Do you need to look at genetic differences between races to see if race is real? Some may argue that you do, and when you do you’ll see that genetic variation is too small to say that race exists. However, other arguments exist that do not look at genetic differences between races, but look at geographic ancestry, reproductive isolation between races, and morphologic differences. Those three variables are enough to prove the existence of race without looking at genetic differences between races. They do correspond to genetic differences between races. The four concepts I will briefly lay out are from Michael Hardimon, professor of philosophy at University of California, San Diego. The concepts are the racialist concept of race, minimalist concept of race concept, populationist concept of race, and the socialrace concept of race. One doesn’t need to look at the racialist concept of race to prove the existence of race, which I will prove below.
Michael Hardimon published Rethinking Race: The Case for Deflationary Realism earlier this year. In the book, he makes the case that race exists if minimalist race exists (I will get into what minimalist race entails below). Nevertheless, race deniers will say that even by looking at variables such as morphology, reproductive isolation, and geographic ancestry, race as a concept is scientifically invalid. This is patently false.
Concepts of race
The racialist concept of race
Hardimon’s first race concept is the racialist concept. The racialist concept (keep in mind, this is, as Hardimon writes on page 17 of his book Rethinking Race “the specific concept I have dubbed “the racialist concept” which “is hierarchal“) as defined by Hardimon holds that “racialist race is the idea of a fundamental division between groups and individuals” (Hardimon, 2017: 17). I think that Hardimon strawmans the racialist concept as he as defined it, but that’s for another day.
He also says that the racialist concept “is closely associated with racism” while the terms racialism and racism are “sometimes used interchangeably” (Hardimon, 2017: 17).
His argument against the racialist concept of race (as he defines it) is as follows (Hardimon, 2017: 21):
A third line of argument starts from the idea that in order for racialist races to exist, certain things must be true of human genetics, namely the following:
(a) The fraction of human genetic diversity between populations must exceed the fraction of diversity between them.
(b) The fraction of human genetic diversity within populations must be small.
(c) The fraction of diversity between populations must be large.
(d) Most genes must be highly differentiated by race.
(e) The variation in genes that underlie obvious physical differences must be typical of the genome in general.
(f) There must be several important genetic differences between races apart from the genetic differences that underlie obvious physical differences.
Note: (b) says that racialist races are genetically racially homogeneous groups; (c)-(f) say that racialist races are distinguised by major biological differences.
Call (a)-(f) the racialist concept of race’s genetic profile.
Now that his argument against the racialist concept (as he defines it) is laid out, you can see why I said that I think he strawmans the racialist concept. But I’ll get into that another day.
He then cites Lewontin’s (1972) analysis of blood groups by race as evidence against the racialist concept. Lewontin found that 85.4 percent of total human variation fell within populations. He also found that populations that populations classically defined as human races (Caucasians, Africans, Mongoloids, South Asian Aborigines, American Indians, and Oceanians) accounted for 8.3 percent of total human variation. Total variation between the classically defined races accounted for 6.3 percent of the variance.
It’s worth noting that the numbers given by Lewontin are true; where he goes wrong is assuming that there is no taxonomic significance for race based on the data he got from his analysis. “Call this Lewontin’s cleaver,” writes Hardimon on page 22.
Then in 2002, 31 years after Lewontin published his analysis, A.W.F. Edwards published his paper Human Genetic Diversity: Lewontin’s Fallacy. (Edwards, 2003). In the paper, Edwards argues that Lewontin’s conclusion is incorrect. Edwards (2003: 800-801) writes in his conclusion (emphasis mine):
There is nothing wrong with Lewontin’s statistical analysis of variation, only with the belief that it is relevant to classification. It is not true that ‘‘racial classification is … of virtually no genetic or taxonomic significance’’. It is not true, as Nature claimed, that ‘‘two random individuals from any one group are almost as different as any two random individuals from the entire world’’, and it is not true, as the New Scientist claimed, that ‘‘two individuals are different because they are individuals, not because they belong to different races’’ and that ‘‘you can’t predict someone’s race by their genes’’. Such statements might only be true if all the characters studied were independent, which they are not.
Of course, Lewontin’s conclusion is fallacious because small genetic differences do not entail that racial classification that race has no taxonomic significance (Richard Dawkins accepts the taxonomic existence of race). As you can see from the quote from Edwards, he does not object to Lewontin’s analysis of the races, he objects to his conclusion—namely that races do not exist based on the within-race variation being greater than between-race variation.
