For the 100 years preceding the 1970s, progressives in Europe and America pursued a politics of solidarity. The left demanded the creation and expansion of the welfare state. Public policy should redistribute income and subsidise, if not deliver directly, essential services such as education and health. The ideal was a society in which the inequalities associated with social class would fade away.

That ideal remains in place, but from the 1970s onwards it has been gradually supplemented by another ideal-the promotion of diversity. Groups that once experienced discrimination would now be accorded recognition. The plethora of languages and cultures created by immigration and the greater tolerance of domestic minority groups, such as gays, would be celebrated in the name of multiculturalism, not trampled in the name of assimilation. Because different groups have different values and understandings of right and wrong, the state would have to be neutral between them. The good society became one in which no person would have to live with a sense of shame because his or her gender, race, sexuality or able-bodied status is different from the majority's.

Herein lies the progressive dilemma of the 21st century. Solidarity and diversity are both desirable objectives. Unfortunately, they can also conflict. A sense of solidarity creates a readiness to share with strangers, which in turn underpins a thriving welfare state. But it is easier to feel solidarity with those who broadly share your values and way of life. Modern progressives committed to diversity often fail to acknowledge this. They employ an over-abstract and unrealistic notion of affinity, implying that we ought to have the same feelings of generosity or solidarity towards a refugee from the other side of the world as we do towards our next door neighbour.

In the 1940s and 1950s, the early days of the British welfare state, this was not such an issue. People believed that they were paying the social welfare part of their taxes to people who were like themselves and who faced the same risks and problems. For most people, paying tax was a kind of enlightened self-interest. Just 25 years later, Britain had become a much more diverse place. This was not just a matter of ethnic diversity. Rather, big differences in values began to emerge between (and within) the generations. It was also the beginning of the end of a long process of national homogenisation which had begun in the late 18th century and encompassed the creation of empire; the forging of new national institutions in the Victorian period; and the two world wars of the first half of the 20th century. Britishness-encapsulated in institutions like the BBC-was shot through with more particular regional and class identities, but it had become a powerful binding force. By the 1970s, that binding force began to weaken and it has been gradually unravelling ever since. In some instances this is welcome, but the price paid is a diminution in solidarity. It seems plausible to suggest, for example, that this weakening is one factor behind the emergence in the 1970s of popular support for tax resistance-if the ties that bind you to increasingly diverse fellow citizens are loosened, you are likely to be less inclined to share your resources with them.

The great 19th-century theorists of progress, from John Stuart Mill to Karl Marx, distrusted the claims of particularistic groups. And TH Marshall's 1949 essay, "Citizenship and Social Class," one of the founding documents of the modern welfare state, posits an inevitable tension between social class, which is particular, and citizenship, which is universal. Marshall is famous for describing three kinds of citizenship rights: civil, political and social, each of which are associated with a particular century, beginning with the 18th. Citizenship rights, especially social ones, promoted what Marshall called "class abatement." He accepted a certain amount of economic inequality as inevitable. What he wanted eliminated was the badge of inferiority associated with a class system as rigid as that of 19th-century Britain.

The theorists of the welfare state were unabashed nationalists. William Beveridge, in his 1942 report, recognised that a system of social insurance would require "a sense of national unity overriding the interests of any class or section." In a 1942 lecture he said: "One of the weaknesses of many reformers in the past is that they have not taken account of the immense feeling of patriotism in the British people, or that loving pride which we have in our country." That was not just wartime rhetoric; the welfare state's communitarianism predated the second world war. In 1931, RH Tawney, in Equality, wrote that "what a community requires... is a common culture, because, without it, it is not a community at all."

In contrast to Friedrich Hayek and other free market theorists, Marshall and Beveridge did not believe that an expansion of the state came at the expense of individual freedom. On the contrary, government power and individual capacity for self-fulfilment were dependent on each other. Government was required not only to frame new social policies but to confront entrenched forces, such as class, which prevented people from full participation in their society. The distrust that welfare state theorists expressed toward particularism was the flip-side of the faith they had in the state and social citizenship. Where the claims of class-or, for that matter, race and religion-were weak, the state would need less power. But because the social effects of class had been so deeply etched into British life, the state would have to be strong in response.

the politics of diversity starts out with assumptions similar to the politics of national solidarity: where we end up in life should not be dependent upon the conditions into which we are born. But from that common starting point, theorists of diversity go on to propose relationships between individuals, groups and the state which are widely at odds with an earlier generation of welfare state advocates.

