Introduction

Race is a complex and volatile topic. Slavery, genocide, and all kinds of violent acts have been—and continue to be—done in the name of “race.” Those of us who work on this topic should proceed carefully and use clear language. Such care need not be taken in the name of political correctness. Just as we handle dangerous goods with care, we should be careful with dangerous ideas, such as race.

I take the above to be obvious. However, I am going to argue that it is in tension with a common goal among race theorists, which is to convey to their students and to the public that race is a ‘social construct.’ In this paper I will argue that ‘social constructionism about race’ is so ambiguous that it is not meaningful, that some versions of the view are false, that the true versions are either overly broad or better described as anti-realism about race, and that any practical usefulness this label may have had has been outlived. The “social constructionist consensus about race”—which I will show is no consensus at all—ought to be rejected in favor of a version of anti-realist reconstructionism about race, which says that there are no races, only racialized groups (Miles 1989; Small 1994; Blum 2002; Darder and Torres 2003; Hochman 2017, 2019b; Glasgow 2019; Mavundla 2019). This view plays the roles most social constructionists want their view to play, and it does not suffer from social constructionism’s ambiguity problems. It also has the benefit—the not insignificant benefit—of being true.

‘Social constructionism about race’ is a highly ambiguous phrase. Social constructionism has been endorsed by those who do not believe that there are biological races (e.g., Gannett 2004; Hochman 2013a), by those who do believe that there are biological races (e.g. Outlaw 1992; Kitcher 1999), by those who believe that there are social races (e.g. Sundstrom 2002; Diaz-Leon 2015), and even—as we will see—by those who believe that there are innate psychological differences between the “races” (e.g. Reich 2018a; Carl 2019).

If ‘social constructionism’ can mean almost anything, then—without further clarification—it means almost nothing. Or rather, we might say that it means whatever the reader believes it to mean, which could be any number of things. In this paper, I argue that the social constructionist consensus about race that has formed in most corners of academia (at least in the English-speaking world) is an illusion. Social constructionism about race cannot operate as a meaningful consensus view. Too many positions fall under this label, some of which are mutually exclusive. I will discuss four versions of the view, arguing that two of these are false, that one is true but overly broad, and that the other is true but better described as anti-realism about biological race.

It may be most useful to think of social constructionism about race in terms of the rhetorical roles that it plays. Many scholars assume that social constructionism about race stands in contrast with biological racial realism, so adopting the label signals a rejection of race as a biological category. I argue that biological racial realism is compatible with all but one version of social constructionism about race, and that social constructionism is therefore poorly suited to one of its intended roles.

Another role that social constructionists have wanted their view to play is to stand in opposition to racial hereditarianism, the study of putatively innate psychological differences between human biological “races.” This may have been the only common thread holding social constructionists together. However, Harvard geneticist David Reich has recently endorsed both social constructionism and racial hereditarianism. Reich accepts that “It is true that race is a social construct” but argues that geneticists should fearlessly study “average genetic differences among ‘races’” anyway, including “the genetic influences on behavior and cognition” (Reich 2018a).

Reich is a leading figure in what he calls the ancient DNA revolution. While behavioral genetics is not his area of expertise, he wants to connect his work reconstructing human prehistory to this field. It is unsurprising that, as Jeffrey Long notes in his review, Reich deals with this area “with less objectivity and thoroughness than he applies to his primary research” (2017, 304). It is surprising, however, that he simultaneously endorses racial hereditarianism and social constructionism about race. This combination will strike many as strange, if not utterly confused. Social constructionism is fundamentally opposed to hereditarianism, right? Many social constructionists would like this to be true. However, I will argue that most forms of social constructionism about race are consistent with racial hereditarianism. There is one version of social constructionism about race that is not, but this is the version I argue is better described as anti-realism about biological race. Those who want a position about “race” which is at odds with racial hereditarianism should not endorse social constructionism about race, but, rather, racial anti-realism.

The fact that social constructionism about race and racial hereditarianism are compatible is a problem for those social constructionists who want their view to reflect their rejection of racial hereditarianism. And some racial hereditarians may not like it either, as they see social constructionism about race as a symptom of political correctness “gone mad.” However, some racial hereditarians, like Reich, may want to align themselves with social constructionism about race, given that it is the hegemonic “position” about race.