On page 22-23, Hardimon writes about Edwards’ objection to Lewontin’s conclusion:
Lewontin’s locus-by-locus analysis (which does not consider the possibility of a correlation between individual loci) does not preclude the possibility that individual loci might be correlated in such a way that people could be grouped into traditional racial categories. The underlying thought is that racial classification would have “taxonomic significance” were it possible to group people into traditional racial categories by making use of correlations between individual loci. However, Lewontin’s argument that there are no racialist races because the component of within-race genetic variation is larger than the component of between-race variation is untouched by Edwards’s objection.
In 2002, Rosenberg et al, in their paper Genetic Structure of Human Populations confirmed Lewontin’s analysis. They looked at 377 autosomal loci in 1,056 individuals from 52 populations and found that within-population differences between major groups (Africa, Europe, Asia, the Middle East, Central and South Asia, East Asia, Oceania, and America) accounted for 3-5 percent of genetic variation while genetic differences between individuals accounted for 93-95 percent of genetic variation. So Rosenberg et al (2002) confirmed Lewontin’s (1972) analysis—though do recall that Lewontin’s conclusion is incorrect. According to Hardimon’s interpretation of the racialist concept of race, both Lewontin’s and Rosenberg et al’s analysis disprove the racialist concept of race, but that doesn’t mean that there is no scientific basis for the biological reality of race (Hardimon, 2012).
The minimalist concept of race
The minimalist concept of race is similar to the racialist concept, though there are some stark differences. It does not say that there are intrinsic differences between races—call them essences if you will), but it does say that you can distinguish races by patterns of different physical features such as skin color, hair type, nose shape, morphology, etc, which then correspond to differences in geographic ancestry in geographically, genetically isolated breeding populations.
The minimalist concept of race further states that (i) races are distinguised from other races by patterns of visible physical features; (ii) the members are linked by a common ancestry which is peculiar to members of the group; and (iii) this group must originate from a distinct location.
The minimalist concept of race does not require: that the fraction of human genetic diversity between minimalist races is larger than the fraction of diversity within them; it is compatible with within-race diversity being large and between-race diversity being small; it does not require most genes to be highly differentiated by race; it does not require the existence of a lot genetic differences between races that underlie more than the phenotypic differences already noticed; the concept does not imply that there can be predictions made from yet unstudied characteristics; it finally does not require any genetic differences between races other than those found in the genes that underlie differences in physical appearance between race. This is called the minimalist concept of biological race (Hardimon, 2017: 66) and it survives all objections from Lewontin’s and Rosenberg et al’s analysis of between-race genetic variation.
This is my favorite race concept, personally, because it covers any and all objections from the race-denialist crowd—people who deny any genetic differences between races—because the only genetic differences it counts on are those physical traits that are already noticed.
Hardimon (2017: 29) writes:
Such readers should feel free to regard the minimalist concept of race, that is, as a concept that, though in many respects similar to the ordinary concept, is nonetheless distinct from it. What I would insist on is that minimalist races (groups satisfying the minimalist concept of race) are *races* (that is races so properly called)—either because the minimalist concept of race just is the ordinary concept of race or because it captures enough of the ordinary concept of race for minimalist races to be counted as races. My view is that if it can be shown that minimalist races exist, races exist. And if it can be shown that *minimalist race* is real, race is real.
The populatonist concept of race
The populationist concept of race is a nonessentialist, non-hierarchical concept of race that slightly differs from the minimalist concept of race. The populationist concept of race can be said to be a scientific concept of race (as can the minimalist concept) because it characterizes races as groups belonging to different groups of biological descent, they are distinguished by patterns of phenotypic differences, and these phenotypic differences trace back to geographically separated and genetically isolated founding populations.
The populationist concept of race also holds that “A race is a subdivision of Homo sapiens—a group or population that exhibits a distinctive pattern of genetically transmitted phenotypic characters that corresponds to the group’s geographical ancestry and belongs to a biological line of descent initiated by a geographically separated and reproductively isolated founding population” (Hardimon, 2017: 99). So with these criteria, you can see that even if you do not accept the racialist concept of race (as Hardimon defines it), you can still be a race realist. The populationist concept is likely to exist, and if the populationist concept of race exists then race is real.
