According to one popular version of the race narrative, natural scientific concepts of race were conceptualized, in the 18th to early 20th century, such to imply discrete categories; and human “races,” were, accordingly, imaged to have significant discontinuities until post-World War II anthropologists, such as Frank Livingstone (1962), came to realize that human variation was relatively continuous, a fact which, the story goes, demonstrated that human “races,” as traditionally understood, did not exist — instead, only “populations” do. Anyone sufficiently familiar with actual 18th to early 20th century discourses on the matter, would find this tale outlandish. They would recall, for example, Wallace’s (1864) account, in “The origin of human races and the antiquity of man deduced from the theory of natural selection,” of the competing pre-evolutionary views about human variation:

In favour of the unity of mankind it is argued that there are no races without transitions to others; that every race exhibits within itself variations of colour, of hair, of feature, and of form, to such a degree as to bridge over to a large extent the gap that separates it from other races. It is asserted that no race is homogeneous; that there is a tendency to vary; that climate, food, and habits produce and render permanent physical peculiarities, which, though slight in the limited periods allowed to our observation, would, in the long ages during which the human race has existed, have sufficed to produce all the differences that now appear …

Proponents of the dominant, monogenist position, for example, the influential natural historian Alexander von Humboldt, in Cosmos (1866, Vol. 1 pp. 351-359), argued that human races were merely different variety-lineages of the same species, as evinced by “the many intermediate graduations, which have been made know to us by the rapid progress of geographical knowledge”; as with his mentor Blumenbach, Humboldt also pointed to the obvious semi-discordant variation, for example, how when comparing Sub-Saharan Africans to Pacific Islanders “we see that a black skin, woolly hair, and a negro-like cast of countenance are not necessarily connected together.” In short, for the majority of natural historians, continuous variation was not taken as evidence against “race,” but as evidence that these lineages were not “species” in the 18th century sense.

Wallace continues:

The advocates of the original diversity of man, on the other hand, have much to say for themselves. They argue that proofs of change in man have never been brought forward except to the most trifling amount, while evidence of his permanence meets us everywhere. The Portuguese and Spaniards, settled for two or three centuries in South America, retain their chief physical, mental, and moral characteristics ; the Dutch boers at the Cape, and the descendants of the early Dutch settlers in the Moluccas, have not lost the features or the colour of the Germanic races ; the Jews, scattered over the world in the most diverse climates, retain the same characteristic lineament everywhere…

In contrast, proponents of the minority, polygenist position, for example, Josiah Nott and George Giddon (1854) argued, in accordance with accepted 18th century systematic principles, that since human races (or lineages) exhibited differences which could not plausibly be accounted for by the environment, they must be either different species-lineages or species hybrids. They did not deny continuous variation. They just attributed this to original similarity and to admixture.  In “Types of Mankind,” for example, Nott and Giddon (1854) take the standard polygenist stance:

We hold that a variety which is permanent, and which resists, without change, all known external causes, must be regarded as a primitive species – else no criteria exists by which science can be governed in Natural History… Hybridity has been considered a test for species; but, when we come to this theme, it shall be proven that, in many instances, what have been called varieties are really distinct species; hence, that hybridity is no test.

According to them, the only reliable test for species was the presence of constant differences. Since original species could hybridize, they could form amalgamations. And since they could hybridize and vary in degree of original distinction, they could exhibit continuous variation, making any classification scheme somewhat arbitrary:

Though many other classifications might be added, the above suffice to testify how arbitrary all classifications inevitably must be; because no reason has yet been assigned why, if two original pairs of human beings be admitted, we should not accept an indefinite number…

Inasmuch as these types are more or less fertile inter se, and as they have, for the last 5000 years, been subjected to successions of wars, migrations, captivities, intermixtures, &c., it would be a vain  task at the present day to attempt the unravelling of this tangled thread, and to make anything like a just classification of types…

As Wallace notes, both sides had a point and the differences of opinion are reconcilable through “the theory of “Natural selection” promulgated by Mr. Darwin. ”

Now, it goes without saying that early evolutionists did not see a contradiction between race and seamless intergradation. The degree of discontinuity was, in part, what justified the conventional ranking of certain races as species versus subspecies versus infrasubspecific divisions. More generally, it should be clear that, as commonly understood in natural history, “race” could not have implied the types of discontinuities is it now often said to have, since the term was used to describe species-lineages, constant variety-lineages, and inconstant variety-lineages. And since only insofar as this term described species-lineages and insofar as certain understandings of species were adopted, was there an expectation — though not a definite requirement — of substantial discontinuity between divisions. But if “race” never implied this, why did Frank Livingstone (1962) juxtapose “races” with clines in the sense of population continua?