Indeed, Reich’s work has been welcomed by racial hereditarians, who only complain that he puts ‘race’ in scare quotes (Wade 2018). But does he give a convincing endorsement of racial hereditarianism? I will argue that he does not. Reich justifies hereditarian research by erecting and then knocking down a strawman. This strawman is a scientific “orthodoxy” against research into population differences. I argue that no such orthodoxy exists. Instead, hereditarianism is in tension with a developmentalist tradition that questions the value of heritability measures for understanding developmental outcomes (see Tabery 2009 and citations within). Reich does not engage with this tradition and thus overlooks one of the primary reasons that scientists are skeptical of hereditarian research.

Racial hereditarianism also relies on the existence of human biological races. If there are no races, then “race” is not a valid category of investigation and “race” cannot be the cause of anything. Racialization can be a cause, and racism can be a cause, but “race” cannot. I draw on Reich’s own work to throw the validity of biological racial realism—and consequentially racial hereditarianism—into question.

Empirical issues aside, there are serious ethical issues surrounding hereditarian research. Reich is sensitive to these issues, and offers what he calls a “rational framework” for undertaking this kind of work. However, I show that the implementation of this framework renders hereditarian research redundant.

For these reasons, I will argue that Reich’s defense of racial hereditarianism fails. However, it offers a striking example of how social constructionism about race is ill-suited to perform the rhetorical role that almost all card-carrying constructionists want it to play: to act as an alternative to racial hereditarianism. Critics of racial hereditarianism should be worried about this. The hegemonic status of social constructionism about race, combined with its ambiguity, makes it ripe for exploitation by those who want to give racial hereditarianism an appearance of respectability.

Social constructionism about race, four ways

In my first two publications on “race” I self-identified as a social constructionist (Hochman 2013a, b). I even went so far as to suggest that my critique of the new wave of biological racial realism helped to show how to be a social constructionist in the post-genomic era. I later realized that researchers both within and outside of philosophy mean radically different things by ‘social constructionism about race.’ In this section, I outline four major versions of social constructionism about race, without assuming that they are exhaustive (for instance, I do not cover Ásta’s conferralist version of social constructionism; see Ásta 2018). My aim is to show how the social constructionist consensus isn’t meaningful because of the sheer ambiguity surrounding the phrase.

  1. (1)

    Social constructionism about race as a view about social influences on racial classification

Our first version of social constructionism about race is the one defended by Reich, and it is an intuitive application of the construction metaphor. As Reich explains, “race is a ‘social construct’, a way of categorizing people that changes over time and across countries” (2018a; see also Carl 2019). Underlying this view is the assumption that racial classification is a social product—a product of human actions and intentions.

  1. (2)

    Social constructionism about race as an anti-realist position about biological race

According to Bence Nanay, “Antirealism about race often takes the form of a version of social constructivism: the claim that race is merely a social construct” (Nanay 2010, 256). Anti-realism about biological race in the human context is simply the view that there are no human biological races. It does not deny the existence of human biological diversity, only that this diversity is well-described as racial. As Robin Andreasen explains, there is a “widely-held presumption that social constructivism is always an antirealist thesis” (2000, 653). It was under this presumption that I endorsed social constructionism about race. I was also influenced by Lisa Gannett’s definition. She writes that, “Biological race is a socially-constructed category. The races biologists once claimed to have discovered in nature were, in actuality, the illegitimate offspring of an invented classification scheme they had imposed on nature” (2004, 323). On this account, social constructionism about race claims that race is a biological illusion.

  1. (3)

    Social constructionism about race as a realist view about biological race

This version of social constructionism traces back to Lucius Outlaw, who writes that “human groups, though historical, socially constructed realities, also have natural histories, and this makes for particularly thorny conceptual challenges” (Outlaw 1992, 446). Philip Kitcher takes up these challenges, arguing that “races are both socially constructed and biologically real. Biological reality intrudes in the objective facts of patterns of reproduction, specifically in the greater propensity for mating with other ‘blacks’ (or other ‘whites’ respectively); the social construction lies in the fact that these propensities themselves have complex social causes” (1999, 106). Kitcher is using the idea of social constructionism in a literal way here, arguing that social factors—such as racism—cause biological races to stay in existence. This form of social constructionism has also been endorsed by Quayshawn Spencer, who argues that “some entities are biologically real exactly because they are socially constructed” (2015, 52).