Defining race as geographically and reproductively isolated breeding populations that share a common line of biological descent with similar phenotypic characters is as barebones a concept of race as you can get—and it is perfectly in line with how most people view races on the basis of phenotypic characterization. The populationist concept of race supposes that numerous concepts from the racialist concept of race are true—but do not presuppose any to-be-studied differences between those races. The strength of the populationist argument, as you can see, is very strong and it holds up to numerous lines of criticism very well. Although both the populationist and minimalist race concepts do not presupposed any to-be-studied differences between races, this still is not good enough for race deniers.
It is clear that without even looking at the brain and physiological differences between races, that race does indeed exist and it does—contrary to popular belief—have implications for people’s health of certain races.
The socialrace concept of race
Finally, the last concept of race laid out by Hardimon is the concept of socialrace. The concept of socialrace takes a race to be a racialist race, it refers to a position that is occupied by a social group that is a socialrace, and the socialrace concept refers to the system of social positions that are socialraces. This concept of race is, clearly, different from the minimalist and populationist race concepts but does indeed correlate with popular notions of race (and would correlate with the minimalist and populationist concept of race very well). The socialrace concept is, basically, what is believed to be racialist races.
The concept of socialrace is a concept of race as a social group (Hardimon, The Ontology of Race: 31)
The socialrace concept differs from the minimalist and populationist concept of race in that it looks at so-called social—not biological—correlates of race. Though, still, the socialrace concept can be said to show the reality of race since how one socially defines themselves correlates almost perfectly with geographic ancestry (which is a prerequisite for the existence of the minimalist concept of race and the populationist concept of race) (Tang et al, 2005). They showed that self-identified racial categories lined up almost perfectly with geographic ancestry (99.86 percent of the time). So, as you can see, the concept of socialrace also gives credence to the existence of the minimalist and populationist concepts of race.
This concept of race—as its name implies—does not talk race is a biological manner, but a social one, as its name implies. However, due to the extremely high chance that one’s self-identified race (their socialrace) lines up with the geographic ancestry of the classical races, we can see that the socialrace concept further buttresses the argument for the existence for the reality of the minimalist concept of race and the populationist concept of race.
The socialrace concept is kind of like Templeton (2014) defines race: that human races exist in a cultural sense, but not biologic sense. I have shown, though, that races exist in a cultural, social, and biological sense with the arguments presented in this article. Socialrace, culturalrace, whatever you want to call it, it is evidence for the existence of race.
Race exists whether or not the racialist position of race (as Hardimon defines it) is true or not. The minimalist concept of race and populationist concept of race show that race is real while the concept of socialrace further lends credence to the biological models of the minimalist and populationist concept of race. Even still, people who deny race because the genetic distance between races is too small for their to be any meaningful differences between them do not accept that three arguments above (sans the racialist concept) for the existence of race. They’ll still talk about the genetic differences between them and, say, morphology, but the minimalist concept of race and the populationist concept of race define race in enough of a way that genetic differences do not need to be looked at—we can only look at reproductive isolation, morphology, geographic ancestry and physical differences between minimalist and populationist races such as hair, nose, and skin color along with morphological differences.
Minimalist and populationist races exist and are a biological reality. We can take those two concepts to be a scientific basis for race. While we can take the concept of socialrace not as a biological concept, but as a social concept and we can then say that socialrace is socially real while being a significant social reality. That social reality is manifested by noticing different racial phenotypes, along with differences in SES, educational attainment, etc, and placing different races in different average social positions, which would correlate with the concepts of race mentioned above. This also correlated nearly perfectly with geographic ancestry. So, I’m saying it again, the existence of race as a social reality is real; the existence of socialrace buttresses the arguments for both the existence of the minimalist concept of race and the populationist concept of race—both of which are scientific concepts of race.
Minimalist races exist, and is a superficial biological reality, populations races may exist and if they exist, they are a relatively superficial biological reality. Socialraces exist and are a social reality which also lend credence to the minimalist and populationist concepts. I personally am privy to the minimalist race concept because it is shown to be real, so race is real.
In sum, race exists whether you look at genetic differences between races or not, morphology, geographic ancestry, reproductive and genetic isolation are all you need to prove the existence of race. There is a scientific concept of race, and the minimalist and populationist race concepts provide the existence for it, while the socialrace concept does as well. It is clear that for a scientific concept of race, you only need phenotypic variation, morphologic variation between races,
(Also read the American Rennaisance review for the book, A Tactical Retreat for Race Denial. I think it is balanced and fairly written, though a bit biased and doesn’t account for Hardimon’s views well enough in my opinion.)