A reader asked if I might refer him to a cogent, while pithy, elaboration of the natural historian’s concept of race, an exposition which he might cite in future discussions. One of the most lucid articulations which I have encountered can be found in physical anthropologist Alice Brues’ (1913 –2007) book “People and Races” (1977/1990). Brues studied under Earnest Hooton, whose own concept of race was remarkably well articulated and coherent. In undergrad, she majored in philosophy (and psychology), a fact which might help account for the unusual lucidity of her discussion. In the seven pages of her first chapter, she says most of what needs to be said. And in the remaining chapters she makes the other necessary points. The first chapter is copied below both in PDF form and text. The discussion can be summarized as follows (with my notes added and paragraphs numbered).
P1. A race is a division of a species, the members of which share more ancestry with each other than with members of other divisions; different races differ in the frequency of hereditary traits.
[This captures the early notion of races as genealogy-based classifications.]
P2. Individuals are arranged into divisions and groups are distinguished based on “the total appearance of the individual” and their “normal combination of traits”.
[This reflects the common notion of races as natural biological divisions, ones delineated based on an ensemble of ancestry indexing characters.]
P3. The hereditary nature of race explains why when populations relocate from one part of the world to another descendants resemble their region of origin ancestors.
[This expresses the concept’s original purpose, which was to explain why varieties retained their particular traits even after being relocated to foreign environments.]
P4. Variation between races is generally continuous.
[This restates the popular monogenist position that trait discontinuities evidence species, while trait continuity evidences intraspecific divisions.]
P5. History has been marked by shifting patterns of racial differention and racial admixture. Primarily and pure races exist/existed only in the sense of population-lineages which are/were relatively more isolated and differentiated than secondary or admixed ones.
[Races, as divisions of a species, have always been understood as population-lineages which diverged from a common stock or which represent new lineages forged through the hybridization, and subsequent semi-isolation, of two or more such lineages.]
P6. Races are not national, religious, linguistic, or cultural communities. While communities defined by political, religious, and linguistic affinity, if reproductively isolated for a long enough time, form races, they need not necessarily and often do not.
[This restates the idea that the natural historian’s races are somatically, genealogically, and genomically, not socially, delineated divisions. During the 1800s the term was often applied confusedly. Deniker (1900) was one of the first to clarify the distinction between “races” and “ethnic groups formed by virtue of community of language, religion, social institutions, etc.” Of course, it has since been shown that many such ethnic groups correspond with races. Thus, for example, Paschou et al. (2010) found that with enough ancestrally informative alleles they could assign most individuals to their self-identified HapMap reference “ethnic” group of origin. As the authors note, “subtle population genetic structure can be detected at levels that were previously unimaginable”. Nonetheless, the logical distinction between races as genealogically defined divisions of a species and national, religious, linguistic, and cultural communities as such stands.]
P7. Race is related to cultural and societal differences – to the extent it is – on account of inherited differences in temperament and aptitude which influence how groups think and conduct themselves.
[This clarifies the relation between race and behavioral population genetics; it distinguishes biological understandings from quasi-mystical ones.]
P8. The number of races recognized depends on the classification scheme. The major races recognized were the “white” (Caucasoid), “yellow” (Mongoloid), and “black” (Negroid) races. There are many other races, groups which represent either subraces of the major ones, intermediate populations, or groups which do not neatly fit into any classification scheme.
[This restates the point made so many times since the time of Buffon that one can lump and split races.]
P9. Populations are grouped into races based on genetic affinity resulting from once sharing a common breeding population. This affinity is indexed by similarity in characters. Characters are selected insofar are they are thought to index common ancestry. Traits subject to selection are viewed with caution.
[This reemphasizes the point that races are genealogical based classes, the members of which owe their commonality to their ancestors having once shared a common breeding community. It also clarifies the point made by Darwin that when it comes to assessing affinity, resemblance in traits subject to intense selection should be viewed with skepticism.]
