On what principles should a government base its immigration policies? What makes a nation? To what extent do the inhabitants of a nation have the right not to be submerged? On what kind of identity should a state be founded? Unless we can answer these questions, British government policies will continue, as they were for most of the 20th century - since the Aliens Act of 1905, designed principally to keep out European Jews - to be based on the unworthy principle of pandering to prejudice and to be formed with the sole aim of gaining votes.
Let us start with the question of identity, to which there are many answers. Israel, for instance, identifies itself as a Jewish state, and on this ground operates the law of return, under which anyone who qualifies as a Jew is guaranteed admission and settlement. All people of demonstrably German ancestry, no matter how remote, such as those who emigrated to Russia generations ago, are assured of admission to the German homeland. Other states exclude those who do not share the identity it ascribes to itself: the white Australia policy used to refuse admission to anyone other than those of white European descent, while the constitution of Malawi denies citizenship to anyone not of black sub-Saharan race.
The identity of a state may also be founded upon a particular religion. This is true of those countries that designate themselves "Islamic republics", and it was true of almost all European countries during the Middle Ages and for some centuries afterwards: they proclaimed themselves to be Christian kingdoms - after the schism, to be Catholic or Orthodox, after the Reformation, to be Catholic, Orthodox or Protestant. They took for granted their right, when they wished, to expel Muslims or Jews. Modern Israel is a mixed case. Under the law of return, one must prove birth from a Jewish mother: the criterion is racial. Failure to practise the Jewish religion, or even overt renunciation of it, is no bar, but the criterion is nevertheless in part religious, because adherence to any other religion is held to invalidate the claim for admission.
Language may also be seen as essential to a state's identity. Mussolini endeavoured to suppress the use of French or German by the inhabitants of Italy, despite the numbers who spoke these as first languages. In our own day, Turkish governments have forbidden the use of the Kurdish language. In both cases, even school- children have been prohibited from using their mother tongues, not just in classrooms, but in playgrounds.
A self-governing nation needs an identity; and this identity will always be informed to an extent by its history. But what is "its" history? The way in which English history has been taught in England provides one answer to this. The teaching is traditionally imperialistic and triumphalist, but in one respect it has been admirable: it has been taught as the history of a land, not of a people. It used to begin with the Romans; after the Romans left, they were ignored in favour of the Britons left behind. The Angles, Jutes, Saxons, Danes and Normans all sprang into existence when they invaded the land; no one cared what they were doing before that. Few English schoolchildren ever grasp the extent of Canute's realm or even hear that other Normans conquered Sicily. This contrasts notably with the way history is taught in Turkey, where the primary emphasis is on the history of the Turkish people, from their origins in central Asia long before the conquest of Asia Minor.
But the identity of a state cannot be grounded solely in the territory over which its dominion extends. If it is not to be grounded in a common ethnicity, religion or language, it must be grounded in shared ideals, a shared vision of the society it is striving to create. There is no need to represent such ideals as having been handed down from the immemorial past: the ideals that unify France are those of 1789. And if ideals are not wholesome, they will sooner or later be overturned: neither the Italian fascist vision nor that of Nazi Germany could supply a sense of identity that would last for very long. Even if Germany had won the Second World War, the Nazi vision could not have endured because it was racist to its core, and could not be shared by the subjugated European peoples.
But in today's world, many nations were created, some quite recently, in response to a demand for self-determination or, like Israel and Pakistan, as havens for people who feared persecution. How could such states not define themselves by race or religion?
The truth within the principle of national self-determination is that everyone has the right to live in a country in which they and others of a group to which they belong are not persecuted, oppressed or discriminated against, in which their religion, language, race and culture are not reviled or held up to contempt and in which they can fully identify themselves with the state under whose sovereignty that country falls. This may be called the right to be a first-class citizen.
They and their group also have a right not to be submerged. This principle requires the greatest precision, since the standard complaint of racists and xenophobes who object to any level of immigration, however low, is that the country is being swamped; for this reason, I have deliberately chosen the word "submerge" in place of the more emotive "swamp", as once used by Margaret Thatcher.
We each need to be able to feel at home somewhere; those who do not are deprived of a natural human sentiment, and are usually conscious of their deprivation. That feeling of home may be attached to a country, a particular region, or just to a city, town or country district. But it is not just a matter of locality: one should also feel attachments to the institutions and the groups to which we are bound by common endeavours and concerns. Such attachments will usually lead someone to identify with the history of the locality or institution. Once, looking out over the sea at Syracuse, my wife asked a friend, a native of Syracuse, whether she was looking in the direction in which the Athenian fleet had arrived. "Yes," he said, "just over there. But we defeated them." He was not claiming to be descended from the ancient Greek inhabitants of Syracuse, any more than an English Catholic of Italian descent is denying his ancestry when he joins in with "Faith of our Fathers", singing, "Our fathers, chained in prisons dark". This is the "we" that may be used by a member of a college or a cricket team when recalling events in which it took part before he or any of his colleagues were members: it is the "we" of belonging.
