Few concepts in modern times have been less understood and few more liable to misuse than the concept of race when applied to humankind.
Such powerful feelings has it aroused that its use is sharply declining among the writers of physical anthropology textbooks in the USA. Of twenty such textbooks published between 1932 and 1969, thirteen (65 per cent) accepted that races of humans exist, three (15 per cent) claimed that they do not exist, while of the remaining four, two did not mention the subject while two stated that there was no consensus on the subject. Of thirty-eight such textbooks that appeared between 1970 and 1979, only twelve (32 per cent) stated that races of humans exist, whereas fourteen (37 per cent) claimed that races do not exist; of the remaining twelve texts, four were non-committal on the matter, three failed to mention race and five indicated that there was no consensus (Littlefield et al. 1982).
It is of course a moot point how much we may conclude from a study of the contents of textbooks, but it is, to say the least, striking that, during the 1970s, there was in the USA so marked a swing away from the earlier widespread acceptance of the existence of human races. Critics of the study cited have raised the question of the degree to which that change reflects new concepts flowing from new data and novel approaches, and the extent to which the change might have been predicated upon extraneous factors, such as a swing of fashion, political considerations, or the composition of classes of students to which the texts were directed. Nor is it clear whether the tendency in the USA typifies other parts of the world of physical anthropology.
Certainly the change tells us that, even among experts, no less than in the public mind, the concept of race is being critically re-examined and that no consensus, let alone unanimity, among specialists on the validity or the usefulness of the race concept appears to exist at the present time. It is worthwhile therefore to examine the meaning of race. Since race is basically a concept of biology in general, we shall start by examining race as a biological notion.
Race as a biological concept
Many, perhaps most, species of living things comprise numbers of populations which may be dispersed geographically and among varying ecological niches. To impart order to the subdivisions within a species, biologists have used several terms such as subspecies, race and population to classify the various groupings of communities that make up a species. Thus, in a species within which two or more subspecies are recognized, a race comprises populations or aggregates of populations within each formally designated subspecies. Often the term race is qualified: biologists recognize ‘geographic races’ (which may be synonymous with subspecies); ‘ecological races’ where, within a species, there occur ecologically differentiated populations; and ‘microgeographic races’ which refers to local populations (Mayr 1963).
Although students of any group of living things may differ from one another on the finer details of such intraspecific classifications, there has for some time been fairly general agreement that race is a valid biological concept. Classically, the differences among the races in a species have been identified by their morphology, that is, their observable physical structure.
Since the mid-1930s, and especially since 1950, biologists, not content with studying the morphological make-up of populations within species, have been studying the genetic composition of the subdivisions within species. These studies have directed attention to a number of non-morphological traits such as the genes for blood-groups and for specific proteins. When these hereditary characters are analysed, they reveal that there are no hard and fast boundaries between races and populations within a species. For any such genetic marker, it is not uncommon to find that the frequency of the trait in question is distributed along a gradient (or cline) which cuts across the boundaries of races, as delimited by morphology. Such gene clines often do not parallel any detectable environmental gradient; they appear to be neutral in relation to natural selective agencies in the environment.
Different genetic markers within a species may vary along different gradients. Thus, if one were to base one’s thinking about the subdivisions of a species on the distribution of any one genetic marker, one would be liable to reach a different conclusion from that which might flow from the use of another genetic marker.
Hence, newer methods of analysis combine the frequencies of many different genetic markers, in the hope that the resulting sorting of populations will more nearly reflect the objective genetic relationship of the subgroups within a species.
Character-gradients apply as well to some morphological features. That is, some structural features such as body size, ear size, or colouring, change gradually and continuously over large areas. Such gradients, unlike the genetic clines, appear to parallel gradients in environmental features and have probably resulted from the action of natural selection (Huxley 1963). However, the frequencies of the genes governing morphological characters are less commonly used in the study of genetic interrelationships within a species, for several good reasons: first, such traits are often of complex, difficult and even uncertain genetic causation; second, many of them and, particularly, measurable characters are determined not by a single gene-pair, but by numbers of different gene-pairs; third, such characters are especially subject to environmental modification: for example, if animals live in a lush area and eat more food, they would be expected to grow bigger than those of the same species living in a more arid region. This ‘eco-sensitivity’ of the body’s metrical traits renders them less useful in an analysis of genetic affinities.