  1. (4)

    Social constructionism about race as a realist view about race as a social category

The final version of social constructionism I will cover is the view that race is a social kind. Ronald Sundstrom, for example, defends “an ontology of race as a constructed and real category: a real human kind” (2002, 91). In a similar vein, Charles Mills writes that “race is a contingently deep reality that structures our particular social universe, having a social objectivity and causal significance that arises out of our particular history” (1998, 48). According to Esa Diaz-Leon, “Social constructivism about races holds that races are socially real, that is, that races should be identified with socially constructed properties, or social kinds” (Diaz-Leon 2015, 547).

‘Social constructionism about race’ is used to describe a wide range of views about “race”, some of which are mutually exclusive. The first tells us that racial classification is a product of human actions and intentions. The second tells us that biological races do not exist. The third tells us that biological races are real, but literally constructed through social forces. The fourth tells us that social races are real. These four social constructionist positions run the gamut of possible metaphysical positions that one could hold about race. No wonder Lawrence Blum writes that when it comes to ‘race,’ “the language of ‘social construction’ seems to me too fraught with confusion to recommend” (2010, 304).

If all four of the views discussed above count as social constructionism, then the fact that most race scholars are social constructionists doesn’t tell us that there is a genuine consensus. In fact, it suggests that we are just calling a range of different views by the same name.

Will the real social constructionists please stand up?

A skeptical reader might argue that social constructionism is not as hopelessly ambiguous as I suggest. They might claim that some of these versions of social constructionism about race are not really social constructionist positions. However, as I will show in this section, all of the four views above are consistent with what is perhaps the most influential and often-quoted definition in the philosophical literature, the one offered by Ian Hacking.

Hacking explains that social constructionists believe:

  1. (1)

    X need not have existed, or need not be at all as it is. X, or X as it is at present, is not determined by the nature of things; it is not inevitable.

Very often they go further, and urge that:

  1. (2)

    X is quite bad as it is.

  2. (3)

    We would be much better off if X were done away with, or at least radically transformed. (Hacking 1999, 6)

The reader may wonder, “why choose Hacking’s approach to social constructionism, rather than, say, Sally Haslanger’s?” However, Haslanger believes that constructionism “is many different things, and the discourse of social construction functions differently in different contexts…” (2012, 113). Rather than challenging Hacking’s criteria, she introduces what she calls constitutive construction. This contrasts with Hacking’s focus on causal construction. The former has to do with “the degree to which the kind in question is defined by ‘identification’ with the social position” and the latter with “the degree to which explicit classification is a causal factor in bringing about the features that make for membership in the kind” (Haslanger 2012, 128). So Haslanger does not offer a competing definition of social constructionism, but rather adds to Hacking’s account. Indeed, constitutive constructionism fits well with Hacking’s criteria, quoted above.

Hacking’s criteria are only a part of his overall account of social constructionism. He also distinguishes, for instance, between object constructionism and idea constructionism. The distinction is intuitive: the former is about the construction of things, the latter is about ideas about things. This is a useful distinction for understanding social constructionism about race, because ‘race’ is used to refer to postulated objects—“races”—and to ideas—“race” as a system of classification.

Now that we have criteria for what counts as social constructionism, we can ask whether the self-titled social constructionists discussed in the last section endorse constructionism according to these criteria.

