Immigrants to Western Europe, late 1990s Source SOPEMI1999 COUNTRY OF DESTINATION COUNTRY OF 1997 1990 1997 1997 1998 1997 1997 1997 ORIGIN Belgium France Germany Netherlands UK Switzerland Italy Spain TOTAL Turkey 73,8 197,7 2.107,40 172,7 63 79,6 2694,2 Algeria 8,9 614,2 23,1 646,2 Morocco 132,8 572,7 82,8 145,8 131,4 111,1 1176,6 Tunisia 4,7 206 28,1 1,5 48,9 289,2 Iran 101,5 16,5 118 Afghanistan 30 30 Pakistan 17 69 86 Bangladesh 23 69 92 India 12 139 14,3 5,7 171 Caribbean 183,2 122 305,2 0 1178,1 Yugoslavia (former) 52,5 721 46,7 313,5 44,4 Bosnia-Herzgovina 281,4 281,4 Croatia 206,6 206,6 Poland 6 283,3 31,3 320,6 Italy 295,8 252,8 607,9 17,4 89 342,3 22,6 1627,8 Spain 47,4 216 131,6 17,6 94 17 523,6 Portugal '25,3 649,7 132,3 8,7 136,5 38,2 990,7 Greece 10,2 362,5 5,3 16,5 394,5 Ireland 19,5 448 6,4 473,9 France 103,6 103,9 16,5 74 55 28,3 34,3 415,6 Germany 33,3 126,8 75 94,7 40,1 49,9 419,8 0 11376,3 0 18561,9 Other 141,9 835 7.365,80 790,3 918 218,8 868,5 238 total 903,2 3596,6 7.365,80 1549 2066 1340,8 1240,7 499,8 Belgium and France). Morocco, rather surprisingly, has more citizens overseas than Algeria, 1 million as opposed to 650,000, but both heavily concentrated on France. Portugal, with 860,000 citizens in western Europe, particularly France, is the other major supplier of note. 1950-1973 Essentially the period of worker migration divides into the explosive growth stage, between 1950 and the oil crisis of 1973 and the period of entrenchment from 1973 and family reunification until the present. By 1973, there were about 15 million foreign immigrants living in the industrialised countries of north west Europe. There numbers remained stable at that level until 1983 and since that time have shown a decrease. The period from 1945 to 1973 witnessed a massive increase in immigration as the European economies recovered from the war. The recovery manifested itself earliest in Britain and France, with Germany acting as a source of immigrants until the early 1950s. German demand for foreign labour grew from just over 100,000 in 1958 to over 1 million in 1966. By 1968, Germany had overtaken France as the largest concentration of foreign labour Miihlgassner 1984, 73). In France and Germany, immigration took place as a result of planned bilateral agreements between those governments and third parties1. In Britain and the Netherlands, the movement was largely unplanned, although specific industries and services made initiatives which were important beyond the scope of directly involved authorities (Peach, 1991). Essentially, what happened during the period 1945 to 1973 was that there was an outward ripple effect in terms of the catchment area of migration to western Europe. In the period up to the late 1950s, the movement was dominated by local effects -migration from adjacent or close countries : Irish to Britain, Finns to Sweden, Italians to Switzerland, France and Germany and so on. In the later period, to 1973 and beyond, the ripples extended far beyond the European catchment to include the Mediterranean Basin, sub Saharan Africa and South Asia. Within this perimeter, there was a geographical partitioning of territory. The Mediterranean Basin was effectively divided along the axis of Italy into a western French sphere of influence and an eastern German sphere. The Maghreb and sub Saharan 1 France set up a national immigration office (ONI) in 1945 and thereafter concluded a series of bilateral agreements with different governments. Between 1945 and 1950, agreements were made for the recruitment of German and Italian workers. Further agreements were made with Greece (1954) Spain (1961) Morocco, Tunisia, Portugal, Mali and Mauritania (1963) Senegal (1964) Yugoslavia and Turkey (1965). These arrangements were in addition to the supply of Algerian labour, freely available before independence in 1962 and which continued thereafter. Germany concluded bilateral arrangements with Italy (1955) Spain and Greece (1960) Turkey (1961) Morocco (1963) Portugal (1964) Tunisia (1965) and Yugoslavia (1968) (Peach 1992, 116). Africa, Spain and Portugal fell into the French sphere. Yugoslavia, Greece and Turkey fell into the German sphere and Italy was divided between the two. Countries such as Austria and Switzerland showed a German pattern of migration sources. Belgium followed a more French pattern. There was, in addition, a British pattern which was of overseas flows from former colonial territories in the Caribbean and South Asia. The Dutch shared some of the characteristics of the British pattern and some of those of the German (see mapl). 