In sum, race is a biological concept. Races are recognized by a combination of geographic, ecological and morphological factors and, since the 1970s, by analyses of the distribution of gene frequencies for numbers of essentially non-morphological, biochemical components. As long as one focused on morphological traits alone, it was sometimes not difficult to convince oneself that races were distinctly differentiated, one from another, with clear-cut boundaries between them; the progressive application of genetic insights and analyses to the problem revealed that recognizable gene variants (or alleles) are no respecters of such hypothetical boundaries. Often, indeed, one race merges with the next through intermediate forms, while members of one race can and do interbreed with members of other races. Hence, the importation of genetic appraisal into the discussions on race has led to a definite blurring of the outlines of each race, and so to an attenuation of the concept of race itself.
Race in human biology
The biological concept of race, as just sketched, has been applied to the living populations of the human species. At least since the time of the Swedish naturalist and systematist Linnaeus (1707–78), all living human beings have been formally classified as members of a single species, Homo sapiens. The accumulation since the middle of the nineteenth century of fossil remains of the human family has revealed that earlier forms lived which could validly be regarded as different species of the same genus, for example Homo habilis and Homo erectus. Our species, Homo sapiens, probably made its appearance between one-half and one-third of a million years before the present.
As Homo sapiens spread across first the Old World and, more latterly, the New World, the species diversified, in varied geographical zones and ecological niches, into numerous populations. At the present time we have a situation in which living humanity is divided into several major and many minor subdivisions among which the same kinds of variation are encountered as apply to other living things. Thus, the populations show morphological variation, including some gradients associated with environmental gradients, and varying gene frequencies with clines of distribution that, for individual genetic markers, breach the limits of morphologically defined groups of populations.
Physical anthropologists, relying on morphological traits, have for long divided living humankind into great geographical races (also called major races, subspecies and constellations of races). Most classifications recognized three such major subdivisions, the Negroid, Mongoloid and Caucasoid; some investigators designated other major races, such as the Amerind and the Oceanian. Within the major races, several dozen minor races (or, simply, races) were recognized, the number identified varying with the investigator. As with other living groups, historically the classification of living Homo sapiens was based on morphological traits, such as skin colour, hair form and body size. As genetic analysis came to be applied, in respect first of blood-groups and later of a variety of proteins, clines were found which cut across the boundaries of minor and even of major races. Moreover, it was found that the genie variation between the major races was small in comparison with the intraracial variation. Doubts began to be expressed as to whether there was any biological basis for the classification of human races (for example, Lewontin 1972).
The problem is compounded by the fact that, even when genetical analysis became based not just on a few traits such as the ABO, MN and Rh blood-groups, but on a number of traits, different results were obtained according to which combinations and numbers of traits were used. For example, Piazza et al. (1975) analysed frequency data for eighteen gene loci in fifteen representative human populations: they found that the Negroid populations were genetically closer to the Caucasoid populations than either group of populations was to those populations classified as Mongoloid. This, in turn, was interpreted as signifying an earlier phylogenetic split between Mongoloid, on the one hand, and Negroid-Caucasoid on the other, and a later (more recent) split between Negroid and Caucasoid.
However, Nei’s (1978) analysis, based on eleven protein and eleven blood-group loci in twelve human populations, revealed a first splitting between Negroid and Caucasoid-Mongoloid. Subsequently, Nei (1982) and Nei and Roychoudhury (1982) used a still larger number of genetic traits, namely sixty-two protein loci and twenty-three blood-group loci, that is eighty-five gene loci in all, for which data were available for some eighteen world populations. Interestingly, while the protein data revealed a first splitting between Negroid and Caucasoid-Mongoloid, the blood-group data suggest a slightly closer affinity and therefore a slightly more recent splitting between Negroid and Caucasoid.
Clearly, the last word has not been said on the exact pattern of affinities among the living races. Nor is there a consensus as to whether the large size of intraracial genetic variation, compared with interracial, vitiates any biological basis for the classification of human races. As two representative studies, we may cite Lewontin (1972) who believes there is no basis; and Nei and Roychoudhury (1982) who disagree with Lewontin and assert that, while the interracial genie variation is small, the genetic differentiation is real and generally statistically highly significant. Furthermore, it is clear that, by the use of genetic distance estimates, Piazza et al. (1975), Nei and Roychoudhury (1982) and others have been enabled to study the genetic relationships among the mainly morphologically defined human races, to construct dendrograms and to impart some understanding of the pattern of recent human evolution. Thus, the latter investigators have found evidence from protein loci to suggest that the Negroid and the Caucasoid-Mongoloid groups diverged from each other about 110,000 ± 34,000 years before present, whereas the Caucasoid and Mongoloid groups diverged at about 41,000 ± 15,000 years before present. These estimates do depend on a number of assumptions and may be modified with the accretion of more data.
