This paper considers three main issues. First, it deals with the underlying structural causes of international migration, especially labour migration, and the way they are liable to create ethnically distinct sub-populations. In achieving this it presents a typology of forms and characteristics of ethnic migration streams. Secondly, and by way of contrast, the case is argued for the need to consider ethnic migration from the behavioural perspective. Thirdly, reasons are described and suggested for the position of relative disadvantage in which many migrant ethnic minorities find themselves placed.

The main aim of this paper (see Author's note) is to demonstrate that definitions of which groups of people constitute indigenous minorities are subject to change—changes which are often due not so much to our increasing scientific or academic knowledge about particular groups of people, but to political factors. A secondary aim is to document two cases which have so far found little or no mention in the literature, one drawn from Zimbabwe and the other from Australia.

Cultural—‘ethnic’—differences within national populations have recently become the focus of considerable academic and political debate. This paper considers their demographic significance and reviews some facts and theories on the demography of minorities.

It cannot present a comprehensive theory of the demography of ethnic minorities. Such a theory must be part of a more general analysis of cultural effects on demographic variables—which so far is lacking in demography. Neither can it review exhaustively the facts of minority demography; they are not collected or made available in sufficient detail from many countries with minority or multiple nationalities. The aims of the paper are accordingly less ambitious: to consider the political and academic import of minority demographic differences, formulate questions that might usefully be asked about them, and present a few examples in detail.

The title of this paper suggests at least two possible points of departure. First, one can argue that spatial patterns are themselves an important feature in the study of social life which deserves our attention: that there is a significant geographical dimension to social interaction. Secondly, the title also implies that various social groups interact differentially with one another: that interaction is itself a proper focus for social investigation. Both points will be developed in the present paper with respect to ethnic minorities in Britain and the United States.

This paper is based on the author's experiences during first, participation in an anthropological study of migrants, particularly Samoan migrants, in New Zealand over a period of 10 years in the late 1960s–mid- 1970s (Pitt & MacPherson, 1975) and secondly, participation over the last 3 years in a World Health Organization programme concerned with promoting the health and welfare of children, many of whom were migrants living in poverty.

John Stuart Mill and Sir Charles Wood are the British thinkers to whom I wish to pay special tribute in this paper. I must begin, however, by recalling that in London 30 years ago, I was accused of propounding a ‘natural law of white minorities’ and of being unduly deterministic in my prediction that the then still proposed Federation of Rhodesia and Nyasaland, to be entrusted mainly to a white minority, could not succeed. It availed little to reply that my hypothesis of ‘Imperial inertia’, as I called it, and my forecast for the Federation, were made in the hope that the Federal constitution and its mode of implementation might be modified before it was imposed in 1953 by the government of the United Kingdom. I also urged that the key concepts ‘partnership’ and ‘trusteeship’ should be closely defined—it will be remembered that the Federation was formally dedicated to interracial partnership—and that careful regard should be given to what I termed ‘external partnership’, ‘internal partnership’, ‘external trusteeship’ and ‘internal trusteeship’, and to the important, dynamic relations between these concepts and policies, relationships which were scarcely recognized.

The concept of ‘need’ is currently popular in education. It is, as Dearden (1972) has pointed out, an attractive concept for educationalists, because it appears to offer an escape from arguments about values and about power. Those who uncritically accept the doctrine of needs can assume that there are educational experts who can unproblematically decide on needs, which can then be unproblematically catered for. Thus, when ethnic minority children began to arrive in British schools in significant numbers in the early 1960s, it was assumed that the schools—without experience or understanding of other cultures—could make an assessment of the needs of these children.

Rates of social mobility are some of the most revealing indicators of a society's character. In any society there is a tendency for privileged groups to try to close ranks, preserve their advantages and pass them on to their children, but the arguments of social justice, economic efficiency and social stability all suggest that these tendencies towards closure and inherited privilege should be curbed. In western liberal societies the dominant version of social justice is perhaps the one that advocates equality of opportunity: those with equal effort and talent should be equally rewarded whatever their class, sex, age, race or religion. Economic efficiency likewise requires that workers should be paid according to their actual productivity rather than to ascribed (and economically irrelevant) characteristics such as their race or sex. And theories of social order hold that the blocking of legitimate aspirations for social and economic advancement will be a potent source of social dissent and conflict (see Goldthorpe, 1980; Heath, 1981). On all three counts the opportunities open to members of ethnic minorities are of interest, but sociologically the third has particular interest for whereas within native white society failure to secure upward social mobility may lead to individual responses—to personal frustration and stress—an ethnic minority which experiences discrimination and blocked careers as a shared grievance is to that extent more likely to respond with collective protest and militant group action.

It may seem unexpected to find a geographer writing on the topic of ethnic segregation and ethnic intermarriage for a symposium of the Biosocial Society. Social geographers have, in fact, been actively working on ecological models of urban structure since the 1960s, but the development of this field dates back to the Chicago School of Sociology in the 1920s and 1930s. Geographers have become interested in both the spatial organization of society and in the social organization of space. Geographers have measured the ways in which social attitudes of groups manifest themselves in their residential choices of neighbourhood: groups which perceive themselves as socially distant from others are also spatially segregated from such groups. Residential location in neighbourhoods also affects the interactions between individuals in a fairly mechanical way. This field of social geography overlaps the spatial or ecological side of sociology so that practitioners of both schools often have more in common with each other than they have with others working in the fields from which they have evolved.

Broadly, genetic assimilation may be taken to mean the process by which groups acquire similar genes to one another. In the case of ethnic minorities, this usually involves a minority group and the surrounding majority population. Our knowledge of the genetics of human populations is imperfect and we may declare that only a small fraction of the total set of genes we possess may be identified. These genes may be divided into three subsets:

(1) genes which are known and have known effects upon characteristics of those who possess them;

(2) genes which are known and may be identified but have no known effects;

(3) genes which are unknown, presumed to be awaiting discovery.