Defining the ethnic group: important and impossible
The start to this chapter is deliberately personal to highlight some of the complexity of the topic in situations where political sensitivity may be less striking. W hen I was arranging research interviews of U nited Kingdom m igrants in the western suburbs of Sydney, an Australian teenager with a sm art coastal address said to me ‘Oh, among the ethnics!’ Is ethnicity, then, som ething only relating to others? Chapm an (1993) provides a short history of derivation and use of the word. Am I not ethnic too? I confess to some discomfort that I was not allowed a vote in the recent Scottish referendum on devolution, while my brother-in-law with impeccably English ancestry and culture was able to vote simply because of his address. Even more ironically that address is in the village of my paternal grandfather and his crofter antecedents. My other grandparents were also Scottish born and I was feeling some sense of Scottish ethnicity although born and living in England. W hat m atters, then, this self-definition or the geography of today’s electoral registers? I have even been called a ‘bloody C elt’ by an academic supervisor who felt I was short on Lancastrian discipline - the point may have been valid, but the stereotyping within the phrasing doubtful. The self-definition and the supervisor’s description clearly overlap despite my accent and current address. Is this overlap in anyway due to my physical appearance or ju st my surname? Is the residence of myself or of my ancestors more significant, and what of the certain mobility of some of these ancestors?If you wished to classify my ethnic group for medical purposes you might be well advised to consider my Scottish ancestry and pigm entation, as well as the culinary habits of my m other and her mother, as these might be more significant than my curren t address. However, even leaving aside more complicated transatlantic aspects of my personal history, my ethnicity is even harder to describe, for although I may be Scottish in England, I accept that I am considered very English when in Scotland. W hat is more, in both Scotland and England, lifestyle and linguistic differences between different groups, geographic and socio-economic, are many. Even in 1701, Defoe had explained how mixed is the ancestry o f ‘The True-born Englishman’ (Defoe 1701). So, just as my own ethnicity is hard to classify, other individuals have complicated ethnicity and all ethnic groups are similarly hard to define. Exceptional groups
may exist with some truly isolating features, but this is unlikely as it presumes a social homogeneity within that isolated group. Furtherm ore, ethnicity only becomes an issue where cultural groups are in some contact with each other. Any contact is likely to involve some individual mobility and some exchange of ideas and genes.In the 1970s, while working on research concerned with residents of neighbouring Oxfordshire villages anecdotally referred to as ‘the Otm oor population’, I began wrestling with the concept of ‘population’ (M acbeth 1985). Steeped in the contem porary debates about village endogamy (Cavalli-Sforza and Bodmer 1971) in Palma villages or village exogamy and consequent gene flow, as shown in the m arriage registers of these Oxfordshire villages (Boyce, K iichem ann and H arrison 1968), I could not accept the phrase ‘O tm oor population’ (M acbeth 1985). It was not until the late 1980s that Gom ila’s (1976) paper ‘Definir la population’ was drawn to my attention . At about the same time I read A rdener’s (1972) contribution to the concept of ‘population’. From this background my ideas on ‘population’ and so on ‘ethnic group’ developed and were discussed (M acbeth 1993). While this chapter reechoes some of the points made then, they are expanded with reference to the topic of this volume.As m ight be expected of this topic, a large social anthropological and sociological litera ture about ethnicity exists (e.g. B arth 1969, Banks 1996, Chapm an 1993, Cohen 1985, Tonkin et al. 1989), but it rem ains an elusive concept, in teracting intim ately with concepts of race and racism (e.g. Bhopal, this volume, Jenkins 1997) and with nationhood, nationality and nationalism (e.g. E riksen 1993, L lobera 1989, W erbner 1996). W hatever it is, it is m aintained through culture and socialisation and it affects health-related factors in many ways. O f prim e im portance are m arriage preferences or obligations. While these are much discussed by social scientists, it has been left to the geneticists (e.g. Barbujani and Sokal 