One further point may be mentioned here: the extent of genetic differentiation among the living human races, as determined by the study of protein loci, is not always closely correlated with the degree of morphological differentiation. Indeed, evolutionary change in morphological characters appears to be governed by quite different factors from those governing genetic differentiation in protein-forming genes of the human races, on presently available evidence. Genetic differentiation at protein loci seems to occur largely by such biological processes as mutation, genetic drift and isolation, with migration playing an important role in the establishment of current genetic relationships among human races. However, morphological characters have apparently been subject to stronger natural selection than ‘average protein loci’ (Nei and Roychoudhury 1972; 1982).
In short, the race concept can be applied to modern humans, even when one uses the most modern analytical procedures of population geneticists, and such application has been found of heuristic value. Nevertheless, irrespective of sociopolitical considerations, a number of modern investigators of human intraspecific variation find it more useful and more valid to base such studies on populations, as the unit of analysis, and to discard the race concept in these attempts.
Abuses and aberrations of the race concept
Among the various misconceptions that surround the concept of race are ideas about ‘race purity’, the effects of racial hybridization, superior and inferior races, race and mental differences, race and culture. A full review of this vast subject is not possible here: it has been dealt with in a number of studies such as those of Tobias (1970), Montagu (1972), Mead et al. (1968), Kagan (1968), Jensen (1969), Bodmer and Cavalli-Sforza (1970), Scarr-Salapatek (1971), Lochlin et al. (1975), Scarr (1980) and Gould (1981).
Although the foregoing selection of writers adopt widely differing standpoints, especially on the subject of race and intelligence (as supposedly reflected by IQ test results), it would not be unfair to claim that the following reflect the view of a great majority of physical anthropologists, human biologists and human geneticists at this time.
1 Race is an idea borrowed from biology.
2 At a stage when the study of human populations was primarily, if not exclusively, morphological and its objective classificatory, the race concept helped to classify the immense variety of living and earlier human beings of the species Homo sapiens. With the advent of genetic analysis and the discovery that clines of genetic differentiation transcend the supposed boundaries of human races, the race concept has been appreciably weakened.
3 While some population geneticists have found that race still serves a useful purpose in the study of the genetic affinities of living populations, in the determination of the causal factors that have operated to produce genetic differentiation and in the reconstruction of the phylogenetic history of modern human diversity, others have found the concept of such negligible value in these studies as to have led them to discard race entirely. Time will tell whether we are witnessing ‘the potential demise of a concept in physical anthropology’ (as Littlefield et al. 1982 speculated), or whether the concept will survive the politico-social abuses to which it has been subject and which have been regarded by some as the primary cause of its decline from favour among many investigators and writers of textbooks.
4 If, for purposes of this analysis, we accept the existence of human races (as of other living things), we .must note that races differ not in absolutes, but in the frequency with which different morphological and genetic traits occur in different populations.
5 The overwhelming majority of the genes of Homo sapiens are shared by all the mankind; a relatively small percentage is believed to control those features which differentiate the races from one another.
6 The formation of the modern human races is a relatively recent process, extending back in time for probably not much more than 100,000 years. As against this period of recent diversification, at least forty times as long a period of its hominid ancestry has been spent by each race in common with all other races, as it has spent on its own pathway of differentiation. This statement is based on the evidence that fossilized members of the human family (the Hominidae) are known from 4 million years before the present; molecular and some other evidence suggests that the appearance of the hominids may go back to 5 or more million years before the present.
7 Racially discriminatory practices make certain assumptions about race, some overt, some tacit. These include the assumptions that races are pure and distinct entities; that all members of a race look alike and think alike, which assumption, in turn, is based upon the idea that how one behaves depends entirely or mainly on one’s genes; and that some races are better than others.
8 The scientific study of human populations has provided no evidence to validate any one of these assumptions.
9 Genetical and morphological analysis of human populations has failed to confirm that some races are superior and others inferior.
10 Accidents of geography and history, difficulties of terrain, physical environment and communication, are sufficient to account for the contributions which different populations have made to the varying advancement of human culture and to civilization.
11 Culture, language and outlook are not inseparably bound up with particular morphological or genetic racial features; for example, human culture is altering the direction of human evolution, as the species spreads into every corner of the world, and as cultural and racial divergence gives way over large areas to cultural and racial convergence.
12 The myth of the pure race has been thoroughly disproved. There are no pure (genetically or morphologically homogeneous) human races and, as far as the fossil record goes, there never have been.
13 Not only is purity of race a non-existent fantasy, but there is no evidence to support the notion that purity of race is a desirable thing.
14 Racial groups are highly variable entities; for many traits intraracial variability is greater than interracial variability. Intermediates exist between one race and the next.
