Racial Differentiation

In many cases, biologists over the years have found it useful to divide a species into two or more subspecies, or races. The criteria for the definition of races - based on geographic distribution and various features of the body - yield classifications similar to those obtained using genetic markers. Use of genetic markers also shows very clearly that there are no “pure” races. Races are, in fact, generally very far from pure and, as a result, any classification of races is arbitrary, imperfect, and difficult. Yet anyone can see that there are certain relatively clear differences between a typical Caucasoid and a typical Mongoloid or a typical Negroid. What are the causes of racial differentiation? Which differences are cultural and which are biological?

The Human Species

A species is a set of individuals who can interbreed and have fertile progeny.

When Carolus Linnaeus originally defined the concept of a species in 1735, he believed that each species had been separately created by God and had remained unchanged ever since. Therefore he viewed the species as a natural, unchanging precisely definable unit. Already in that century it was clear that the classification into species should bring together individuals capable of interbreeding with each other, but incapable of interbreeding with members of any other species. As taxonomists worked at the task of classifying all known organisms into species, they soon discovered that nature could not be fit perfectly into such rigid pigeonholes.  In a classic work on the nature of species written in 1942 the zoologist Ernst Mayr put it this way: “no system of nomenclature and no hierarchy of systematic categories is able to represent adequately the complicated set of interrelationships and divergences found in nature.”  It is now clear that existing species have changed and that new species have been formed continuously throughout the history of life on earth. Therefore, there exist many situations that are intermediate between the ideal cases of a single species or of two distinct species.

 The members of a species have, potentially at least, a common gene pool. The formation of a new species is a long and complex process. A very important factor in this process of speciation is isolation between two or more populations of the original species. Isolation creates two distinct gene pools, in which evolution may follow different patterns. Isolation may have many origins, the most important one being geographical. Geographical barriers, especially mountains and water, tend to reduce migration and so to create isolation.

 In the absence of intermigration, genetic differentiation between populations may be expected as a result of  (1) selective adaptation and (2) genetic drift. For two isolated populations, initially identical in genetic composition, it would not be surprising if selective adaptation leads to two different patterns of genetic change. This would occur if the two populations occupy differing habitats, so that they are adapting to different environmental conditions—thus producing different selective advantages for various traits in the two populations. Even if the two populations occupy similar habitats, adaptation might occur through different mechanisms in the two populations. Finally, genetic drift and random accumulations of different mutations might be expected to lead to differing evolutionary patterns in the two isolated populations.

 Any gene flow between incompletely isolated populations will tend to reduce or prevent differentiation. Given sufficient isolation, however, differences between the populations will accumulate with time, and the genetic gap between the populations will gradually increase. Formally, the biologist tends to classify the two populations as different species when the capacity to interbreed has effectively disappeared. The effective incapacity to interbreed may be a secondary (and usually late) result of the differences accumulated in this process of genetic differentiation. It is more likely to occur, the longer the separation and the more significant the biological hiatus that has been created. Thus, the formation of different species does not (necessarily at least) occur abruptly. In zoological examples one finds that interfertility often is reduced gradually as the two populations differentiate before it eventually disappears completely.

Because the process of differentiation that eventually leads to different species is usually a continuing one, the differences between isolated groups may be trivial or difficult to detect in the initial phases of the process. In such a case, taxonomists are likely to group these populations into a single classification. At the stage when some differences between the isolated populations have become detectable, the taxonomist is likely to classify the recognizably different populations as different “races” (sometimes called “geographic races” when they are geographically isolated) of the single species. When the process is more advanced and the differentiation between the two populations is quite noticeable, the taxonomist is likely to classify the populations as “subspecies” of the single species.  (However, some taxonomists regard the terms “race” and “subspecies” as equivalent.) In any case, the decision about whether to classify the populations as different races or subspecies, and the decision about what characteristics to emphasize in describing the differences between them, are relatively arbitrary ones.  In general, among mammals, it may take on the order of a million years for two isolated populations to evolve from genetic identity to a situation of intersterility. It has apparently been about 20 or 25 million years since the human populations became genetically isolated from the great ape populations. Therefore, we should certainly expect the human species to be totally separate from the species of our closest cousins, the chimpanzees and gorillas. In fact, the modern great apes are even considered to belong to a different genus from our own. (The genus is a higher category of genetic differentiation than the species.) For obvious reasons, the experiment of interbreeding man with higher primates has not been carried out under controlled conditions. If nothing else, the possible embarrassment of how to bring up any progeny that might result from such a cross should deter any sensible person from yielding to this kind of curiosity. The differences in appearance, the differences in the number and morphology of chromosomes, and the presumably very long isolation of the populations are all indications - but not guarantees - that the differentiation between man and the great apes has gone well beyond the specific level.

