Sewall Wright on race differences, group selection, and cultural selection among humans (1978)

[From Evolution and the Genetics of Populations, Volume 4: Variability Within and Among Natural Populations, pp. 439-457]

Racial Differentiation in Mankind

The existence of conspicuous diversity among human populations in physical appearance has been common knowledge at least since the time of ancient Egypt. The subject is discussed at length in numerous books on physical anthropology and need not be considered here in detail.

There is no question that all mankind constitutes a single species in view of the absence of any physiological bar to hybridization between the most diverse races or of any recognizable loss of vigor in the first or later generations.

There is also no question, however, that populations that have long inhabited widely separated parts of the world should, in general, be considered to be of different subspecies by the usual criterion that most individuals of such populations can be allocated correctly by inspection. It does not require a trained anthropologist to classify an array of Englishmen, West Africans, and Chinese with 100% accuracy by features, skin color, and type of hair in spite of so much variability within each of these groups that every individual can easily be distinguished from every other.

[. . .] It has been indicated earlier that such an evolutionary process as that of man is much more understandable if it occurred by the shifting balance process. Simultaneous sampling drive at thousands of sufficiently neutral loci provides different material in innumerable localities without appreciable cost, material that can give the basis for effective interdeme selection.

Was the population structure of primitive man favorable to this process? There have been a number of studies of the few remaining peoples at the hunting and gathering level of culture that bear on this matter. Birdsell (1972), in an intensive study of Australian aborigines in western Australia, has described their population structure. The primary territorial unit is the band, consisting of a group of related families. Marriage is exogamous but largely restricted to the tribe, a group of bands in which the same dialect is spoken. He estimates the average total number in a tribe to be about 500, with a breeding population of about 185 and an effective number of about 100. This is small enough for the building up of considerable differences among large areas at each nearly neutral polymorphic locus merely by sampling drift. There is thus the basis for operation of the shifting balance process. [. . .]

The actual process of interdeme selection may take different forms. At one extreme, the local appearance of a superior genetic system is followed by expansion of its territory accompanied by complete elimination of its neighbors until it occupies the entire range of the species. At the other extreme there is merely excess diffusion from the superior center. Neighboring populations are graded up until they reach the point (the crossing of a saddle in the surface of selective values) at which mass selection carries them autonomously to the new selective peak, or perhaps beyond, if they contribute something that improves on the latter. The locations of the population with the highest selective peak may shift from place to place in the course of time, as a group of neighboring populations step each other up to heights well above the general level.

Bigelow (1969) has emphasized the importance of tribal warfare in the operation of this process. A tribe that is generally successful because of superior intelligence, capacity for cooperation, and high frequency of the heroic virtues as well as physical prowess, tends to increase its territory and also to grade up what is left of the defeated group by hybridization. The process is illustrated by the incessant tribal warfare of the tribes of American Indians observed by the European settlers in America, in which some tribes such as the Iroquois expanded at the expense of their neighbors.

The heroic virtues, including willingness to sacrifice one's own life for the good of the tribe, are traits that can hardly be developed (insofar as they have a genetic basis) by purely individual selection. They may to some extent arise as a by-product of familial selection in which close relatives with heredities strongly correlated with that of an individual who gives his own life to save them. As noted earlier, the effectiveness of familial selection in general is testified to by the improvement of milk production in cattle and of egg production fowls, mainly by selection of males on the basis of the performance of close female relatives. The importance of this sort of intergroup selection in evolution has been emphasized as noted in chapter 7 by Hamilton. The increase in frequency of traits deleterious on the average to their possessors but beneficial to the deme may also, however, be increased by interdeme selection (referred to as intergroup selection in early articles) if the benefit to the deme sufficiently outweighs the damage to the individual. A more rigorous demonstration of this mode of evolution of "altruistic" characters as been given by Eshel (1972).

Not all interdeme selection in man has consisted of intertribal warfare. According to Birdsell, about 15% of marriage among Australian aborigines were intertribal; not enough as shown by the wide variability of gene frequencies to homogenize the whole population but enough to permit effective interdeme selection if exchange was asymmetrical, predominantly from the more to the less successful. That differences were not swamped was presumably due to the exchange being largely between neighboring tribes which differed little, as indicated by the semiclinal nature of the pattern of gene frequencies. The average effective immigration from the population as a whole, the m of formulas, was thus very much less than 0.15. [. . .]

Evolution Since the Origin of Agriculture [. . .]

We can form a better idea of the course of event than in any earlier period but the interpretation is confused by the exponential progress of a second evolutionary process, barely existent at all in any other animal, and little if any more rapid in the earlier history man than his biological evolution.

