Race in the US is tricky. On one hand, we socially construct races. On the other, these socially constructed races have biological underpinnings. Racial constructivists, though, argue that even though biological races are false, races have come into existence—and continue to exist—due to human culture and human decisions (see the SEP). Sound arguments exist for the existence of biological races. Biological races exist, and they are real. One extremely strong view is from philosopher of science Quayshawn Spencer. In his paper A Radical Solution to the Race Problem, Spencer (2014) argues that biological races are real; that the term “race” directly refers; that race denotes proper names, not kinds; and these sets of human populations denoted by Americans can be denoted as a partition of human populations which Spencer (2014) calls “the Blumenbach partition”.
To begin, Spencer (2014) defines “referent”: “If, by using appropriate evidential methods (e.g., controlled experiments), one finds that a term t has a logically inconsistent set of identifying conditions but a robust extension, then it is appropriate to identify the meaning
of t as just its referent.” What he means is that the word “race” is just a referent, which means that the term “race” lies in what points out in the world. So, what “race” points out in the world becomes clear if we look at how Americans define “race”.
Spencer (2014) assumes that “race” in America is the “national meaning” of race. That is, the US meaning of race is just the referent to the Census definitions of race, since race-talk in America is tied to the US Census. But the US Census Bureau defers to the Office of Management and Budget (OMB). Therefore, since the US Census Bureau defers to the OMB on matters of race, and since Americans defer to the US Census Bureau, then Americans use the OMB definitions of race.
The OMB describes a “comprehensive set” of categories (according to the OMB) which lead Spencer (2014) to believe that the OMB statements on race are pinpointing Caucasians, Africans, Pacific Islanders, East Asians, and Amerindians. Spencer (2014: 1028-29) thusly claims that race in America “is a term that rigidly designates a particular set of “population groups.” Now, of course, the question is this: are these population groups socially constructed? Do they really exist? Are the populations identified arbitrary? Of course, the answer is that they identify a biologically real set of population groups.
To prove the existence of his Blumenbachian populations, Spencer (2014) invokes populational genetic analyses. Population geneticists first must make the assumption at how many local populations exist in the target species. According to Spencer, “The current estimate for humans is 7,105 ethnic groups, half of which are in Africa and New Guinea.” After the assumptions are made, the next step is to sample the species’ estimated local populations. Then they must test noncoding DNA sequences. Finally, they must attempt to partition the sample so that each partition at each level is unique which then minimizes genetic differences in parts and maximizes genetic differences among parts. There are two ways of doing this: using structure and PCA. For the purposes of this argument, Spencer (2014) chooses structure, invoking a 5-population racial model, (see e.g., Rosenberg et al, 2002).
K = 5 corresponds to 5 populational clusters which denote Africans, Oceanians, East Asians, Amerindians, and Caucasians (Spencer, 2014; Hardimon, 2017b). K = 5 shows that the populations in question are genetically structured—that is, meaningfully demarcated on the basis of genetic markers and only genetic markers. Thus, that the populations in question are meaningfully demarcated on the basis of genetic markers, this is evidence that Hardimon’s (2017b) minimalist races are a biological reality. Furthermore, since Rosenberg et al (2002) used microsatellite markers in their analysis, this is a nonarbitrary way of constructing genetic clusters which then demarcate the continental-level minimalist races (Hardimon, 2017b: 90).
Thus, Spencer (2014) argues to call the partition identified in K = 5 “the Blumenbachian partition” in honor of Johann Blumenbach, anthropologist, physician, physiologist, and naturalist. (Though it should be noted that one of his races “Malays” was not a race, but Oceaninans are, so he “roughly discovered” the population partition.) So we can say that “the Blumenbach partition” is just the US meaning of “race”, the partitions identified by K = 5 (Rosenberg et al, 2002).
Furthermore, like Lewontin (1972), Rosenberg et al (2002) found that a majority of human genetic variation is between individuals, not races. That is, Rosenberg et al (2002) found that only 4.3 percent of human genetic variation was found to lie between the continental-level minimalist races. Thus, minimalist races are a biological kind, “if only a modest one” (Hardimon, 2017b: 91). Thus, Rosenberg et al (2002) support the contention that minimalist races exist and are a biological reality since a fraction of human population variation is due to differences among continental-level minimalist races (Africans, Caucasians, East Asians, Oceanians, and Amerindians). The old canard is true, there really is more genetic variation within races than between them, but, as can be seen, that does not rail against the reality of race, since that small amount of genetic variation shows that humanity is meaningfully clustered in a genetic sense.
Spencer (2014: 1032) then argues why Blumenbachian populations are “race” in the American sense:
It is not hard to generate accessible possible worlds that support the claim that US race terms are just aliases for Blumenbachian populations. For example, imagine a possible world τ where human history unfolded exactly how it did in our world except that every Caucasian in τ was killed by an infectious disease in the year 2013. Presumably, we have access to τ, since it violates no logical, metaphysical, or scientific principles. Then, given that we use ‘white’ in its national American meaning in our world, and given that we use ‘Caucasian’ in its Blumenbachian meaning in our world, it is fair to say that both ‘Caucasian’ and ‘white’ are empty terms in τ in 2014—which makes perfect sense if ‘white’ is just an alias for Caucasians. It is counterfactual evidence like this that strongly suggests that the US meaning of ‘race’ is just the Blumenbach partition.
Contrary to critics, this partition is biologically real and demarcates the five genetically structured populations of the human race. Rosenberg et al (2005) found that if sufficient data are used, “the geographic distribution of the sampled individuals has little effect on the analysis“, while their results verify that genetic clusters “arise from genuine features of the underlying pattern of human genetic variation, rather than as artifacts of uneven sampling along continuous gradients of allele frequencies.”