By definition different races are still potentially interfertile, and they often exchange individuals, especially at the boundaries of their geographic distributions.

Sometimes movements of large groups may lead to partial fusion of groups that had formerly split. This process of fusion after fission cannot occur between species whose gene pools are separated irreversibly because interbreeding is not possible.

However human races are defined, there is no evidence of any decrease in fertility in crosses between even the most distant human races (for instance, between Africans and native American Indians). It is clear that human racial differentiation has not had the time or the opportunity to reach a point of incipient speciation.  A much longer period of geographic isolation would probably have been necessary for intersterility mechanisms to evolve. If it is accepted that Homo sapiens s. started racial differentiation at the time the species began to expand and spread throughout the world (about 40,000 years ago), then clearly this time has been too short for any significant reduction of interfertility between the human races to have occurred. It is quite possible that increases in geographic mobility during the past few centuries have reduced isolation of various human population sufficiently to reverse the trend toward racial differentiation in the human species.

Human Races

Race is a more elusive concept than that of species.

A species may be divided into races if the differences between the populations so defined are of some significance, but the level of differences used as a threshold is entirely arbitrary.  When differences are striking, the classification is easy; but even then the taxonomical work may be made difficult by the existence of gradual transitions between the groups so defined.  In most cases, separate races will be defined only if the groups differ in several significant traits. However, in most cases, these traits vary independently of one another in the transitional populations between the races. Thus, boundaries between the races drawn on the basis of one trait will not coincide with boundaries drawn on the basis of another trait. These are precisely the problems faced by the taxonomist who attempts to classify the human species into races. It is not difficult to see why there is nearly complete continuity in the distribution of almost every single trait, as revealed by maps of their geographical distributions.

Human races probably evolved at the time of spread of Homo sapiens s.  The occupation of different habitats in this worldwide expansion created the basis for local differentiation. Three possible complications must be considered.  One is that there probably were, in almost all areas, many successive waves of expansion, so that the process was complex, with later waves contributing to a reduction of the initial genetic differentiation.  A second conceivable complication is that, even in his first expansion, H. sapiens may have found earlier local groups with which he mixed.  The relative importance of this hypothetical mixture will depend on the relative size of the groups with which admixture occurred, if it did so at all. The only serious claim of this nature is with respect to admixture of H. sapiens with Neanderthal man in Europe. Another weaker claim of this sort has been put forward with respect to an analog of Neanderthal in Africa, Rhodesian man.

The recent increase in geographic mobility makes the task of distinguishing races even more difficult.

The third complication is perhaps the most real and formidable of all. In his only expansion, man may have already shown considerable mobility, but the times and rates of this movement are very poorly known. Most migrations, probably including those to Australia and America, could take place only by land, and made use of natural land bridges that were then in existence. Later on, however, man considerably increased his mobility, and much technological progress has been specifically directed at the aim of improving transportation on water and land (and of course, more recently by air). Five centuries ago, transportation technology reached the point where it became possible to explore the whole earth during a fraction of the lifetime of an individual. But in the first exploration and occupation of the earth, it probably took several millennia for the same job.

The process of geographic exploration that started five centuries ago has affected the geographic distribution of man to such an extent that now the Americas and Australia are populated mostly by people of European descent. Today, all continents of the world are inhabited by representatives of the three major human races: African, Caucasian, and Oriental.  The proportions of the three groups still differ considerably in the various countries, and the migrations are too recent for social barriers between racial groups to have disappeared. The trend, however, seems to be in the direction of greater admixture.  Had the incipient differentiation been allowed to continue for a million years or so, different human species would have formed on earth. The dropping of geographic barriers that has followed the recent cultural evolution may in the next centuries erode most of the differentiation that had previously developed. It is for these historical reasons that, when speaking of “aboriginal” groups, one generally refers to the habitat of people before the great geographic discoveries that began in the fifteenth century.

It should not be assumed that human movements were insignificant before the fifteenth century AD.

Early Neolithic man used boats in the Mediterranean, as shown by the utilization of obsidian sources in islands, and the occupation of Crete around 8,000 years before the present. We have seen that the expansion of early farming in the Near East probably involved a creeping occupation of most of Europe, Arabia, Iran, India, and northern and eastern Africa. With the later development of organized trade, sea transportation, and political units of some size, mass colonization became possible. Examples in historical times include the colonization of the Mediterranean by Greeks and by Phoenicians.

All these movements of people are hardly compatible with clearcut, well-isolated geographical races - which in fact are not found. As we shall see, much variation, observed at the level of single gene frequencies, or single anthropometric traits, is clinal - that is, shows a gradual change from one geographical region to another. A cline is a relatively regular change in a biological trait over a stretch of territory.