1974 ■ 1993 This rapid expansion of immigration into western Europe was brought to a dramatic halt by the oil crises of 1973/1974. The Yom Kippur War and the consequential rapid rise of oil prices following the Arab countries’ embargo on exports, produced a major dislocation in western economies. Both France and Germany imposed a halt on immigration and made attempts to make migrants return to their home countries. Between 1974 and 1981, the overall number of foreigners living in Europe stabilised. Generally speaking, the number of workers decreased while the number of dependants increased. For example, between 1973 and 1981 the number of foreign workers in Germany decreased from 2,416,000 to 2,096,000 and in France from 1,813,000 to 1,600,000. On the other hand, the total foreign population increased from 4,127,000 in Germany to 4,630,000 and in France from 4,043,000 to 4,148,000 (Peach 1987, 36). The period since 1974 has seen the maturing of many of the immigrant groups, so that, with some of the earliest established communities, the sex ratios have become more even and the majority of the ethnic minority communities have been born to their parents in the countries of settlement. We are no longer talking about simple immigrant minorities. Taking the sex ratio first, in Britain in the 1991 census, the numbers of Caribbean born men and women are almost identical. The number of Indian born women slightly exceeds that of Indian born men (1055 women per 1,000 men). For more recently arrived groups such as the Bangladeshis, the ratio was 883 women per 1,000 Bangladesh-born men. In Germany the ratio of Turkish women per 1,000 Turkish men rose from 554 in 1974 to 856 in 1989 (SOPEMI 1990, 143). Although it is difficult to establish what proportion of Turks has been born in Germany, 41 per cent were under 18 years of age at the time of the 1987 census, compared with 17 per cent for the German population as a whole. In France in 1985, there were 701 Algerian women per 1,000 Algerian men (SOPEMI 1990, 145) indicating a continuing substantial imbalance of the sexes. THE NEW WAVE: POST 1988 ASYLUM SEEKERS Europe 1985 - 1991 In the period leading up to and following the fall of the Berlin Wall and the associated collapse of socialist systems in Eastern Europe, there was a substantial growth in the number of asylum seekers moving into western Europe from the East. In 1983, there were less than 54,000 but in 1991, the annual figure reached nearly half a million (see table 3). The overwhelming majority of these asylum seekers moved to Germany, whose constitution obliged the government to accept them. In 1991, the number of asylum seekers in Germany reached over a quarter of a million. The political pressure to repeal this provision was intense and in 1993, all party was secured. Table 3: Inflows of Asylum Seekers into selected European countries, 1980 - 1998 (thousands) Belgium Denmark France Germany Greece Italy Netherlands UK Sweden Total 1980 2.7 0.2 18.8 107.8 1.3 9.9 140.7 1981 2.4 0.3 19.8 49.4 0.8 2.9 75.6 1982 3.1 0.3 22.5 37.2 1.2 4.2 68.5 1983 2.9 0.3 22.3 19.7 0.5 3.1 2 4.3 4 59.1 1984 3.7 4.3 21.6 35.3 0.8 4.6 2.6 4.2 12 89.1 1985 5.3 8.7 28.8 73.8 1.4 5.4 5.6 6.2 14.5 149.7 1986 7.6 9.3 26.2 99.7 4.3 6.5 5.9 5.7 14.6 179.8 1987 6 2.7 27.6 57.4 6.3 11 13.5 5.9 18.1 148.5 1988 4.5 4.7 34.3 103.1 9.3 1.4 7.5 5.7 19.6 190.1 1989 8.1 4.6 61.4 121.3 6.5 2.2 13.9 16.8 30 264.8 1990 13 5.3 54.7 193.1 4.1 4.7 21.2 30.2 29.4 355.7 1991 15.4 4.6 47.4 256.1 2.7 27 21.6 73.4 27.4 475.6 1992 17.3 13.9 28.9 438.2 2 2.6 20.3 32.3 84 639.5 1993 26.2 17.1 27.6 322.6 0.8 1.3 35.4 28.5 37.6 497.1 1994 14.7 6.7 26 127.3 1.8 52.6 42.2 18.6 289.9 1995 11.7 5.1 20.4 127.9 1.4 1.7 29.3 55 9 261.5 1996 12.4 5.9 17.4 116.4 1.4 0.7 22.9 37 5.8 219.9 1997 11.8 5.1 21.4 104.4 4.4 1.9 34.4 41.5 9.6 234.5 1998 22 5.7 21.8 98.7 2.6 4.7 45.2 79.8 13 293.5 (\ 190.8 104.8 548.9 2,489.4 48.5 80.6 337.2 485.7 347.2 4,633.1 Source : SOPEMI, 1997, 1998 and 1999 editions Movement into Southern Europe and the Schengen Agreement Attention on worker migration and asylum seeking until the late 1980s has been focused on northern Europe. However, in the late 1980s immigration into southern Europe became a notable phenomenon. Table 1 shows that Greece and Italy, which had not figured on the tables of refugees before 1983, began to register arrivals in that year. Restrictions on immigration had been progressively tightened in northern Europe, beginning with British legislation in the 1962 Commonwealth Immigrants Act, but becoming most clear with German and French legislation to halt worker migration in the aftermath of the 1973 oil crisis. Southern Europe, however, which had been an exporter rather than a receiver of immigration had no comparable legislation. Since the mid 1980s there has been the arrival, often illegally, of migrants from North and sub-Saharan Africa in Italy, Spain and Portugal. Because southern European countries had traditionally been exporters rather than importers of labour, arrivals were not subjected to rigorous immigration controls. However, as those controls tightened in northern Europe, the south became an easier point of access. The Schengen agreement allows for the free movement