Some may claim that K = 5 is “arbitrary”, however, constructing genetic clusters using microsatellites is nonarbitrary (Hardimon, 2017b: 90):
Constructing genetic clusters using microsatellites constitutes a nonarbitrary way of demarcating the boundaries of continental-level minimalist races. And the fact that it is possible to construct genetic clusters corresponding to continental-level minimalist races in a nonarbitrary way is itself a reason for thinking that minimalist race is biologically real 62.
It should also be noted that Hardimon writes in note 62 (2017b: 197):
Just to be perfectly clear, I don’t think that the results of the 2002 Rosenberg article bear on the question: Do minimalist races exist? That’s a question that has to be answered separately. In my view, the fundamental question in the philosophy of race on which the results of this study bear is whether minimalist race is biologically real. My contention is that they indicate that minimalist race (or more precisely, continental-level minimalist race) is biologically real if sub-Saharan Africans, Caucasians, East Asians, Amerindians, and Oceanians constitute minimalist races.
Sub-Saharan Africans, Caucasians, East Asians, Amerindians, and Oceanians constitute minimalist races, therefore race is a biological reality. We can pinpoint them on the basis of patterns of visible physical features; these visible physical features correspond to geographic ancestry; this satisfies the criteria for minimalist races; therefore race exists. Race exists as a biological kind.
Furthermore, if these five populations that Rosenberg et al (2002) identified (the Blumenbachian populations) are minimalist races, then minimalist race is “a minor principle of human genetic structure” (Hardimon, 2017b: 92). Since minimalist races constitute a dimension within the small amount of human genetic variation that is captured between the continental-level minimalist races (4.3 percent), then it is completely possible to talk meaningfully about the racial structure of human genetic variation which consists of the human genetic variation which corresponds to continental-level minimalist races.
Thus, the US meaning of race is just a referent; the US meaning of race refers to a particular set of human populations; races in the US are classically-defined races (Amerindian, Caucasian, African, East Asian, and Oceanians; the Blumenbach partition); and race is both a biological reality as well as socially constructed. These populations are biologically real; if these populations are biologically real, then it stands to reason that biological racial realism is true (Hardimon, 2012 2013, 2017a; 2017b; Spencer, 2014, 2015).
Human races exist, in a minimalist biological sense, and there are 5 human races. Defenders of Rushton’s work—who believed there are only 3 primary races: Caucasoids, Mongoloids, and Negroids (while Amerindians and others were thrown into the “Mongoloid race” and Pacific Islanders being grouped with the “Negroid race” (Rushton, 1988, 1997; see also Liberman, 2001 for a critique of Rushton’s tri-racial views)—are forced into a tri-racial theory, since he used this tri-racial theory as the basis for his, now defunct, r/K selection theory. The tri-racial theory, that there are three primary races of man—Caucasoid, Mongoloid, and Negroid—has fallen out of favor with anthropologists for decades. But what we can see from new findings in population genetics since the sequencing of the human genome, however, is that human populations cluster into five populations and these five populations are races, therefore biological racial realism is true.
Biological racial realism (the fact that race exists as a biological reality) is true, however, just like with Hardimon’s minimalist races, they do not denote “superiority”, “inferiority” for one race over another. Most importantly, Blumenbachian populations do not denote those terms because the genetic evidence that is used to support the Blumenbachian partition use noncoding DNA. (It should also be noted that the terms “superior” and “inferior” are nonsensical, when used outside of their anatomic contexts. The head is the most superior part of the human body, the feet are the most inferior part of the human body. This is the only time these terms make sense, thus, using the terms outside of this context makes no sense.)
It is worth noting that, while Hardimon’s and Spencer’s views on race are similar, there are some differences between their views. Spencer sees “race” as a referent, while Hardimon argues that race has a set descriptive meaning on the basis of C (1)-(3); (C1) that, as a group, is distinguished from other groups of human beings by patterns of visible physical features, (C2) whose members are linked be a common ancestry peculiar to members of that group, and (C3) that originates from a distinctive geographic location” (Hardimon, 2017b: 31). Whether or not one prefers Blumenbachian partitions or minimalist races depends on whether or not one prefers race in a descriptive sense (i.e., Hardimon’s minimalist races) or if the term race in America is a referent to the US Census discourse, which means that “race” refers to the OMB definitions which then denote Blumenbachian partitions.
Hardimon also takes minimalist races to be a biological kind, while Spencer takes them to be a proper name for a set of population groups. Both of these differing viewpoints regarding race, while similar, are different in that one is describing a kind, while the other describes a proper name for a population group; these two views regarding population genetics from these two philosophers are similar, they are talking about the same things and hold the same deflationary views regarding race. They are talking about how race is seen in everyday life and where people get their definitions of “race” from and how they then integrate it into their everyday lives.
“Race” in America is a proper name for a set of human population groups, the five population groups identified by K = 5. Americans defer to the US Census Bureau on race, who defers to the Office of Management and Budget to define race. They hold that races are a “set”, and these “sets” are Oceanians, Caucasians, East Asians, Amerindians, and Africans. Race, thusly, refers to a set of population groups; “race” is not a “kind”, but a proper name for known populational groups. K = 5 then shows us that the demarcated clusters correspond to continental-level minimalist races, what is termed “the Blumenbach partition.” This partition is “race” in the US sense of the term, and it is a biological reality, therefore, like Hardimon’s minimalist races, the Blumenbach partition identifies what we in America know to be race. (It’s worth noting that, obviously, the Blumenbach partition/minimalist races are one in the same, Spencer is a deflationary realist regarding race, just like Hardimon.)