It probably does not need to be said that there has been some variability across time regarding how biological race was conceptualized. The above understanding, though, represents a, if not the, typical one, one according to which races were said to be descent based divisions of a species, the relationship between which was indexed based on ancestrally informative characters. Now, some object that this concept is radically different from that of Comte de Buffon, the later Blumenbach, Darwin, Deniker, Hooton, and others. I would like to see a demonstration of this, one based on primary sources. Others object that such a concept is radically different from our population geneticists’ and biological anthropologists’ “biogeographical ancestry” group, “population structure”, “genomic clusters” and so on, despite the fact that the latter concepts purport to cut out roughly the same types of groups, that the classes cut out in line with these nominally different concepts are little different, and that many critics of “race-thinking” recognizing the equivalence of the concepts and thus complain:
“It appears that many scientists do not even believe this distinction makes a difference; they have concocted a thinly disguised euphemism for race they hope will not stir up as much controversy. Geographic ancestry has not replaced race — it has modernized it.” (Roberts, 2011)
“One might fancy that typological days are safely locked away in the cobwebs of history, but the same typological thinking is still here, all around us (Weiss & Long, 2009). In one of its more rigorous forms it is called “structure” analysis, after the first of several programs that perform it (Pritchard et al., 2000)… It has become a routinely used approach by scientists who would not dream of using words like “race” and have no discriminatory or eugenic intent, but who may not be aware of the history of such concepts.” (Weiss and Lambert, 2011)
“Race thinking in science is still with us today, despite a few brief retreats as recently as the year 2000. Increasingly, the use of race in certain geneticists’ circles can be seen as acceptable on several registers. Scientists who organize studies by race, even if they prefer the euphemism “continental genetic ancestry,” now hope to include racial minorities in projects with social justice and real capital effects.” (Fullwiley, 2014)
“Thus, though the “population” concept is touted as an advancement in freeing genomics from racial bias, it is merely a terminological mask for “race” in genomics… The language employed to talk about “race” without talking about it overtly then takes the form of racial euphemisms like “population.” (Williams, 2015)
 It is conventional to start with definitions of terms to be used. We will not define a human being because each of our readers knows one of them with special immediacy. Race has been considered a hard word to define, though it is probably no more so than many other commonly used words that refer to collections of events or objects. A race is: a division of a species which differs from other divisions by the frequency with which certain hereditary traits appear among its members. Among these traits are features of external appearance that make it possible to recognize members of different populations by visible inspection with greater or less accuracy. Members of such a division of a species share ancestry with one another to a greater degree than they share it with individuals of other races. Finally races are usually associated with particular geographic areas.
 It is important to realize that we can describe races only on the basis of differences between populations. The frequency with which a trait occurs in a population can be evaluated only if we have a statistical sample. A single individual is not a race, and no single individual will match in every respect the average of the population from which he comes; most individuals, in fact, are quite non-average in one way or another. If we try to make fine distinctions in Europe, for instance, we will find that the differences between populations are rather small, and are confusingly overlaid with a great deal of individual variation. In this case, certain individuals could not be identified with any particular area, and we might say of a man, “He could have come from almost anywhere.” However, if we had a sample of a number of individuals from the same area, we would be more confident about deciding where the population came from: and the larger the sample, the greater would be our confidence. We might say, “There couldn’t be so many people who look like this anywhere but in nation X.” If we were to compare two very different population groups, such as Norwegians and Tanzanians, we might find overlap in a single trait, but not in the total appearance of an individual. Out of a thousand of each group, we could easily find a swarthy Norwegian whose skin color was similar to that of an unusually fair Tanzanian. And we could find a Norwegian with a kind of bushy hair that occurred sporadically among Europeans, and comes within the limits of variation of African blacks. But this hair form is not correlated with skin color in European populations, and it would be one-in-a-hundred-thousand chance to find the two variants in the same individual. Still to be taken into account would be many differences in facial features between the two races. In this case we could not probably mistake our aberrant Norwegian for a Tanzanian, though we might reasonably guess that he was from any of a number of places, and confess that he was an anthropological puzzle. Races, then, even quite distinct ones, can not be clearly defined in terms of one or a few characteristics; they must be defined in terms of their normal combination of traits.
 The hereditary nature of the traits that are significant for distinguishing races is shown by the fact that when populations move from one part of the world to another, descendants continue to resemble their ancestors. Some Americans look European, and some look African, because their ancestors came from these regions many generations ago. If American born descendants of Japanese are taller and heavier than their parents, we do not say that their racial character has changed; we conclude that height and weight are only partly correlated with race because they are also affected by environment. Obviously there is a correlation between physical appearance and geographical area of residence or ancestry, but if there is a causal relation, it is not so simple or immediate that the children of Old World settlers in the New World grow up to look like American Indians. The nature of the relation between area and race will prove a very interesting one.
 The importance of common ancestry in determining racial characters is also shown by intergradation between races wherever populations have been incompletely separated in the past. This effect, combined with individual variation within groups, makes it impossible to set definite boundaries to races, in most cases. In a simpler and more innocent world, before people had moved about we might have made a leisurely trip away from home, in any direction, and observed a changing appearance of the people we saw, that would have given us a realistic view of racial differences. A few hundred miles away we might have noticed that certain variations we knew among the people of our own village – blond, brunet; tall, short; beaky nose or turned-up nose – were commoner or rarer than we were accustomed to. We might have seen a number of individuals who would look rather unusual back home. As we went further, we might begin to see individuals who looked like no one we had ever seen at home; and farther away still, we might find a population in which nobody looked like anybody back home. At what point in these travels could we say we had met a different race? We had observed an increasing racial difference from the population at our starting point, but where would we draw lines between races?
 If we recognized as distinct races only populations so different that no individual of one of them could ever be mistaken for a member of the other, most of the people in the world would not belong to any of our races. Only when people whose ancestors originally came from widely separated areas have later been brought into contact with one another by travel or migration, do we have a situation in which clearly different elements in a single community (that is, elements with little or no overlap of physical appearance) can be defined as separate races. This is an abnormal, and largely a recent, phenomenon, that has given rise in some areas to attempts to define race in a legal way. However, if we look at races in their natural habitats, continuity is the rule rather than the exception.