1990, Bittles and Neel 1994, Modell 1997) to link these patterns to distributions of gene frequencies. As people find m ates w ithin their cultural group, genes cluster and are passed on predom inantly within that cultural group making ethnicity an independent social variable. In this way the clustering of the genes is not so much the cause of ethnicity as the outcome, the dependent variable. It is also true that situations do exist where characteristics under partial or total genetic control may be used for social classification and this can lead to a two-way process of classification and gene clustering, via stereotyping.T his early re fe ren ce to g en etic c lu s te rin g does not im ply genetic determ inism of everything biological. W ith cultural control of so many aspects of hum an lifestyle, ethnicity is involved in many non-genetic factors affecting biology since concep tion . E thn ic ity , for exam p le , is involved in the development of food preferences (e.g. Rozin et al. 1986, M acbeth 1997), the cultural influences on food choices (e.g. Bourdieu 1979, MacClancy 1992, Messer 1984), the m arketing and even the availability of foods, the eating patterns and so the content and tim ing of food intake, and, of course, the
cuisine (MacClancy 1992). Cultures, including diets, however, have features which change and others which seem to w ithstand change. Frequently there is a recognition of change in fam iliar cultures but a belief that ‘o ther’ cultures endure unchanged. The la tte r is probably an inaccurate description of the daily activities or diet of any group. Furtherm ore, ‘ethnic’ foods need not be regularly consumed: neither haggis nor paella is consumed often by those for whom each is considered ethnic. The recent growth in the m arket of British cookbooks assigned to different ethnicities and localities provides delicious entertainm ent and reinforces stereotypes, but may poorly reflect reality in the daily lives of those presum ed to eat these dishes (e.g. M acbeth 1998).Ethnicity and culture also affect the use and abuse of other consumptions, the extent of family support or independence, male and female activities (e.g. M acCormack and S trathern 1980), exercise (e.g. MacClancy 1996) and, within the framework of society at large, even work and study routines and ethics. The role of religion and beliefs must not be forgotten, especially where these are expressly concerned with health and health care. Clearly the ways in w hich cu ltu ra l p re fe ren ces d e te rm in e hum an biology are m yriad. Furtherm ore, what might be considered ethnic tends to be inextricable from what is considered socio-economic in the life experiences of the so-called ‘minority groups’ in W estern recipient countries (see Nazroo and Davey Smith, this volume).All this does not exclude the significance of locality, the rest of the environment, both physical and social. The list can continue for pages, but the significant aspect is that despite some sophisticated statistical attem pts there is no way to disentangle all these factors and their in teracting effects on our m ultifactorial characteristics. It is not ju st tha t the genetic and the non-genetic in teract, but also the ethnic and the economic, the locality and the ancestry, etc. All can be shown to affect biology.It would be superfluous in this volume to review in detail recent medical literature about ethnic differences, but readers will be aware of the increase in reference to ethnicity in the last quarter century. There is, however, an aspect of th is l i te ra tu re w hich i ts e lf deserves m ore a tte n tio n - the categorisation of the people studied. Table 2.1 provides a few examples of c lassifica tions used. W hile ‘W hite , B lack and A sian’, (or equ iv alen t paraphrases) are the most common in the U nited Kingdom, ‘W hite, Black and H ispan ic’ are m ore com m on in A m erican artic les. If ‘A sians’ are m entioned in the U nited States, Chinese and japanese are probably referred to, while in the U nited Kingdom those from the Indian subcontinent are being considered. In the medical papers of other nations there are other divisions, but most are rem iniscent of outdated and discredited concepts of biologically discrete racial groups. Take, for exam ple, S te e r s al (1995), who refer to ‘W hite, Indo-Pakistani, Afro-Caribbean, Black African, M editerranean, O rien tal and o thers’. If one unites the African categories and replaces M ed iterranean w ith A m erican Indians, the e igh teen th -cen tu ry ideas of Blumenbach (Ju rm ain ^ a /. 1999) are clearly reflected.