Let’s begin with the first version of social constructionism, defended by Reich, according to which race is socially constructed because it is a way of categorizing people that changes over time and between contexts. This is indeed a form of constructionism about race according to Hacking’s criteria. It is idea constructionism. Reich defends Hacking’s (1), (2), and (3) in relation to racial classification. He writes about “the inconsistent definition of ‘black’”, noting that:

In the United States, people tend to be called “black” if they have sub-Saharan African ancestry—even if it is a small fraction and even if their skin color is very light. In Great Britain, “black” tends to mean anyone with sub-Saharan African ancestry who also has dark skin. In Brazil, the definition is different yet again: a person is only “black” if he or she is entirely African in ancestry. (Reich 2018b, 249)

Clearly, Reich (1) does not think that racial classification is determined by the nature of things. Confusingly, Reich warns against racial classification in his book but uses it in his opinion pieces (a tension to which we will return). In the book he writes, in accord with Hacking’s (2), that “race vocabulary is too ill-defined and too loaded with historical baggage to be helpful” (Reich 2018b, 253). When Angela Saini interviews Reich for her book Superior: The Return of Race Science, she quotes him as saying that “‘Latinos’ is a crazy category that encompasses groups with different ancestry mixes” (Saini 2019, 121–22). If Reich thinks that conventional racial classification is “crazy,” it is reasonable to infer that he also believes (3) that racial classification should be transformed.

Using Hacking’s definition of social constructionism, the first version of social constructionism is genuine. What about the second version of social constructionism, which says that race is a biological illusion? This is a form of object constructionism: it says that the object “biological race” does not exist. This clearly meets Hacking’s criteria, as it says (1) that racial classification is not determined by the nature of things; typically, (2) that racial classification has been destructive; and (3) that we ought to either eliminate it or replace it with some other form of classification, such as classification in terms of ancestry or racialized group.

Recall that the third version of social constructionism says that “Races might quite literally be social constructs, in that our patterns of acculturation maintain the genetic distinctiveness of different racial groups” (Kitcher 1999, 246–47). This is a form of object constructionism that says that biological races do exist. This version also meets Hacking’s criteria. Kitcher, for instance, denies (1) the inevitability of race, claims (2) that the continued existence of biological races is bad because it is a product of racism, and—insofar as this is true—promotes (3) the unification of the species through a policy of “Netflix and admix”—although he doesn’t use those exact words.

The fourth version of social constructionism claims that race is a social kind. This is also a form of object constructionism, where the object is social. It also fits Hacking’s criteria. On this view, racial classification (1) need not have been invented, but was the product of (2) malevolent social forces such as White supremacist ideology (Mills 1998). Social constructionists who endorse this version of the view tend to be critical of eliminativist positions, at least in the short term, but they do want to (3) shake the foundations of “racial formation” to the core.

As I have shown, all four versions of social constructionism about race meet Hacking’s criteria. Of course, one might disagree with Hacking’s criteria for what counts as social constructionism and advance others. Alternative criteria might entail that some of the versions of social constructionism about race discussed above are not genuinely constructionist views. However, the fact would remain that scholars occupying all of the major metaphysical positions about race are touting their views as forms of social constructionism. So, even if the social construction metaphor were being misused, that very misuse would be problematic and confusing. If social constructionism is meant to be an exclusive club, it needs some new bouncers.

The compatibility of social constructionism about race and biological racial realism

I can imagine readers who are still not convinced that these positions are all forms of social constructionism. As Andreasen explains, “Although there are many ways of being a constructivist about race, most constructivists assume that their view is incompatible with the idea that races are biologically real” (1998, 201).Footnote 1 A skeptical reader could argue that social constructionism about race is a more coherent “ism” than I suggest because it is fundamentally opposed to biological racial realism. However, in this section I argue that only one version of social constructionism about race precludes biological racial realism.

Consider the first version of social constructionism about race, which says that racial classification is dependent on the social context. This version is quiet on the metaphysics of race. It does not tell us whether or not human biological races exist, only that racial classification is affected by social factors. Indeed, it leaves open the possibility that there is a biological race-classification system that genuinely carves nature at its joints.

While the first version of social constructionism about race is compatible with biological racial realism, the second is not. This is unsurprising, as its central claim is that biological races do not exist.

What about the third version of social constructionism about race? This version is not only compatible with biological racial realism, it is a version of biological racial realism, because its central claim is that biological races are literally socially constructed.

Let’s move to the fourth and final version of social constructionism about race, the social kind version. Most social kind theorists about race reject race as a biological category. However, there is no shortcut from “race is a social kind” to “race is not a natural kind.” One must argue for both. As Hardimon observes, “There is no tension between conceiving of race as socialrace and conceiving of it as biological race” (2017, 162; see also Pierce 2014, 28). In other words, the possible existence of social races does not preclude the possible existence of biological races. Such races might map perfectly to each other, imperfectly, or not at all. The point is that both sorts of races could exist.