Proponents of identity politics argue that prejudices against women or racial minorities are so deep that it is naﶥ to expect them to abate in the near future. It therefore serves the cause of social justice to take groups as well as individuals into account. Individuals, after all, are not isolated neutrons bouncing around in a scientist's cloud chamber; they are constituted by the language, ethnicity and gender of their birth; their outlook on the world is shaped by the world's outlook on them. So supporters of identity politics attribute to groups many of the qualities that earlier liberal theorists attributed to people: they have consciousness, they have interests, and, in a just society, they will have rights.

If groups within the nation state receive greater recognition, it must follow that conceptions of over-arching national solidarity must receive less. In the recent report sponsored by the Runnymede Trust, The Future of Multi-Ethnic Britain, Bhikhu Parekh-the report's co-ordinator-argues along exactly these lines. The Parekh report can be read in part as a response to an earlier Home Office document, Race Equality in Public Services. In language indebted to Marshall and Beveridge, the Home Office announced its commitment to creating "One Nation," in which racial diversity would be "celebrated," "everyone recognises their responsibilities," and "everyone is treated according to their needs and rights." Parekh is sceptical of the "One Nation" idea and asks whether we can prevent it being "oppressive or jingoistic."

Beveridge may have spoken of the "loving pride" he had in his country, but the Parekh report famously states that "Britishness, as much as Englishness, has systematic, largely unspoken, racial connotations." National unity in Britain never existed, the report argues, nor should it exist now; a "fixed conception of national identity and culture" is the enemy of diversity and underpins racism. People living in Britain cannot adhere to "the values of one community." Such agreement as those living in Britain require in order to function as good neighbours can be provided by "international human rights standards... that form part of the moral dialogue in all parts of the world."

The Parekh report argues that in a multicultural world people have many identities, some overlapping, some competing. To uproot them from the particular groups which do so much to fashion their identity, and turn them into citizens whose first loyalty must be to the nation-state, is, potentially, to deprive them of important cultural tools: their language, if it is different from the dominant one; their faith, if it is not Christian; their literature, if it is not represented in schools; their sense of self-worth, if society devalues their traditions. As Parekh once put the point in a discussion of the Rushdie affair, Muslim immigrants in Britain "did have an obligation to obey the laws of the land, but British society too had obligations to them... To insist that they and other minorities should accept the British way of life amounted to treating them as second-class citizens." What Marshall and Beveridge saw as a precondition for a properly functioning welfare state Parekh views as domination. "Immigrants owe loyalty to the British state," he says, "but not to British values, customs, and way of life."

The Parekh report does talk about striking "a balance between cohesion, equality and difference," but then proceeds to ignore the problem. Its main recommendations amount to little more than an attempt to broaden the elite which administers the state. More money should be spent on non-English community leaders; minorities should be appointed in greater numbers to boards and commissions; there should be more non-whites in the media and the arts. Aside from demands for increased child allowances and support for asylum seekers, the report seeks no substantial extra funds for social programmes.

Like nearly all advocates of diversity, Parekh is a man of the left who thinks of himself as a friend of the welfare state. Yet welfare plays little or no role in his report's conception of British society. It offers a history of Britain designed to demonstrate how arbitrary the idea of Britain has been. It highlights the break with Rome, the union of the Scottish and English crowns, and Irish incorporation. Notably absent from the list are the 1945 reforms which established the modern welfare state. The report's history of the welfare state encompasses in one sentence three centuries of social reform, from the elimination of the Poor Laws to the outlawing of child labour. This is history through the looking glass. Many Britons, of all backgrounds, take greater pride in the welfare state's contemporary accomplishments, than in the nation's dim, often mythic, past.

But the report's biggest confusion is over whether the state has the right to impose western liberal values on minorities. This is a familiar conundrum for multiculturalists who are also liberals-and Parekh fails to resolve it. The report urges agreement on fundamental values, including "the equal moral worth of all human beings." But then it goes on to suggest that "different individuals and communities should be free to lead their self-chosen lives." "Society may legitimately ban forced marriages," it says, "or those based on duress or deceit, but it should respect the custom in many cultures of basing marriages on introductions arranged by parents." At one point, the report seems to suggest that polygamy ought to be tolerated. Then, in a different mood, it states: "It is legitimate to ban the unequal treatment of women even though this may enjoy cultural authority in certain communities." The philosopher Susan Muller Okin recently asked whether multiculturalism was bad for women. The preference the report gives for immigrant ways of life suggests it is.