Well, he did not.

Rather, Livingstone argued that there are only character gradients [the original meaning of “clines”] which crisscross, and therefore that “race” [meaning common ancestry] had no explanatory power. Notably, he granted the common evolutionary understanding of race, as descent group, but rejected its applicability to humans and other species, on the grounds that, on the infraspecific level, common descent did not explain variation. As such, he tells us:

To apply a concept of the Linnean system to a group of populations implies something about the evolutionary history of these populations, and it also implies that these populations are similar in whatever characters were used to classify them together because of close common ancestry. It is this implied explanation of whatever genetic variability is used to group populations into races which I consider to be false.

Ignore the conflation of systematic concepts (e.g.,morph, variant, and race) with taxonomic ranks (e.g., genus, species, subspecies). Livingstone is correct that Darwinians used the term “race” (fr. lineage) to refer to groupings into which individuals were arranged by propinquity of descent; and they imagined that commonality in descent, in turn, explained similarity in form — not all, but some, similarity, since they recognized convergent evolution.

As contemporary population geneticists commonly employ a doppelganger concept – just giving it a different label e.g., “descent group” and “ancestry population” — and since, when it comes to humans, they end up delineating roughly the same major descent groups e.g., “Sub-Saharan Africans” (“Negroids”) and “West Eurasians” (“Caucasoids”) — and since they find that ancestry is often a useful epidemiological variable, we know that Livingstone was wrong.

But where did he go amiss? Here is what he tells us:

The causes of intraspecific biological variation are different from those of interspecific variation and to apply the term subspecies to any part of such variation not only is arbitrary or impossible but tends to obscure the explanation of this variation.

So the evolutionists’ 1st error: Conclude that the general causes of infraspecific and specific variation are the same…

Of course, the breakthrough with biological thinking came when evolutionists like Wallace grasped the problem of drawing a clear ontic and epistemic distinction between specific and infraspecifc variation. In, Note on the theory of permanent and geographical varieties, written the same year as his essay on evolution, On the Tendency of Varieties to depart indefinitely from the Original Type, Wallace elucidated the problem:

The adoption of permanent and geographical varieties has this disadvantage, that it leaves the question “What is a species?” more indeterminate than ever; for if permanent characters do not constitute one when those characters are minute, then a species differs from a variety in degree only, not in nature, and no two persons will agree as to the amount of difference necessary to constitute the one, or the amount of resemblance which must exist to form the other…

Now the generally adopted opinion is that species are absolute independent creations, which during their whole existence never vary from one to another, while varieties are not independent creations, but are or have been produced by ordinary generation from a parent species. There does, therefore (if this definition is true), exist such an absolute and essential difference in the nature of these two things that we are warranted in looking for some other character to distinguish them than one of mere degree, which is necessarily undefinable. If there is no other character, that fact is one of the strongest arguments against the independent creation of species, for why should a special act of creation be required to call into existence an organism differing only in degree from another which has been produced by existing laws? …

To escape this difficulty there is but one way: you must consider every group of individuals presenting permanent characters, however slight, to constitute a species; while those only which are subject to such variation as to make us believe they have descended from a parent species, or that we know have so descended, are to be classed as varieties. The two doctrines, of “permanent varieties” and of “specially created unvarying species,” are inconsistent with each other.

He deduced that either one should recognize all “permanent varieties,” no matter how slight the difference, as Linnaean species, as polygenists argued, or one should grant that divisions within species can vary in permanently hereditary character, which opens the door to the transmutation of species. Others, of course, had recognized the basic logic, but reemployed it as a reductio ad absurdum against monogenism. For example, commenting on one of Prichard’s suggestions regarding the origin of human racial differences, de Gobineau (1853) notes:

No argument can be based on the accidental deviations from the normal which are sometimes found in certain individual instances, and which, if transmitted, would certainly give rise to important varieties…To cite only one instance, Prichard quotes Baker’s account of a man whose whole body, with the exception of his face, was covered with a sort of dark shell…. Four sons were born to him, all resembling their father. One survived; but Baker, who saw him in infancy, does not say whether he reached manhood. He merely infers that since the father has produced such offspring, “a race of people may be propagated by this man, having such rugged coats and coverings as himself ; and if this should ever happen, and the accidental original be forgotten, it is not improbable they might be deemed a different species of mankind.”