Because of this compatibility, I have argued elsewhere that social constructionists who believe that race is a social kind should—like Hardimon—talk about “social race” rather than “race” simpliciter (Hochman 2017, 66). Otherwise, they are likely to be misinterpreted as referring to “biological race”, given the history of race as a biological category. There is a danger that social constructionists who do not believe in the existence of biological races could spread belief in their existence by talking about “race” in contexts where people will interpret the term as referring to a biological category (see Miles 1988).

In sum, there is no easy way to separate social constructionism about race from biological racial realism. Most social constructionists believe that to “say that race is a social construction is at least to say that it is not a biological kind, a ‘natural kind’” (Blum 2010, 304). This is the central message most social constructionists want their view to convey. However, most versions of social constructionism are compatible with biological racial realism. The social constructionist club does not discriminate based on metaphysical views.

Is race really socially constructed?

So far, I have shown that the social constructionist consensus about race is an illusion, and is not suited to function as an alternative to biological racial realism. The illusion of a consensus is protective. Given that social constructionism is adopted by those defending all of the major metaphysical positions about race, social constructionism is unfalsifiable. It is easy to be the correct position when you are every position.

Nevertheless, we can evaluate each view individually. Because of space constraints, I will only be able to do this in a cursory way. I will argue, briefly, that only two of the four constructionist views are correct, and that while these two correct versions can be described as social constructionist, they are better described differently.

  1. (1)

    Social constructionism about race as a view about social influences on racial classification

Let’s begin with the first version of social constructionism about race, endorsed by Reich. Recall that according to this version “race is….a way of categorizing people that changes over time and across countries” (Reich 2018a). This is, of course, accurate. As Michael Root has observed, “Some men who are black in New Orleans now would have been octoroons there some years ago or would be white in Brazil today” (2000, 631–32). If we read this simply as a statement about how one person can be racialized differently in different times and places, rather than about the metaphysical possibility of changing “races,” then this version of social constructionism about race is obviously true.

  1. (2)

    Social constructionism about race as an anti-realist position about biological race

This is another version of social constructionism that I believe to be true. I have argued extensively against the existence of human biological races on the following basis: we are not a very genetically diverse species; our biological traits are predominantly smooth in their distributions across geographic space; there are no major human lineages; and there is nothing scientifically privileged about conventional “racial” classification (Hochman 2013a, b, 2014, 2016, 2019a, 2021; see also Atkin 2017; Templeton 2013; Maglo et al. 2016).

In my view, attempts to revive race as a biological category tend to make one of three mistakes. They are empirically false (e.g., Andreasen 1998; see Templeton 2013); they change the definition of race to such an extent so as to change the topic or to trivialize it (e.g. Pigliucci and Kaplan 2003; Sesardic 2013; see Hochman 2019a); or they appeal to and rely upon folk understandings of “race” in order to solve what is essentially a scientific problem (e.g. Spencer 2014; Glasgow and Woodward 2015; Hardimon 2017; see Hochman 2014).

  1. (3)

    Social constructionism about race as a realist view about biological race

Recall that this is the view that social factors, such as racism, literally construct biological races through how they affect human mating patterns. Is this version of social constructionism about race true? If there were human biological races, then it would be likely that their maintenance would be due to social forces, because social factors such as racism influence partner choice. However, if race fails as a biological category (as I suggest above and argue further below) then this form of social constructionism is false—it fails to get off the ground.

  1. (4)

    Social constructionism about race as a realist view about race as a social category

This is the dominant view in the philosophical literature, so I will spend a little longer on it. Some have critiqued social kind approaches to race on a semantic basis. They argue that race just is a biological concept, or that it would be too confusing to the public to suggest otherwise (Atkin 2012; Blum 2010; Glasgow 2009; see also Mallon 2004). I am sympathetic with these arguments, but I believe that a stronger argument can be made. I believe that race does not qualify as a social kind: that there are no social races, only groups misunderstood to be biological races—racialized groups (Hochman 2017, 2019b).