of populations without passports between the signatory states within the EC. Thus, once across the external boundaries of the Community movement of the undocumented becomes difficult to control. Italy began to regularise its illegal immigrants in 1987 (SOPEMI 1990, 21). Of these, the largest group at the end of 1990 came from North Africa and sub Saharan Africa (80,000 Moroccans, 41,000 Tunisians and 25,000 Senegalese) but there were also 30.000 Yugoslavs and 34,000 Filipinos. In 1991, Italy had a sudden influx of about 38.000 Albanian refugees. About 17,000 of them were immediately repatriated (SOPEMI 1992, 66). There was also a sharp increase (from 99,509 in 1989 to 238,365 in 1990) in the number of residence permits granted to foreigners from Africa (largely North Africans). Spain has, in recent years, received what has been called the ‘new wetbacks’ The original wetbacks were the Mexicans who entered the USA illegally over the Rio Grande. The new wetbacks come across the Straits of Gibraltar, in frail fishing craft and are often made to swim the last dangerous stretch. Some do not make it. Between 1989 and 1992, 36 bodies have been found by the Spanish authorities on either side of Tarifa, in the main entry zone (The Economist. 12.9.92, 56). In 1985 and 1986, Spain mounted an operation to regularise illegal immigrants and again in 1991. About 40 per cent of the 100,000 or so who were successful, were Moroccan. In 1991, Spain introduced visa requirements for citizens of Maghreb countries (SOPEMI, 1992, 77). Conclusion Western Europe has changed from a sub-continent of emigration to a sub-continent of immigration. It has changed from a set of countries, which between 1918 and 1945 have sought to make themselves internally more homogeneous, into a set of coun- tries with dramatically changed ethnic compositions. The chillingly labelled process of ‘ethnic cleansing’ in the former socialist states of eastern Europe has a major potential to add to this change in the west. The migration frontiers of Europe have changed, during the 1990s from the south to the east. The south continues to send migrants, but they are coming to the countries of southern Europe. The process of absorption of the new minorities is very different in the main receiving countries. There are, I believe, two main types (1) the assimilationist model of the British, the French and the Dutch and (2) the accommodationist model of the Germans and the Swiss. This is a bold generalisation which hides many differences between countries placed in the same categories, but which nevertheless, holds important truths. The assimilationist model stems from the colonial histories of the British, French and Dutch. Trapped by their own political rhetoric, they gave citizenship to their colonial subjects. They did not expect any but the elite to use their rights to gain access to the metropolitan countries. The political integration of minorities into those states has been achieved. The vote was given before the minorities arrived. The accommodationist model of the Germans is fraught with greater difficulties. The Turkish minority is large and well established. The second generation is substantial and German educated. However, with re unification, with the continued inflow of the Aussiedler, the inflow of asylum seekers wished upon Germany by an imposed post war constitution, coupled with a deep economic depression, the political forces ranged against giving citizenship to Turks born in Germany is substantial. The focus of this paper is the worker migration; data on asylum seekers and their settlement into European countries is more recent and less well known. The earliest movement was into Britain starting in 1948 and later into France and Germany. Thence it has spread to fill in the Benelux countries, Switzerland and Austria. From there it has spread to the Nordic countries and now to the southern Atlantic and Mediterranean shores, with Portugal, Spain, Italy and Greece becoming the new front door of Europe. What has been the impact of these flows on the cities themselves ? What strikes me is that despite the different ethnic groups involved, despite the different numbers and proportions formed by these groups, despite the different constitutional positions of the minority population in the various countries, despite the different time-scales over which the movements have taken place, there is an over-riding similarity of the patterns of settlement between the European countries. Not only is there a commonality of pattern, but that pattern is not that of the African American Ghetto. It is the pattern of the ethnic enclave, not the pattern of the ghetto. To begin with, we need to understand the difference between the ghetto and the enclave. The Chicago School made a massive mistake when it fused these two models into a single type. The Chicago School was based on what we can call the three