How then can we define a particular race? The answer is that we cannot. We can try to define the word race itself, but individual races, like many other biological phenomena, can only be described. Some have said that for this reason we should not name races, although we can measure degrees of racial difference between populations. It is quite likely, however, that we will continue to name races in the future, just as we name colors — though colors intergrade infinitely, and people often can not agree on which word to apply to a specific hue. Other sciences has similar problems of terminology. Biologists give names to “biomes” – characteristic plant-animal associations of different areas – which intergrade everywhere at their boundaries. And a geographer speaks of “climates,” although a climate is only a statistical generalization derived from variety of weather events that occur with different frequencies in different areas. Yet it is meaningful to say that the climate of Colorado is colder than that of Florida, even though there are many days in each year when it gets warmer in Denver than it does in Jacksonville.
 Some races are grouped into “primary” races, which are unique or extreme in certain respects, and “secondary” races, which appear to be intermediate between neighboring populations. This is a relative matter, because nearly all populations have some distinctive features that cannot be explained in terms of mixture between any of their neighbors. One indirect inference from this kind of classification is generally wrong: the notion that at some time in the past races were “purer,” that is, more distinct from one another and more uniform within themselves, than they are now. This may be true, to a degree, in some cases: for example, in Europe during the last few thousand years. In the Neolithic period, populations were smaller, and regional differences may have been clearer at that time than they are now. But it is not safe to extrapolate this further backwards. Alternating periods of migration and isolation have probably occurred throughout human history, with race formation (that is, acceleration of regional differences) at some times and places, and race mixture at other times and places…
 It is important to distinguish certain differences between human groups which, although they may correlate with race, are not the same thing. Sometimes the name of a nation is used as if it were a designation of race, as the “Irish race” or the “German race.” If the boundaries of these nations were quite permanent and well-sealed against immigration, the populations within them might in time come to be genetically meaningful units. However, national boundaries have often shifted in the past, and peoples have crossed boundaries, either in groups as invaders, or individually as immigrants, so there may be more difference, genetically, between different parts of a single nation than between adjacent parts of different nations. In the case of some large modern nations like the United States, extremely divergent races are present under a single flag. It is unwise, therefore, to use terms that confuse nationality with race. Languages do not define races either, though common language is sometimes evidence of common ancestry, and may be a guide to understanding the history of population movements and racial affinities. Language differences also inhibit communication and may thus discourage intermarriage. But languages can be imposed on groups by political means: the “Latin languages” of Europe do not define a “Latin” race; these languages are relicts of the Roman Empire, which, at its greatest extent, included a rather diverse lot of people. And in some areas, languages that are not detectably related may be spoken by peoples whose physical type is quite similar, as among the Plains Indian tribes of North America. Nor does religion define race. Though in some areas there may, for historical reasons, be racial differences between adherents of different religions. And race is not culture, though different racial groups in an area may have cultures that are different to a greater or less degree. Many dubious statements have been made in the past about causal relations between race and culture. If we define race in genetic terms, the only meaningful relation between race and culture would be the one that was the results of inherited differences in temperament and aptitude, which determined to some extent the various modes of thought and actions of various populations.
 Thus, there are more races than most people think there are. In part, of course, the number of races we recognize depends on how we group together, in major classifications, various combinations of local populations that are distinguishable but not radically different. No one, so far, has reduced the number of major races to less than three, however. We will recognize three racial groups that correspond in a general way to the old popular designations of “white, yellow, and black,” and will discuss numerous varieties and subvarieties of these. Blumenbach’s “red” race of the Americas shares many characteristics with the “yellow” of Asia, and the two are generally classed together as a single very widespread, major race. Most of Blumenbach’s “brown” race would be included with them also. In addition, we shall call attention to other human populations, smaller in number, which cannot be honestly fitted into any simple scheme and should be considered races in their own right.
 When we find two populations who are alike in many traits, though not all, and who live in geographically contagious areas, we naturally assume that they have been part of a single breeding unit in the past, and may still be to some degree. We will therefore class these peoples as member of one race, but make a distinction of subrace, in view of their minor differences. This kind of reasoning is less convincing when geographical continuity of residence is interrupted, and when genetic similarity is less conspicuous. In this case, we base our judgement of genetic relatedness on the plausibility, in historical and geographical terms, of former membership in a single breeding population, and on our assessment of the “importance” of traits the populations share as opposed to those they do not share. We can automatically discount, to a proper degree, traits that are known to be subject to immediate environmental effects on living individuals. Any trait is suspect of being subject to strong selection in relation to environmental differences will not be considered very reliable for tracing genetic relatedness far in the past, because strong selection may have produced significant local changes. Another reason for doubting the usefulness of a trait for tracing ancestral relation may be that the conditions are not genetically the same. We do not know, for instance, whether very dark skin that occurs in some widely separated and otherwise different peoples of the world is caused by the same genes at the same loci. We do know that what we call frizzy hair is not all the same when examined closely, and that different types occur in different populations… One solution to the problem of evaluating the importance of different traits in assessing relationship [is] simply to use as many traits as possible and assume that all are equally important, or at least that ratings of similarity or differences made on this basis are approximately correct. This may lead us to underrate relationship, however, where there has been intense selection or exaggerated drift due to small population size.