For the social kind theorist, “race” is made real by social properties and relations. The risk of a social kind approach to race is that it could reify race, giving it a false appearance of reality. There is no question that groups have been racialized, so there is nothing wrong with talking about racialized groups. However, constructionists need to offer a positive account of the social properties and relations that constitute “race” in order to show that it is a social kind. Being misunderstood as belonging to a biological kind is not, by itself, a social property or relation. More needs to be shown.

Drawing together a range of work by social constructionists, I have elsewhere suggested that, “From a social constructionist perspective, race could thus be defined as a social kind, distinguished on the basis of real or imagined differences, which is used to differentially distribute power and privilege between groups who—as a result of their classification—have shared histories, experiences, and opportunities” (Hochman 2017, 68). However, this definition produces false positives: unracialized ethnic and religious groups would count as races, as would children and adults, men and women, and gay people, under certain circumstances. Social kind approaches to race appear to suffer from an inflation problem: when race is defined in terms of social properties and relations, it loses its conceptual specificity, and as a consequence, its value as a concept (Hochman 2017).

In response to my argument, Phila Msimang has introduced the minimalist account of social race: “A social race”, he explains “shares with other social kind accounts some basic characteristics. These are the broad essentializations of a group (e.g., stereotyping); the purported inability of individuals to transition between groups in a defined social context (i.e., the fixation of immutability); the purportedly inherent genealogical heritability of group belonging (e.g., inheritance in the form of biologicization, autochthony, or a generational curse from God, etc.)” (Msimang 2019, 15).

This is exactly the right sort of response, and it does help. Using Msimang’s criteria of stereotyping, fixity, and genealogy we can see why children and adults are not races (they don’t fit the genealogy or the fixity conditions) and why men and women are not races (they don’t fit the genealogy condition or, as many would argue, the fixity condition). However, Msimang’s minimalist account of social race still produces false positives. There are unracialized ethnic and religious groups that fit Msimang’s three conditions. Consider also homosexuality. Homosexuality has historically been stereotyped and understood to be both fixed and heritable. A social kind approach to race is surely subject to a reductio ad absurdum argument if it allows there to be a “gay race.” While I do not have space to develop my argument here, I believe that we should reject the social kind version of social constructionism about race because it has a range of highly counterintuitive, and even some absurd, consequences (see Hochman 2017, 2019b).

If my arguments are good, then only two of the four versions of social constructionism about race are correct. If we look at the true versions, (1) borders on triviality. As Naomi Zack explains, “Anything that is the result of human interaction and intention in contexts where past actions, decisions, and agreements have present consequences is, trivially, a social construction” (2002, 106). If everything that is produced by or otherwise affected by human interactions and intentions is a social construct, then almost everything is socially constructed. For example, my greyhound Pegasus would be a social construct, because she was the product of the intentions of those who bred her to be a fast runner. This is not a very useful definition of social constructionism, because it lets too much in.

The other true form of social constructionism about race, (2), is the one I used to endorse. I stopped, first, because I didn’t want to be misunderstood to be endorsing other versions of social constructionism, and second because I realized that it is more accurately described as anti-realism about biological race. So, while the anti-realist version of social constructionism is a genuine form of social constructionism, for the sake of clarity we ought to call it anti-realism about biological race instead.

The compatibility of social constructionism about race and racial hereditarianism

Little seems to unite social constructionists about race. Until recently, one might have been able to argue that social constructionists were united in their rejection of racial hereditarianism. However, if this was once the thread that bound social constructionists together, it is no longer. While some race theorists have argued that “race” is both a biological kind and a social construct (e.g. Outlaw 1992; Kitcher 1999; Andreasen 2000; Spencer 2015; Hardimon 2017), Reich (2018a) goes a step further and defends hereditarian investigation into cognitive and behavioral differences between the so-called “races.”

Reich’s combination of social constructionism about race and racial hereditarianism is unusual. Are these a coherent set of positions to endorse? In this section I will argue that most versions of social constructionism about race are indeed compatible with racial hereditarianism.

We can immediately discount the second version of social constructionism about race, which I argue is better described as anti-realism about biological race. Racial hereditarianism depends on the existence of biological races. If races don’t exist, then “they” are not open to scientific investigation.