This is the evolution of culture with its line of transmission largely from speaker to listener, supplemented in the last three thousand years by transmission from writer to reader. It began to become of major importance with the origin of language, but during the hunting and gathering phase of human life the slow advance of culture is indicated by that in the fashioning of stone tools and weapons. It was probably accompanied by relatively rapid diffusion of knowledge of such advances as were made. The success of tribes thus probably depended to a greater extent on capabilities, determined by their genes, than on the possession of techniques not known to their neighbors.

The mode of evolution of culture is analogous to that of the genetic system. Invention is the analog of mutation. Diffusion of culture is the analog of gene flow. Cultural variation is continually subject to selection on the basis of utility. There is random cultural drift, exemplified by the breaking up of languages into dialects. Finally, the most favorable conditions for cultural advance is local isolation, providing the basis for simultaneous trial and error among many variants and the diffusion of the more successful ones in analogy with the shifting balance process in biological evolution. We think here of the multiple competing cultures in ancient Southwest Asia, the evolution of culture among the city states of ancient Greece, and in much divided Europe from the Dark Ages to modern times. The great empires of the ancient Southwest Asia of Alexander and of Rome constituted an overbalancing final phase in the process, giving widespread diffusion but less progress by trial and error.

There has undoubtedly always been a considerable but incomplete correlation between the two kinds of evolution. The state of the culture has been to a considerable extent an index of the rank of populations genetically in the distinctive human line of evolutionary advance, and reciprocally the demands of culture have been the primary selective agent in this advance in its later stages. Aspects of culture are continually being borrowed, but whether such borrowings are effectively integrated into the existent culture to form new peaks (as most conspicuously in the recent period in Japan), or are adopted only superficially and to the detriment of the previous culture, is also an index of genetic capability.

The treatment of either the genetic capabilities or the cultures of peoples as if they could be ranked on single scales is, of course, a gross simplification. If the multiple genetic aspects of mental ability could be measured more independently of culture than is the case, it would no doubt be found that each local race has its own unique combination of favorable qualities. At present only IQ seems to have a repeatability that permits evaluation of the contributions of genetic and nongenetic variabilities to its variability, discussed in the previous chapter, and this only within a particular culture.

On the other hand there have probably always been wide differences among the peoples of the world in average intellectual ability and cultural level from the standpoint of progress toward the situation in civilized man. This was presumably related to the environmental conditions. Men could not endure the northern winters without fire, the use of which is documented by hearths found in France dating back over half a million years and somewhat later in Hungary and in China but only about one-tenth as far back in Africa (Campbell 1974).

The capacity to anticipate and plan for the future is a mental attribute which would be favored under northern conditions and selected for insofar as it has a genetic basis. This would presumably have come to be more advanced in the temperate zone than in the tropics. [. . .]

Linguistic evidence indicates the establishment of an important center of diffusion in east-central Europe some 5,000 years ago from which wave after wave peoples moved in all directions. The Hittites carried an Indo-European language of the western (centum) type into Asia Minor and established an empire some 4,000 years ago. A thousand years later the Iranians, who had moved east into what is now southern Russia and Turkestan, brought an Indo-European language of the eastern (satem) type into the original cultural center and later established the Persian Empire. They also carried another Aryan dialect to India. Other tribes moving south from the east-central European center reached Greece in several waves which, after mixing with the indigenous people, produced classical Greek civilization.

Other waves moved to the southwest into Italy, giving rise to Latin and other Italic languages; to the west, giving rise to Celtic languages in what is now southern Germany, France, and the British Isles; and to the northwest into what is not northern Germany and Scandinavia, to give rise, in much altered form, to the Germanic languages. Subsequent migrations greatly expanded the areas occupied by derivatives of Latin and Germanic branches at the expense in Europe of the Celtic. All the tribal migrations were undoubtedly accompanied by much intermixture with indigenous peoples, but the diffusion of language also undoubtedly implies considerable gene flow.

[. . .] The history of Europe, especially western Europe, was thus prevailingly one of inflow of genes up to the relatively recent period in which it itself became a center of massive outflow.