generation model. The first generation was geographically concentrated and socially ghet-toised in the inner city. It did not speak the language, was employed at the lowest levels, was uneducated and married its own kind. The second generation moved out of the inner city to the next urban ring. It was better educated, spoke the English language in America, improved economically; moved up socially and out geographically. It began marrying out. The third generation was suburbanised and homogenised with the general population. It was spatially mixed with the core society and assimilated socially. The problem with this three generational model was that it assumed that the ghetto was part of the general model and not a distinctive spatial formation in its own right. In the Chicago School model, the second generation ethnic enclave was conceptualised as a low-fat version of the first generation ghetto - the same, but a bit less concentrated. In reality, the first generation was never ghettoised in the way in which the black population was ghettoised. The key difference between the black ghetto and the urban enclave was that the ghetto was doubly exclusive. Black areas were almost entirely black; almost all blacks lived in such areas. The ghetto was doubly exclusive Ethnic enclaves, on the other hand, were doubly dilute. Minority ethnic groups only rarely even formed a majority of the population of the areas with which they were identified; only rarely did a majority of the minorities identified with particular areas (their so-called ‘ghettos’) live in such areas. Thomas Philpott’s The Ghetto and the Slum, (1978) demoinstrates the point for Chicago in 1930 at the end of the era of mass migration (table 4). Table 4: 'Ghettoization ' of Ethnic Groups, Chicago, 1930 Group Group’s Group’s ‘Ghetto’ Total ‘Ghetto’ Percentage of Percentage City Population Population Population group ’Ghettoized’ ‘Ghetto’ Group’s Population Irish 169,568 4,993 14,595 2.9 33.8 German 377,975 53,821 169,649 14.2 31.7 Swedish 140,913 21,581 88,749 15.3 24.3 Russian 169,736 63,416 149,208 37.4 42.5 Czech 122,089 53,301 169,550 43.7 31.4 Italian 181,861 90,407 195,736 49.7 46.2 Polish 401,316 248,024 457,146 61.0 54.3 Negro 233,903 216,846 266,051 92.7 81.5 Source : Philpott (1978,141, table 7) It can be seen that while the African American population formed 81.5 percent of the population of the black ghetto and while 92.7 per cent of the black population lived in the black ghetto, only 3 per cent of the Irish lived in Irish areas and they formed only one third of the population of Irish areas. The two most concentrated white groups were the Italians and the Poles. Just under half of the Italians lived in Italian areas and they formed just under half of the the population of Italian areas. The Poles were a little more concentrated : 61 per cent lived in Polish areas and they formed just over half of the population of Polish areas. However, even their levels of concentration were different in kind rather than different in degree from the situation of African Americans. All the European minorities lived in mixed areas. Hardly any of the blacks did. While white ethnic enclaves dissolved over time, black ghettos intensified and expanded territorially in a compact form. European attempts to understand minority settlement in their cities as the American ghetto model are entirely misplaced. The model of the ghetto and the model of the ethnic enclave are totally different in kind, in form, in origin and in their futures. Europe has enclaves not ghettos. To give an example, if we examine the concentration of minority ethnic groups in London in the 1991 census, along the lines of the Chicago table, we find that black groups look much more like the Irish in Chicago in 1930 than like the African Americans (table 5). Table 5: Ghettoisation' of Ethnic Groups at ED level in Greater London 30 per cent cutoff Group Group’s City Population Group’s ‘Ghetto’ Population Total ‘Ghetto’ Population Percentage of group ’ghettoized’ Percentage ‘Ghetto’ group’s population Non-white 1346119 721873 1589476 53.6 45.4 Black Caribbean 290968 7755 22545 2.6 34.4 Black African 163635 3176 8899 2.0 35.6 Black Other 80613 Indian 347091 88887 202135 25.6 44.0 Pakistani 87816 1182 3359 1.4 35.2 Bangladeshi 85738 28280 55500 33.0 51.0 Chinese 56579 38 111 0.0 34.2 Other Asian 112807 176 572 0.2 30.8 Other Other 120872 209 530 0.2 39.4 Irish born 256470 1023 2574 0.4 39.8 Source: Peach, 1996 It is not even that the US levels of concentration and segregation for African Americans have decreased substantially since 1930. If we examine the proportion of African Americans in Chicago living in areas where they form a high percentage of the area’s population, we find the same situation holding as it did in 1930. Of the total black population of 1.3 million living in Chicago PMSA in 1990, 88 per cent were living in areas which