In 1956 the sociologists Edward Shils and Michael Young wrote: "Over the past century, British society, despite distinctions of nationality and social status, has achieved a degree of moral unity equalled by no other large national state... constitutional monarchy and political democracy has played a part in the creation and maintenance of this moral consensus." Since that was written, the monarchy has become tarnished as a symbol of moral unity-its place, arguably, taken by the NHS-and Britain has become far more diverse in values, lifestyle and ethnicity. There can be no return to that moral unity. But Marshall's ideas about citizenship matter in this diverse and individualistic world more than they did in the Britain that had emerged out of the second world war. It is when we are most different from each other that we most require agreement on what makes us common members of a society. Agreement to abide by the laws of the land is not enough. Of course the idea of Britishness cannot be fixed; it must evolve. (Consider how the history of empire was taught 100 years ago and how it is taught now-reflecting not only the different faces in the classroom but changed attitudes towards race, gender and violence.) But this does not mean that we can afford to abandon the idea of Britishness altogether in favour of a "community of communities" as Parekh prefers. The hard question is which values and symbols, which cultural and linguistic norms, are the minimum we require to achieve a cohesive Britishness.

america and europe approach this argument from different places on the solidarity/diversity spectrum. Notwithstanding America's nationalism and the strength of its civic culture, it has-as an immigrant society-usually placed diversity above solidarity. This is one reason for the weakness of the American welfare state, and why the group most strongly committed to it, African-Americans, is also the most hostile to unrestricted immigration. But scepticism towards the state in the US precludes a significant role for government in the management of social solidarity. The British Social Attitudes: special international report of 1989 reported that only 36 per cent of Americans thought it was "definitely the government's responsibility to provide health care for the sick," compared with 86 per cent of Britons. There were similar gulfs of 16 per cent (US) to 45 per cent (Britain) on providing a decent standard of living for the unemployed and 17 per cent to 48 per cent on reducing income differences between rich and poor.

In continental Europe the stress on social solidarity, managed by the state, is even greater than in individualistic Britain. For reasons linked to feudalism, Christianity, and the fear of revolution or invasion, the idea of a benign national state protecting all its citizens is a powerful one-most famously embodied by French republicanism, assimilating citizens into a state religion transcending class and ethnicity. But the historic preference for solidarity before diversity is now a source of great tension as well. Nowhere is this truer than in the small, egalitarian, formerly ethnically homogeneous states of Sweden and Denmark.

The governments in both these countries have issued reports along the lines of the Home Office's Race Equality in Public Services. Faced with the prospect of immigrants making up almost 20 per cent of the population within the next 20 years, Danes and Swedes have reasserted national values. The Swedes have embraced the idea of diversity but rejected multiculturalism; the Danes have rejected multiculturalism and barely accepted diversity at all. Yet those reports have been widely accepted by the left in both countries; there is nothing comparable to the Parekh report's challenging of common national norms in the name of difference.

Like Marshall and Beveridge, the theorists of Scandinavia's welfare states insisted on the need for national solidarity. A feeling for what she called "patriotic pride" marked Alva Myrdal's 1941 Family and Nation. Sweden, she happily pointed out, was a relatively uncontested idea. Its boundaries had been fixed for over 100 years. There were, moreover, no big subnationalities within its borders at the time she wrote; there were 34,000 Finns, but "there was never any urge to enforce their total assimilation or to keep them out of the Swedish communion."

But the days when Alva Myrdal could take Swedishness for granted are gone. While Sweden, like most European countries, invited in some guest workers during the booming 1960s and 1970s, the past two decades have seen a second wave of immigration unprecedented in Swedish history. According to a 1997 government report (The Future and Diversity), 17 per cent of all children born in Sweden have at least one foreign-born parent. Within another decade, one quarter of people living in Sweden under the age of 17 will be immigrants or the children of immigrants. Similar conditions exist in Denmark. A recent white paper (Better Integration) projects that by 2020 there will be 800,000 "foreigners" living in Denmark ("foreigner" denotes immigrants and their children, including those born in Denmark). They will then form 13.7 per cent of the population.

In both Sweden and Denmark, the second wave of immigration differs from the first, not only in numbers. Unlike the Turks and Yugoslavs who came as guest workers, the new immigrants include many Africans, Pakistanis and Iraqis. They are arriving at a time when the economy, with its tight labour market, cannot easily absorb them. New immigrants in Sweden, according to the government, are badly educated, have poor language skills, and feel alienated and discriminated against. They also behave differently. In Denmark, where the fertility rate is 1.7 children per couple, immigrants from Somalia reproduce at a rate of 5.6 and those from Iraq at a rate of 4.5.