Many people, probably most, have at least a dual identity: Catalan as well as Spanish, Welsh as well as British, Bengali as well as Indian, Syrian as well as Arab. People vary in which of their different identities is the more important to them. But in all these cases, the attachment is not only to a body of people with whom they share a language and a culture, but also to a land, the land where those people live. Not everyone enjoys this association of land and people; not everyone can say "The place where I was born and where I live is where my people belong", nor even "The place where I came from is where my people belong". But to those who do, it is a consolation. No one needs to be a fierce nationalist to feel happy that those with whom he or she most closely identifies have a place that is peculiarly theirs.
Moreover, cultures are fragile: they can be dissipated by the impact of other cultures. They vary in how robust they are, which is in part a matter of their prestige. But the culture of a people that is genuinely in danger of being submerged by an influx of people of different cultures, and particularly of people with especially robust cultures, is unlikely itself to be at all robust. That is why it is an injustice that immigration should be allowed to swell to a size that threatens the indigenous population with being submerged.
But is there any real danger of this happening in Britain? An immigrant presence will have only a faint, and usually beneficial, effect unless the number of immigrants is very large, or their culture powerfully dominant. Quite apart from the proportionately small numbers involved, the culture of West Indians is very similar to the British, being to a great extent derived from it; and neither it nor any of the cultures of the Indian subcontinent is in the least dominant. A far greater threat is posed by the culture of the United States, which really is dominant and which exerts a gravitational pull even from a distance. Hardly a household in Britain is not now pestered by the American custom of trick-or-treat on Hallowe'en. The "received" British dialect of English is ever more closely assimilated to the American dialect. Hardly anyone in Britain now says, for example: "It looks as if it were . . . " Everyone says: "It looks like it is . . . " - which is pure American. We have adopted the American usage of "corn" to mean maize and "a billion" to mean a thousand million. Every few months, a new Americanism takes root. A culture can be submerged without an immigrant presence.
Even if Britain were threatened with being submerged by its immigrants, it would be in no position to complain. In Malaya and Fiji, the British colonial authorities positively encouraged immigration that threatened to submerge - or swamp - the native population. In Malaya, the influx of Chinese to promote commerce, and of Indians to work the rubber plantations, came very close to reducing the Malays to a minority in their own land. In Fiji, the Indian immigration actually did make the Fijian population a minority for a period, with consequences still visible today.
It is worthwhile to discuss the concept of submergence and the right not to be submerged, because they are so often invoked by those engaged in propaganda against admitting small numbers of immigrants. In normal circumstances - in countries that are neither part of a colonial empire nor under the rule of oppressive invaders - there is no danger whatever that even a relatively high level of immigration will threaten the native culture or population. A vigorous culture will assimilate new features, to its own benefit, or ignore them if they cannot fruitfully be assimilated.
But what are the corresponding duties of immigrants? Must they assimilate? Most immigrants will have adapted to a large degree, adopting some of the indigenous customs and superimposing them on those they brought with them; but they will remain conscious of being members of a minority, and they will identify themselves in part with that minority as well as with the country in which they have chosen to live. Even if they have been driven from their own lands by starvation or persecution, they must at some stage make a decision to stay where they are.
The children of immigrants - given an acceptance of them by the surrounding society - will think somewhat differently. They will retain some of the customs of their parents, but will regard themselves as full members of the national community into which they were born: in their eyes, that national community now embraces their customs and their culture, as well as those traditional to that country. As time passes, assimilation may become complete: British descendants of Huguenot refugees may know of their ancestry, but are in every other respect indistinguishable from the rest of the population.
Or it may not become complete, in which case there will simply be a segment of the population whose customs differ in certain respects from those of the majority. Immigrants and their descendants have the problem of how to fuse distinct cultural traditions within their own lives, a problem that becomes less acute for each generation. It is a problem that only they can solve, and with which it is for them alone to concern themselves; the business of others is only to leave them free and without pressure to solve it as they choose.
The world at the turn of this century is one in which there has long been no possibility of crossing any but a very few frontiers unhindered, but in which travel is swifter and easier than ever before, and there are manifold calamities - persecution, violence, war, hunger - pressing people to flee the lands in which they are living. We can therefore say with assurance that, in the world as it now is, no state ought to take race, religion or language as essential to its identity. If it does, it will inevitably find living within its borders minorities not of the favoured race, practising religions other than the favoured one, speaking languages different from the majority tongue. These minorities will be liable to persecution or discrimination. Whether or not such discrimination is severe, members of these minorities will feel themselves, at best, to be "second-class citizens". However much they would like to do so, they will feel unable to identify themselves wholeheartedly with the country to which they belong: Christians will be constantly reminded that it is an Islamic state to which they owe allegiance, or Muslims will constantly have recalled to them "our Christian traditions". There is no country in today's world that does not have racial, religious or linguistic minorities; even if it lacked them, they would soon arrive.
Should countries such as Israel and Pakistan, founded as refuges from persecution, be exceptions to this principle? No. Their means of declaring their national identities must indeed be especially delicate. They have the right to include in them their role as a refuge for those of a particular people or faith, but they must not make membership of that people or adherence to that faith a part of what it is to belong to the nation. It is the destiny of such a nation to provide a haven for those subject, or potentially subject, to persecution on particular grounds; but it is also the destiny of that nation to create a unity from the disparate inhabitants entitled to live in that country and be its citizens, a unity founded upon common ideals of justice.
Sir Michael Dummett is the former Wykeham Professor of Logic at Oxford University. This essay is extracted from his On Immigration and Refugees, to be published by Routledge on 29 May (£7.99)