were 30 per cent or more black (cf the London figures for the 30 per cent cut off in table 5); 82, per cent were living in areas which were half or more black. Two thirds were living in areas which were 90 per cent black; 29 per cent were living in areas which were 99 per cent or more black and 8 per cent were living in areas 100 per cent black (table 6). Table 6: Percentage of the African American Population of Chicago PMSA, living in tracts of a given black percentage, 1990 Black Percentage of Tract Black Population living Percentage of the total Black in such tracts Population of Chicago in such tracts 100 per cent 111,804 8.4 99 per cent or more 381,347 28.7 90 per cent or more 884,725 66.5 50 per cent or more 1,087,600 81.7 30 per cent or more 1,163,969 87.5 Total Black Population 1,330,636 Total Population 6,069,974 Source: Based on data from GeoLytics CensusCD+Maps US Census 1990 data www.GeoLvtics.com The levels of segregation of African Americans in the US is such that they have been termed ‘hypersegregated’ by Massey and Denton (1993). In a way, the American and the European literature has been dominated too much by the ID and has paid too little attention to the absolute levels of concentration of minority groups. The standard measure of segregation which has dominated the literature is the index of dissimilarity ID. Recent advances in the European literature allows us to look in a broadly comparative way at levels of segregation in different European countries. Glebe and O’Loughlin (1987) led the way with their 1987 volume. Complete issues of Tijdschrift voor Sociale en Economische Geografie (Volume 88, 2, 1997) and of Urban Studies (Volume 35, 10, 1998)were devoted to an analysis of the levels of segregation of minority populations. In addition, Musterd, Ostendorf and Breebart (1998) have synthesised and sys-tematised measurements of segregation for Amsterdam, Brussels, Frankfurt, Dussel-dorf London and Manchester. They show that North Africans in continental Europe and South Asians in Britain manifest the highest levels of segregation of the composite groups which they consider (Amsterdam 40, Brussels 60, Dusseldorf 25, Frankfurtl8, London 44, Manchester 49). While these figures are not exceptionally high, Musterd et al’s methodology of aggregating minorities into ‘South Asian’ and ‘North African’ groups disguises some of the critical factors. Bangladeshis in Britain have an excep- tionally high rate of segregation, averaging 73 on the Index of Dissimilarity across the 11 cities in which they numbered 1,000 or more in 1991 (Peach, 1996, table 9), while Pakistanis averaged 61. Nevertheless, European rates of segregation are moderate when compared with those of African Americans. BRITAIN AS A MODEL FOR FUTURE EUROPEAN EXPERIENCE To some extent we can use the UK experience as a predictor of future developments elsewhere in Europe. The UK was the earliest country to experience large-scale non-European immigration. In this respect, it was ten years in advance of Germany, so that there is a longer track record here than elsewhere. On the other hand, most of Britain’s immigrant population came from its former colonial areas and had British citizenship. This marks a major difference between them and say, the experience of Turks in Germany. The first thing to say about the British experience is that there are two main components, the South Asian and the Afro Caribbean and other African descendants. There are sharp differences both within and between these groups. However, one can detect two basic models of accommodation to British society: the Irish model and the Jewish model. What this boils down to is that although the Caribbean population has a notably more blue collar socio-economic profile than the white population and has a higher proportion living in social housing than the white population, it has nevertheless made significant upward economic mobility and shows a significant degree of intermarriage and co-habitation with the white population (table 7 ). In addition, the mean unweighted average Index of Dissimilarity (ID) for the Caribbean population compared with whites in the 17 cities where more than 1,000 Caribbean population were living in 1991, was 45 (Peach, 1996, table 11). In Greater London, which contained just under 60 per cent of the British Caribbean population, the ID at ward level had shown a continues decrease between 1961 and 1991 from 56 to 41 as outward suburbanisation took place (Peach, 1996, table 13). The South Asian populations (Indians, Pakistanis and Bangladeshis) all show far greater social closure. All three groups show a strong preference for arranged marriages, for married couple households and extended families. However, while the their social conventions show strong similarities of traditional family values, their economic fortunes differ considerably. The Indian population has a white collar profile