Priding themselves on their inclusiveness and solidarity, Sweden and Denmark give immigrants the right to vote in local elections and expend great effort thinking about how to integrate them. But these are also countries with big welfare states, in which people are used to passing on large parts of their income to strangers-until recently, strangers quite like themselves. In both languages, the term "welfare state" is synonymous with "welfare society," and the solidarity which underpins it is premised upon society-wide norms and a common morality.

Appeals to common moral understandings are not vague incantations. The Scandinavian welfare states assume agreement over issues which can be highly divisive in many parts of the world. In Scandinavia, for example, it is common to view gender equality as the most recent chapter in the advancing history of rights detailed by Marshall. Social policies emphasising gender equality are needed, because without them, women cannot be full citizens, participating in public and private life. Support for gender equality is so deeply entrenched in these societies that the welfare state is not morally neutral between different conceptions of the family. Were the welfare state to tolerate patriarchal or authoritarian treatment of women in the name of pluralism, it would be violating the welfare state's insistence on equality.

Because the welfare state is not neutral with respect to gender equality, it also takes sides on the question of language. Immigrants are expected to learn the language of their new country as rapidly as possible: "It is expected of foreigners who wish to live in Denmark that they will make an effort to learn Danish and adjust to Danish society," says Better Integration. The reason is clear. Immigrant men, to the extent that they find jobs, will join the larger society. Without command of the language, women who stay at home will be unable to claim the rights offered to them by the welfare state. Hence the state must work its way inside ethnic families in order to ensure that everyone becomes part of the same society.

Official documents in both Sweden and Denmark insist that, for the welfare state to protect itself, immigrants must abandon any beliefs and practices that violate the norms of Scandinavian solidarity. The objective of integration, the Swedish report holds, should be "equal rights and opportunities for all without regard to ethnic or cultural background." Echoing Marshall's ideas about class abatement, the report argues that the purpose of policy "should be to abate ethnic segregation." The Danish report goes even further. "Foreigners who are legally in the country must be introduced to Danish society in a fruitful fashion... society has to demand that foreigners act to become self-sufficient and integrated into Danish society by becoming conversant with its basic values. It is a truism that foreigners as well as Danes must obey the country's laws and rules."

Compared to the Parekh report, there is no emphasis in either document on group rights or the protection of group customs; it is individuals who have rights, not the ethnic communities themselves. Both, moreover, stress the reciprocal obligations of immigrants far more than the Parekh report.

"The Danish welfare state faces a dilemma," wrote law professor Stig Jorgensen in the newspaper Information. "On the one hand, foreigners receive more from the state than they contribute to the common insurance pool, but on the other hand, we cannot accept different treatment between them and Danes." The only solution seems to be to try to make foreigners more like Danes as quickly as possible. It is not enough, writes the director of the Federation of Danish Social Workers in the same newspaper, to insist that Amra from Iran learns Danish and throws away her chador. "If she is to become integrated as a citizen, employers have to employ her, her neighbours have to invite her in for coffee, landlords have to rent her a room..." To counter the politics of exclusion of right-wing parties, social democrats support a politics of inclusion. But what the Danes call integration, the Parekh report dismisses as assimilation.

The Scandinavian insistence on solidarity and integration carries a price-as it does elsewhere in Europe. Sweden and Denmark were once viewed as among the most tolerant places in the world. Yet the defence of the welfare state in the face of the new immigration has revealed an undercurrent of racism. Also, Swedish and Danish industrial policy, which favours high-paying jobs and strong unions, exacerbates the problem by confining migrants to the margins. So, 54 per cent of men and 67 per cent of women in Denmark from non-European societies are unemployed. Since 1985, 92 per cent of Turks living in Albertslund, a Copenhagen suburb, have married other Turks. In most cases, one partner either still lived in Turkey or had arrived in Denmark that year.

For all the talk of integration, Swedes and Danes-unlike the French and British with their colonial pasts-have little experience of mixing with people who are different from them. The Swedish report on diversity says that "in order for Swedish to work as a bridge between people, it is necessary to increase the tolerance Swedes have for those who do not speak the language perfectly or speak it with an accent," a formulation which suggests the existence of considerable intolerance in daily life. Integration is proclaimed-but nothing like it has been achieved.