Such a conclusion is possible. Individuals, however, who are so different as these from the species in general, do not transmit their characteristics. Their posterity either returns to the regular path or is soon extinguished. All things that deviate from the natural and normal order of the world can only borrow life for a time; they are not fitted to keep it. Otherwise, a succession of strange accidents would, long before this, have set mankind on a road far removed from the physiological conditions which have obtained, without change, throughout the ages. We must conclude that impermanence is one of the essential and basic features of these anomalies. [Emphasis added.]

As de Gobineau (1853) points out, if one grants the Unitarian logic that individual degenerations can accumulate and can become fixed within an infraspecific lineage, then there is no reason why species could not transmuted across the generations. (To note, to avoid the charge of heresy, de Gobineau proposed that human races were human varieties whose differences had somehow became fixed during the cataclysms of creation and so had species-like differences, despite not being originally distinct; regardless, he rejected the idea that in ordinary times varieties could become permanent.) So a few years before Wallace, de Gobineau (1853) had also pointed out the inconsistency between the idea of “specially created unvarying species” and that of the acquisition of permanent hereditary variation.  de Gobineau (1853) just took the possibility of species transmutation as a reason for rejecting the latter.

Continuing with Livingston:

If one genetic character is used, it is possible to divide a species into subspecies according to the variation in this character. If two characters are used, it may still be possible, but there will be some “problem populations,” which, if you are an anthropologist, will be labelled composite or mixed. As the number of characters increases it becomes more nearly impossible to determine what the “actual races really are.

So the evolutionists’ 2nd error: Suppose that “correlated variation” can be used to index common descent on both levels…

Now, this “non-concordant variation” argument might “work” against a concept of race in the sense of “hereditary variety” if we took “variety” to mean an infraspecific group of organisms defined only by one or a set of hereditary characters — as opposed to population-lineages distinguished by these characters. But it only would do so vis-a-vis a phyletic based-systematics which has the goal of grouping organisms by overall relatedness.

The problem would not be, in this case, that there were no “races,” since infraspecifc hereditary variation is now obvious, but that this-sense “races,” would not necessarily cut out “good” phyletic unites, or “races” understood as “communities of descent.” Yet, Livingston starts with the concept of race as descent group and is arguing, erroneously, that these can not be delineated, on the infraspecific level, by applying multivariate analysis to phyletic-informative characters.

Livingston continues:

In this way race or common ancestry and migration have been used to explain much of the genetic variability among human populations. Unfortunately such explanations neither accord with our knowledge of the population structure and movements of hunters and gatherers, nor take into consideration the basic cause of biological variation, natural selection. The incompatibility between race and natural selection has been recognized for a long time; so that if one’s major aim was to discover the races of man, one has to disregard natural selection.

So the evolutionists 3rd error: Fail to recognize the incommensurability of “descent” and “modification”…

Evolutionists, of course, solved this intractable-to-Livingston problem by establishing Haeckelian monophyly as a classification principles: organisms are arranged into divisions by descent and then divisions are deferentially ranked to express degree of modification.

In retrospect, it seems strange that Livingston’s argument – cited, for example, in the background materials to PBS’s “Race – Power of an Illusion” – gained the traction it did. The argument is oddly similar to that made by Georg Forster and other 18th to mid-19th century polygenists, who argued that race, meaning descent, did not explain infraspecific biological variation, and who did this by affirming a tight distinction between infraspecific and interspecific variation. In the age of ancestry testing, Livingston’s argument has only persisted, in the form of a slogan, owing to an equivocation between cline as “character gradient” (original meaning) and cline as “population continua” (new meaning) and the heavy recruitment of fake history to justify the view that “race” once implied significantly discontinuous descent groups. This revision has allowed certain contemporary geneticists to feign a distinction between their “biogeographic ancestry groups” and 19th century evolutionary races.

Of course, “race”-denial is a lucrative business; and “race”-recognition is sociopolitically risky for researchers, so one can only expect bromides to the effect of:

Well, yes, ancestry with respect to descent group is explanatorily useful. But race is something else — as Livingston pointed out, “there are no races, there are only clines.”