15 Members of all races are capable of interbreeding with members of all others, that is, all that have been put to the test.
16 The supposed evils attendant upon race-crossing do not bear scientific scrutiny: neither sterility, diminished fertility, nor physical deterioration, has been proven to be a biological consequence of race-mixing. If there are unfortunate effects from such crossing, they are social (not biological) and they appear to result from the way in which other members of the populations in questions look at and treat the ‘hybrids’.
17 The study of the races of humankind has been based on physical (that is morphological, physiological and biochemical) and genetic traits; mental characteristics have not been used in the classification of the human races, nor have they been found useful for such a purpose.
18 Scientific studies have not validly demonstrated any genetically determined variations in the kinds of nervous systems possessed by members of different human races, nor any genetically determined differences in the patterns of behaviour evinced by members of different races.
19 The claim that genetic factors contribute as much as 75 or 80 per cent of the variance of IQ test-score results and are therefore largely responsible for Black-White differences in mean test-score results has been seriously questioned in a number of investigations. It has been shown that a heritability estimate of 0.75 does not apply to American Blacks, among whom a much smaller percentage of the variance of test-score results has been shown to be genetically determined, and a larger proportion environmentally determined. The immense literature that has accumulated since Jensen (1969) put forward his hypothesis that American Blacks are genetically inferior in intelligence to Whites has revealed many flaws that were implicit in the reasoning behind the hypothesis. The main conclusion that many of these studies have reached is that ‘currently available data are inadequate to resolve this question in either direction’ (Bodmer and Cavalli-Sforza 1970). A number of investigations have led to the development of environmental hypotheses. For example, Scarr (1980) found evidence in her studies to support a twofold hypothesis: such differences as exist between comparable populations she attributed partly to environmental factors and partly to cultural factors. On this additional cultural hypothesis, her work led her to stress a different relevance of extra-scholastic or home experience to scholastic aptitudes and achievement: ‘The transfer of training from home to school performance is probably less direct for Black children than for White children’ (Scarr-Salapatek 1971). Clearly, at this stage of our ignorance, it is unjustified to include intelligence, however tested, among the validly demonstrated, genetically determined differences among the races of humankind.
Phillip V.Tobias
University of the Witwatersrand
References
Bodmer, W.F. and Cavalli-Sforza, L.L. (1970) ‘Intelligence and race’, Scientific American 223.
Gould, S.J. (1981) The Mismeasure of Man, New York.
Huxley, J.S. (1963) Evolution: The Modern Synthesis, 2nd edn, London.
Jensen, A.R. (1969) ‘How much can we boost IQ and scholastic achievement?’, Harvard Educational Review 39.
Kagan,J. (1968) ‘On cultural deprivation’, in D.C.Glass (ed.) Environmental Influences: Proceedings of the Conference, New York.
Lewontin, R.C. (1972) ‘The apportionment of human diversity’, Evolutionary Biology 6.
Littlefield, A., Lieberman, L. and Reynolds, L.T. (1982) ‘Redefining race: the potential demise of a concept in physical anthropology’, Current Anthropology 23.
Lochlin, J.C., Lindzey, G. and Spuhler, J.N. (1975) Race Differences in Intelligence, San Francisco, CA.
Mayr, E. (1963) Animal Species and Evolution, London.
Mead, M., Dobzhansky, T., Tobach, E. and Light, R.E. (1968) Science and the Concept of Race, New York.
Montagu, A. (1972) Statement on Race, 3rd edn, London.
Nei, M. (1978) ‘The theory of genetic distance and evolution of human races’, Japanese Journal of Human Genetics 23.
——(1982) ‘Evolution of human races at the gene level’, Human Genetics, Part A: The Unfolding Genome, New York.
Nei, M. and Roychoudhury, A.K. (1972) ‘Gene differences between Caucasian, Negro and Japanese populations’, Science 117.
——(1982) ‘Genetic relationship and evolution of human races’, in M.K. Hecht, B.Wallace and C.T.Prance (eds) Evolutionary Biology 14.
Piazza, A., Sgaramella-Zonta, L. and Cavalli-Sforza, L.L. (1975) ‘The fifth histocompatibility workshop: gene frequency data: a phylogenetic analysis’, Tissue Antigens 5.
Scarr, S. (1980) Race, Social Class and Individual Differences, Hillsdale, NJ.
Scarr-Salapatek, S. (1971) ‘Race, social class and IQ’, Science 174.
Tobias, P.V. (1970) ‘Brain size, grey matter and race—fact or fiction?’, American Journal of Physical Anthropology ns 32.
This is the complete article, containing 3,641 words
(approx. 12 pages at 300 words per page).