the problems that the Scandinavian countries are experiencing highlight the difficulty of clinging to the old ideal of solidarity in a globalising world with high levels of immigration. At the same time the Parekh report, and the many others like it, go too far in the other direction-underestimating the extent to which social solidarity requires strong national cultures. Britain seems to be navigating this divide better than many. For example, it requires all immigrants to learn English but subsidises minority languages; it imposes a common core curriculum in schools, but still allows some to call themselves Islamic, Hindu or Jewish. And attitude surveys in Britain have been recording a steady decline in racial prejudice. According to Yasmin Alibhai-Brown's book True Colours, 74 per cent of whites (88 per cent of young whites) said they would not mind if a close relative married an Afro-Caribbean. (In the hugely successful television series Big Brother, two of the most popular characters were a lesbian and a black man.) But there is also a significant minority who do not embrace diversity. And working-class Britons are still far more likely to hold conservative views across the range of moral and social issues, including immigration. According to the latest British Social Attitudes Survey, 60 per cent of working-class people think that sex between two adults of the same sex is wrong, which compares with only 37 per cent of the salariat. There are similar moral gulfs between the generations. The same survey records that 72 per cent of 25 to 34 year olds believe it is "not wrong at all" if a man and woman have sex before marriage, while only 32 per cent of 55 to 64 year olds agree.

In the midst of this moral diversity, Britons still need a unifying idea of Britishness that can encompass diversity but is not eclipsed by it. As Melanie Phillips and others have argued, it is those at the bottom of the pile, economically and educationally, who suffer most from cultural rootlessness. Yet the Parekh report's rejection of the idea of a British way of life echoes Margaret Thatcher's famous remark that there is no such thing as society. Right-wing libertarians want business to do whatever it decides is in its best interest, irrespective of its impact on the common good, and left-wing multicultural libertarians want minorities to do whatever is in their interests, irrespective of its impact on the common good. Multiculturalists seem to believe that it is sufficient to base solidarity on abstract concepts of international law rather than real people with recognisable values and motives.

The story of modernity has been about people's ability to master forces once considered outside their control. Unwilling to grow old without some guarantee of economic security, or to face a labour market which could throw them out of work with no means of support, citizens of modern liberal democracies have pooled their exposure to risk in welfare states. In the absence of common threats it is the shared interests we have in the effectiveness of institutions like the NHS that binds us together most. And despite the limits of social democracy, the collapse of communism, and the ascendancy of Reaganism and Thatcherism in the 1980s and 1990s, there is no mass support for a politics which would return us to the kinds of lives people led before the 20th century.

Modern individuals want to feel some control over their moral environment for the same reasons they want control over their economic environment. Without broad common standards of right and wrong, some agreement on the nature of marriage and the family, respect for law, and some consensus about the role of religion in public life, our ability to live together happily is weakened. This does not mean that we must return to a time when racial and religious diversity was discouraged, and moral conformity trampled on individual freedom. It is one of the shining accomplishments of modernity that individuals have learned to share their fates with people very unlike themselves. No society in western Europe or North America is any longer homogeneous in ethnic, racial or religious terms, and all of them are richer for it. But the willingness to extend solidarity to those with whom one does not share such common bonds-or, indeed, attitudes towards pre-marital sex and homosexuality-depends on the mutual responsibilities imposed by citizenship.

The unavoidable negotiation between solidarity and diversity will be resolved in different societies in different ways, depending on their history and culture. Both principles must, of course, coexist, and there is no blueprint for dealing with the conflicts when they arise. But a few general points can be made. First, the conflict between solidarity and diversity must at least be more openly acknowledged, especially on the left, and political rhetoric and public policy should not shy away from pressing the claims of solidarity even sometimes against diversity. Second, the opinions and attitudes of the mainstream majority should not be ignored or treated with contempt. (They barely feature in the Parekh report except as representatives of official Britain.) It is true that they are not the people who are facing discrimination for being gay or black, but they are a necessary part of the solution. There is a wide spectrum of opinion in Britain, ranging from those who embrace difference to those who fear and hate it. Michael Ignatieff's useful distinction between positive and negative tolerance insists on a negative minimum of respect for the law and a public display of civilised behaviour, but also acknowledges that some people will not want to embrace difference and, within the law, should be free not to. Third, holding on to an idea of social solidarity through the welfare state in an increasingly diverse society requires conformity with some minimum, over-arching idea of Britishness-to which everyone signs up.

Immigrants come to societies like Britain for a reason. They expect to find ways of life, either economic or political, unavailable at home. How ironic it would be if, in the name of making them welcome, advocates of diversity were to undermine the conditions that made these economic and political miracles possible. Fortunately, that is unlikely to happen. Solidarity does not need diversity, although it can be strengthened and rejuvenated by it. But diversity does require solidarity because, without a common agreement on morality, no principle, including the principle of diversity itself, can ever be safe. n