comparable to that of the white population and indeed rather higher in the professional class (Robinson, 1996) while the Pakistani and to a greater extent the Bangladeshi population is more blue collared, coupled with a low rate of female participation in the forma labour market. All three groups have higher average IDs from whites than do the Caribbeans. The most segregated are the Bangladeshis with an average unweighted ID Ethnic Diversity and the City Table Z- Ethnic Marriage and Co-habitation Patterns, Great Britain, 1991 Unions expressed as percentage of male’s ethnic group White Carib African Other Indian Pakistani Bangla- Chinese Asian Other Total Total Blac Black Black deshi Other Other Per cent White 99.49 0.08 0.03 0.05 0.06 0.01 0.00 0.06 0.12 0.11 100 126803 Black Carib 27.27 67.76 0.97 1.21 0.48 0.24 0.00 0.24 0.36 1.45 100 825 Black African 17.08 5.69 74.02 1.42 0.71 0.36 0.00 0.00 0.00 0.71 100 281 Black Other 51.70 2.04 1.36 42.18 0.68 0.00 0.00 0.00 1.36 0.68 100 147 Indian 6.93 0.10 0.21 0.05 91.06 0.93 0.00 0.26 0.21 0.26 100 1935 Pakistani 5.05 0.00 0.00 0.12 0.72 93.26 0.00 0.00 0.48 0.36 100 831 Bangladeshi 3.00 0.00 0.86 0.00 1.72 0.43 93.13 0.00 0.00 0.86 100 233 Chinese 12.59 0.00 0.00 0.00 0.74 0.00 0.00 86.67 0.00 0.00 100 270 Other Asian 14.71 1.07 0.27 0.27 1.07 1.07 0.27 0.53 79.14 1.60 100 374 Other Other 50.46 0.46 0.23 0.46 1.62 0.93 0.00 0.46 1.16 44.21 100 432 132131 Source: Peach, 1999 of 73; the Pakistanis have an average index of 61 with whites and for the Indians, the ID with whites was 45, just slightly higher than for the Caribbean/white comparison. However, while the Caribbean population was showing trends of decreasing concentration, all three South Asian groups indicated the opposite trend. Indeed, it was notable for the Indian population, with its white collar distribution, that suburbanisation did not equate to dispersal, as in the Chicago three generational model. It followed much more what I have termed the pluralist model of voluntary segregation. France, Germany, Belgium and the Netherlands Turning to the picture which we find in France, Germany, Belgium and the Netherlands, there has been a surge of publications. The pioneering effort was produced by John O’Loughlin and Gunther Glebe in 1987. In the meantime the literature has expanded substantially and in the last three years there have been two issues of major journals, Tijdschrift voor Economnische en Sociale Geografie, 1997 and Urban Studies 1998 (vol 35) devoted to analysis of the position of minority ethnic populations in a European context. Musterd, Ostendorf and Breebaart (1998) have also sought to synthesise and systematise these researchers into a single volume. Despite the fact that the precise composition of ethnic minorities, the history of their movement and the constitutional situations of minorities varies very considerably between European countries, there are nevertheless some over-riding similarities. * Worker migration has been drawn in as a replacement population to jobs which the native population was avoiding. They have filled gaps in the occupational and housing structures. • The movement has generally been to large cities • The movement has generally been to the depressed parts of large cities, though these locations have not always been inner cities. In France, for example, it has been predominantly to les banlieus. the high rise social housing on the periphery. In Amsterdam, the suburban Bijlmermeer seems to have become depressed as a result of minority settlement. • Although the American black ghetto model has been universally invoked by the press and to represent the settlement patterns, in reality these patterns follow what I have termed the ethnic enclave or urban village model rather than the dual characteristics of the black ghetto. • There are areas where minorities form a high percentage of the local population, but they do not constitute a majority of the population over large areas, (The highest proportion that minorities formed at District level in Cologne in 1995 was 50.6 per cent Friedrichs, 1998,1758) • nor do high percentages of minority populations live in areas where they form a majority. • Taking the Index of Dissimilarity at the level of ward, tracts or Stadtteile (about 12,000 persons) as a guide, and benchmarking the values <39 as Low; <59 as moderate and >60 as high, we find that the overwhelming majority fall into the moderate to low category. • Among the high categories are two opposite extremes ;very poor, very recent Muslim populations such as the Bangladeshis in Britain; and rich short-sojourners like the Japanese in London and Diisseldorf. Cultural Landscapes Finally, we come to the transformation of cultural landscapes. If we think of the diversity of the city, we can see a chronological development from the arrival of non-European faces, different clothes, different languages and