This background can explain why the job of classifying races is so difficult—the more so, the more detailed we try to be.

On the most general level, however, geographical and ecological boundaries (which acted as partial barriers to expansion and migration) help to distinguish three major racial groups: Africans, Caucasians, and a highly heterogeneous group that we may call “Easterners”.  The Easterners include subgroups that were separated in various older classifications, such as American Natives (American Indians) and Orientals (Chinese, Japanese, Koreans). Some regard Australian aborigines as a separate race, but they do not differ much from Melanesians. From the Melanesians, we can trace a sequence of relatively gradual changes through the transition to Indonesians, then to Southeast Asians, and on to East Asians. American Natives and Eskimos probably both came from a related Northeast Asian stock from (or through) Siberia into North America. Eskimos, however, came much later than American Indians, and they subsequently expanded further eastward to Greenland.

The African continent contains, in the north and east, populations that have various degrees of admixture with Caucasians by all criteria of analysis. In the western, central, and southern parts of the continent, Africans are relatively homogeneous - although some isolated groups of hunter-gatherers (like Pygmies and Bushmen) show cultural and physical peculiarities that suggest they should be considered somewhat separately. In fact, the Pygmies at least have attributes that indicate they may be “proto-African” groups - populations that have been the least altered by more recent events.

Figure 19.6 outlines a tentative history of African people that is in reasonable agreement with archeological and genetic data currently available. Caucasians are spread over Europe and in Southwest Asia as far as India, where there is a relatively gradual transition with Easterners. From Arabia toward East Africa and in North Africa, there is an almost continuous transition toward the African type as one proceeds toward Central Africa.

Linguistic classifications are only moderately useful in defining human races.

Among other possible classifications of human populations, languages prove to be only relatively useful. Languages evolve much faster than genes do. In a few thousands of years, the languages of isolated populations change enough seriously to compromise intelligibility between the groups. Moreover, social and political  events may lead to the substitution of one language for a totally unrelated one in the space of a few generations - or at least such events may cause substantial transformations of an original language. Because of factors like these, linguistic distributions seldom correspond to biological or genetic distributions. Even so, some correlations between the linguistic and biological picture may remain, when analyzed over relatively small distances, so that in some areas populations more similar genetically also prove to be more similar linguistically. This is not a universal finding, however - nor should it be construed as due to any direct interaction between genes and language. (There is no evidence of any innate tendency to speak any particular human language rather than another.) When similarities are found between linguistic and genetic traits, they must be simply reflections of historical events that have had similar influences on both linguistic and genetic evolution.

The correlation between linguistic and biological traits does tend to hold, for instance, among Caucasians. Most Caucasians who inhabit Europe and parts of Asia today speak languages of the Indo-European group. The area where Indo-European languages are spoken overlaps almost exactly with the northern part of the expansion of early farming. In the southern area of the Neolithic expansion, semitic languages are spoken. This is the present situation, but the languages used when expansion occurred in the Neolithic are totally unknown. Linguists do attempt to recreate the evolution of languages by noting the kinds and degrees of similarities and differences between languages, but the theories set forth must remain largely speculative. Only with the development of writing, some 4,000 years after the initial onset of farming, do we obtain a record of the languages spoken by various human populations.

Apart from the linguistic aspects, there is considerable genetic similarity (as we see in detail later) among all Caucasians, ranging from Europe to western Russia and to India and Arabia. As we have mentioned, populations around the fringes of this area show admixture of Caucasian traits with African or Easterner traits. However, the Caucasian population is not entirely homogeneous even in the areas of little admixture. We tend to side with those taxonomists who prefer to group the human species into a few large racial groups (such taxonomists have been called “lumpers”).  Others (“splitters”) prefer to distinguish a large number of groups differing in relatively subtler ways. The splitters have proceeded to subdivide the Caucasian race into smaller groups.

A fairly natural classification, which follows geographic boundaries, is that between European and extra-European Caucasians. Further splitting - for example of the European group into Northern, Alpine, and Mediterranean subgroups becomes more and more ambiguous and uncertain. It is true that the man on the street can usually guess quite accurately whether an individual’s ancestors came from north or south Europe - the typical Scandinavian looks very different from the typical Italian or Greek. Yet the validity and usefulness of this breakdown into smaller and smaller groups is doubtful, at least on the basis of present data.

The bases of the groupings we have been discussing are largely geographic and, as we shall see, the broadest geographic groupings do correspond to a large extent with the available genetic data.
 

- W.F. Bodmer and L.L. Cavalli-Sforza, 1976.  Genetics, Evolution, and Man.  WH Freeman and Company, San Francisco.  pp.559-574.