Related posts:


Anonymous said...

wright's parents were first cousins which bodes ill, though i have cited him before for his idea of fitness landscapes...a topic un-understood by HBDers.

the theory:

there has been selection for preferring those one is more closely related to.

the reality:

people prefer people like themselves whether in the same family or race or not.

even with immediate family patricide, matricide, fratricide are common enough. and this especially because immediate family are often so different from one another. the concern for and/or attachment to one's immediate family can be explained by familiarity more than by family resemblance. take steve jobs. his dad was a syrian grad student in the useless polysci. he wanted nothing to do with his dad. he hated him. yet, at his behest, he was buried next to his adoptive parents!

the one harvard admit i went to school with described his brother as a bonehead. his brother was popular especially with girls, an athlete, but never graduated from college. this harvard guy went on to earn his mba from wharton and last i heard works in The City.

i'm a total is the sense that great dissimilarity in appearance whether from differences in geographic origin or from say having a baby's arm coming out of your head is disconcerting...and thus irritating and productivity retarding.

and this is exaggerated even more by the histories of the groups to which these conspicuous people belong. why? because "we all know" that blacks were slaves, have high rates of criminality and incarceration, are de facto segregated just like "we all know" that ca. 1/3 of the richest americans are jews, that jews score high on iq tests and have won a disproportionate number of nobels, and are a "hold-over/coelacanth/atavistic-er/i can't remember the noun" for "christians" and for chistendom, that is europeans and europe.

Anonymous said...

and...physical "attraction" overpowers all other feelings of affinity...that is, appearance, and thus race, is top of the list of what makes someone "attractive" or not. the quotes are to distinguish generic affinity from sexual attraction.

in my own case i can say those women i've been most attracted to/attached to, but didn't marry, were all (at least) majority european but with ancestry quite unlike mine. one was swart and welsh, another was half hawaiian, another was half ashkenazi/half russian, etc.

my continental ancestry is a minor part. my most recent immigrant ancestor was a ship's captain from liverpool. he came over in...late 19th c.

Anonymous said...

and me?

i don't look much like any of them, but of all "celebrities" i look most like the german defender christoph metzeleder and the american actors william hurt and andrew mcarthy.

every one of my 16 great great grandparents had blue or blue-green eyes. no brown eyes in the family tree afaik.

and the "spanish" side was the fairest and tallest. my great great granddad was 6'3" with blond hair and blue eyes. he lived to 99 and drinking 2 fifths a week! the tall and robust are usually shorter lived, but those men with such "misfortune" can always look to this guy:

Anonymous said...

"There is no question that all mankind constitutes a single species in view of the absence of any physiological bar to hybridization between the most diverse races or of any recognizable loss of vigor in the first or later generations."

Are there any scientific studies that support this claim or are we just not allowed to go there? Modern men could breed with Neanderthal but Neanderthal weren't considered homo sapiens though that may change to a subspecies given the political need to make people all the same.

It seems to me just a line drawing exercise.

And what of vigor? What scientific studies have followed mixed race people over first and later generations? Why should the above just be assumed?

Anonymous said...

yes i just changed my registration from my age 18 "no party" to dem so if it's hillary vs sanders i can vote for the super jewy jew sanders in my closed primary state.

but if it's hillary vs trump i'll vote trump.

trump inherited 200-600m, so he's not the typical white trash conservatard, that is, he is white trash but he isn't a conservatard.

trump and sanders are more similar than two front-runners have ever been.

if it comes down to trump v sanders i'll have to ask how serious either is about closing the open borders. trump is a rich gentile and sanders is a jew, so they both have open border sympathies.

hey, i love long as they're in israel. outside of israel i want to gas them.

n/a said...


"wright's parents were first cousins which bodes ill,"



"Are there any scientific studies that support this claim or are we just not allowed to go there? [. . .] And what of vigor? What scientific studies have followed mixed race people over first and later generations? Why should the above just be assumed?"

Some degree of outbreeding depression for crosses between distant races would not surprise me, but I think it's likely to be minor. I think studies so far have tended to simply show interracial crosses intermediate between the parent races on the quantitative traits that have been looked at. You would probably need very large, well-controlled studies to detect or completely rule out outbreeding depression.

Anonymous said...

Thanks for your comment n/a. Bone marrow matching for mix-race patients points to one such outbreeding depression. We can hope that possible out-breeding depressions are minor but the fact that we don't know they are bothers me. We know about the possible inbreeding depressions and we even have laws against inbreeding and yet out-breeding we are left to speculate about.

Lemniscate said...

While interbreeding between West Africans and Europeans may not produce easily detectable outbreeding depression, this hardly represents the maximal genetic distance between races. The Khoe-San seem to be the most diverged from the rest of humanity, but the population is so small and endogamous it would be hard to get data on mixed marriages with Khoe-San. Nevertheless, I wonder if there would be significant outbreeding depression in a Khoe-San x Australian-Aboriginal marriage. Given that there is evidence for outbreeding depression between 'out of Africa' humans and Neanderthal, and given that Australian Aboriginals have some fraction of archaic non-African admixture and Khoe-San probably have some fraction of archaic African admixture, it would seem likely to me.