music, different shops selling unfamiliar goods in the early days to the arrival of new building forms in the later days. In other words, there has been a progression in the cultural landscape from the soft features of people, dress and sound to the new, hard features of building forms. This latter development is particularly associated with the Islamicisation of the wroker and ayslum-seeking populations and is possibly further advanced in Britain than elsewhere in Europe, because of the longer period of settlement. We have run a project in Oxford over the last three years examining the growth of Muslim Mosques, Hindu temples and Sikh Gurdwaras. There are about 1,000 such places of worship which are officially registered in England and Wales and possibly double that number if the smaller unofficial domestic places of worship were to be included. Altogether, the official listing of places of worship for Muslims, Sikhs and Hindus in England and Wales amounts to just under 1,000 buildings. Of these, only about 200 or 20 per cent are purpose built, but, about half of these, or about 10 per cent of the whole, are both purpose built and decorated. Some of them are of cathedral size and make a huge impact on the cultural landscape. The contested nature of some of these buildings can be seen from the severe modulation of their characteristic features. Minarets, in some cases, appear as almost the prehensile representation of a former feature. Cupolas are sometimes understated. Building materials fit in with the surrounding styles. Even so, these buildings are more confident and self-assertive than mosques in German cities. In Diisseldorf, with which I am familiar and in which there are many Turkish families, mosques tend to be located out of sight, in courtyards. Mosques, Gurdwaras and Mandirs are the concrete representation of a self-confident multi-cul-tural society. We have passed the stage of soft additions to the cultural landscape and are now in the era of the concrete. CONCLUSION Conclusion and Discussion Minorities now form a significant proportion of the population of large European cities. In Frankfurt am Main they formed 30 per cent in 1995, in Dusseldorf, Cologne and Munich (Friedrichs 1998) and London , for example, they formed about 20 per cent of the population. Multi ethnic populations are now a permanent part of the city and their permanence is marked by their new religious buildings. Yet European cities are diverse in their particular patterns, in the combination of ethnicities, in the availability of social housing, private rentals and the locations of housing (Kesteloot, van Weesep and White, 1997). In many ways, Kesteloot and van der Haegen (1997) shows that Brussels has a very different pattern of settlement from other cities. Amersfoort and Cortie’s (1973) and Amersfoort and Klerk (1987) work on Amsterdam also shows very distinctive patterns of a suburbanised working class ethnic Surinamese minority. Yet despite these high percentages and relative recency of settlement, they do not seem to be moving on the American black ghetto model. ID Levels of segregation for groups in aggregate or taken singly are generally below 60. Even where groups are above 70 as is the case for Bamgladeshis in Britain and Turks in some German cities and Brussels, they do not form the majority of population of large areas and the proportion living at high densities is generally below 50 per cent of the group. One theme that unites the experience of European cities and the minority popula- tions is a growing Islamicization of the minority, both workers and asylum seekers. This, together with a dissatisfaction with the dry statitistical approach is leading us to a greater appreciation of life-styles, life worlds and sultural landscape. It offers another line of unifying pan Euroepan approaches to minorities and dicversity in cities. There are, of course, very many different types of Islam. In a general sense, then, there are similarities of European cities in comparison to the African American model. There are similarities to the process that has been observed historically with regard to other American minorities. We have seen the growth of privileged new sojourner groups as well as the well known labour immigrants. The Japanese are significant for their high levels of spatial segregation, yet at the same time they teach us that we should not get too worried about high levels of ID. There are positive reasons for groups wishing to stick together as well as the negative reasons that have received the largest amount of attention. REFERENCES Amersfoort, H van and Cortie C (1973)’ Het Patroon van de Surinamse vestiging in Amsterdam 1868 t/m 1970, Tijdschrift voor Economische en Sociale Geografie, 64,283-94. Amersfoort, H van and de Klerk, L (1987) ‘Dynamics of Immigrant Settlement: Surinamese, Turks and Moroccans in Amsterdam 1973-1983, in O’Loughlin, John and Gunther Glebe (Eds) Immigrants in West European Cities. Franz Steiner, Verlag, Stuttgart, 199-222. Burgers, Jack (1998) ‘In the margin of the welfare state : Labour market position and housing conditions of undocumented immigrants in Rotterdam’ Urban Studies 35, 1855-1868 Daley, Patricia O (1998) ‘Black Africans in Great Britian : Spatial Concentration and Segregation’ Urban Studies 35, 10, 1703-1724. Friedrichs, Jurgen (1998) ‘Ethnic Segregation in Cologne, Germany, 1984-1994’ Urban Studies 35, 10, 1745-1764 Giffinger, Rudolf (1998) ‘Segregation in Vienna : Impact of Market Barriers and Rent Regulations 1’ Urban Studies 35, 1791-1812 GLEBE, G. (1986) Segregation and Intra-Urban Mobility of a High-Status Ethnic Group: The Case of the Japanese in Duesseldorf, Ethnic and Racial Studies, 9, pp. 461-483. Guillon, M (1974) ‘Les repatries d’Algerie dans las region Parissienne’ Annales de Geographie, 83 644-675. Kempen, Ronald van and A Sule Oziiekren (1998) ‘Ethnic segregation in Cities :New forms and explanations in a Dynamic World’ Urban Studies 35, 10, 1631-1656. Kempen, Ronald van and Jan van Weesep (1998) ‘Ethnic Residential Patterns in Dutch Cities: Backgrounds, Shifts and Consequences’ Urban Studies 35, 1813-1834 Kemper, Franz-Josef (1998) ‘Restructuring of Housing and Ethnic Segregation : Recent Developments in Berlin’ Urban Studies 35, 1765-1790 Kesteloot, Christian and Cees Cortie (1998) ‘Housing Turks and Moroccans in Brussels and Amsterdam : The Difference Between Public And Private Markets’ Urban Studies 35, 1835-1854. Kesteloot, C, van Weesep, J and White P (1997) ‘ Introduction to the Special Issue on Minorities in West European Cities’ Tijdschrift voor Economische en Sociale Geografie, 88, 2, 99-104. Kesteloot, C and van der Haegen, H (1997) ‘Foreigners in Brussels 1981-1991: Spatial continuity and social change’ Tijdschrift voor Economische en Sociale Geografie, 88,2, 119 Lee, T. R. (1973) Race and Residence: The Concentration and Dispersal of Immigrants in London, Clarendon Press, Oxford Miihlgassner, D (1984) ‘Der Wanderungsprozess’ in Lichtenberger, E (ed) Gastarbe-iter.- Leben in Zwei Gesellschaften, Wien, Bohlau Verlag. Musterd, Sako, Wim Ostendorf and Matthijs Breebart, (19989 Multi-ethnic Metropolis : Patterns and Policies, Dordrecht, Kluwer Academic Pulbishers O Loughlin, John and Gunther Glebe (Eds) Immigrants in West European Cities. Franz Steiner, Verlag, Stuttgart, Peach, Ceri (1999) ‘London and New York: Contrasts in British and American Models of Segregation’ International Journal of Population Geography, 5, 319-351 Peach, Ceri (1998) South Asian and Caribbean Minority Housing Choices in Britain’, Urban Studies 35, 10, 1657-1680. Peach, C (1997) ‘Postwar Migration to Europe : Reflux, Influx, Refuge’ Social Science Quarterly, 78, 2, 269-283. Peach, C (1996b) ‘Does Britain have ghettos ?’ Transactions of the Institute of British Geographers, 22, 1, 216 - 235. Peach, C.. (1991), The Caribbean in Europe : Contrasting Pattern of Migration and Settlement in Britain, France and the Netherlands, Research Paper in Ethnic Relations 15, Centre for Research in Ethnic Relations University of Warwick. Peach, C (1987) ‘Immigration and Segregation in Western Europe sincel945’. Chapter in John O’Loughlin, and Gunther Glebe (Edsj Immigrants in West European Cities. Franz Steiner, Verlag, Stuttgart, 20-34. Phillips, Deborah- (1998 ‘Black Minority Ethnic Concentration, Segregation and Dispersal in Britain’ Urban Studies 35, 10, 1681-1702 Philpott, T.L. (1978) The Slum and the Ghetto, New York, Oxford University Press. Robinson, V. (1986) Settlers, Transients and Refugees, Oxford, Clarendon Press SOPEMI (various years) Trends in International Migration: Continuous Recording System on Migration, Paris, Organisation for Economic Co-operation and Development, White, Paul (1988a) Skilled International Migrants and Urban Structure in Western Europe, Geoforum, 19, pp. 411-422. White, Paul (1998b) ‘The Settlement Pattern of Developed World Migrants in London’ Urban Studies 35, 10, 1725-1744 Woods R I (1976) Aspects of the scale problem in the Calculation of Segregation indices,: London and Birmingham, 1961 and 1971 Tijdschrift voor Economische en Sociale Geografie, 67, 3, 169 - 174. POVZETEK ETNIČNA RAZNOLIKOST IN MESTNA SREDIŠČA Ceri Peach Prispevek raziskuje značilnosti povojne imigracije v zapadno evropskih mestih. Obravnava se dotika razlik in podobnosti med različnimi evropskimi državami terse loteva primerjav in vprašanj o tem, ali so evropski modeli prilagajanja priseljencev različni od tistih, ki so značilni za Združene države Amerike. Prispevek C. Peacha podaja splošen pregled migracijskih gibanj in nekaterih pojavov v nekoliko daljši zgodovinski in komparativni perspektivi. Ceri Peach je redni profesor na Univerzi Oxford, geografski oddelek, Velika Britanija. Je avtor številnih poznanih del o migracijah in demografskih ter geografskih študij.