However, as I have argued above, the other three versions of social constructionism about race are consistent with biological racial realism. Racial hereditarianism is just biological racial realism plus the claim that there are “innate” psychological differences between “the races.” Is there any conceptual obstacle in the way of social constructionists who want to make the extra hereditarian claim?

I do not believe that there is. The first version of social constructionism about race says that racial classification is context dependent: it is quiet not only on the metaphysics of race, but also the possibility of heritable psychological differences between putative biological races. Skipping to the third version of social constructionism, it tells us that biological races exist, but it does not tell us what they are like, so it too is compatible with racial hereditarianism. The fourth version tells us that there are social races, and is quiet on the possible existence of biological races and what they would be like if they existed. It is also compatible with racial hereditarianism.

So, most versions of social constructionism about race are not only compatible with biological racial realism, they are also compatible with racial hereditarianism. The fact that Reich endorses racial hereditarianism and social constructionism about race is striking because it is widely assumed that the two are incompatible. However, if we distinguish anti-realism about biological race from social constructionism about race, racial hereditarianism and social constructionism are perfectly compatible.

Social constructionism is not only ambiguous and, in some of its forms, mistaken—it also cannot perform one of its traditional rhetorical functions: to act as an alternative to racial hereditarianism. Reich’s unusual pairing of social constructionism with racial hereditarianism indicates that the power of the constructionist metaphor to keep racial hereditarianism at bay may be dwindling.

How we got here: Reich on “the orthodoxy”

I have argued that Reich’s version of social constructionism about race is one of the true versions of the idea. I have also argued that it is consistent with racial hereditarianism. This may be bad news for social constructionists who want their view to act as an alternative to racial hereditarianism, but it is good news for racial hereditarians who want to show that their view, while unpopular, is consistent with the hegemonic position in race theory.

In this paper, I am arguing that social constructionism cannot perform one of its historically useful functions: to signal the rejection of racial hereditarianism. However, this assumes that racial hereditarianism is mistaken, or otherwise problematic. While a full critique of racial hereditarianism is beyond the scope of this paper, in the sections leading to the conclusion I will argue that Reich’s defense of racial hereditarianism fails.

In this section, I discuss Reich’s belief that there is an anti-hereditarian orthodoxy that rejects any research into population differences. I argue that there is no such orthodoxy, but instead a dominant developmentalist tradition that is highly critical of hereditarian research. Ignoring this tradition is Reich’s first mistake in his defense of hereditarianism.

Reich worries that the “consensus view of many anthropologists and geneticists [that race is a biological illusion] has morphed, seemingly without questioning, into an orthodoxy that the biological differences among human populations are so modest that they should in practice be ignored—and moreover, because the issues are so fraught, that study of biological differences among populations should be avoided if at all possible” (2018b, 250). However, this is an “orthodoxy” without an orthodox. As K. Ann Horsburgh writes in her review of the book, “I do not know any anthropologist or geneticist who believes this. And neither does the author, or he would have cited them” (2018, 656).

Reich insists that “We cannot deny the existence of substantial average genetic differences across populations, not just in traits such as skin color, but also in bodily dimensions, the ability to efficiently digest starch or milk” and so on, but as far as I can tell, nobody is denying these things (2018b, 255). Reich is “worried that people who deny the possibility of substantial biological differences among populations across a range of traits are digging themselves into an indefensible position”, but he appears to be the one doing all of the digging (2018b, 254).

Reich’s talk about physical differences between populations is a stepping stone to more controversial territory. “If selection on height and infant head circumference can occur within a couple of thousand years”, he suggests, “it seems a bad bet to argue that there cannot be similar average differences in cognitive or behavioral traits” (Reich 2018b, 258).

While the so-called orthodoxy that Reich critiques in the book is about differences in populations, vaguely defined, in the opinion piece Reich writes that “The orthodoxy maintains that the average genetic differences among people grouped according to today’s racial terms are so trivial when it comes to any meaningful biological traits that those differences can be ignored” (2018a). Against this “orthodoxy” Reich defends racial hereditarianism.

But the orthodoxy is a strawman. Anti-realists about biological race acknowledge the existence of geographically structured human biological diversity. We just argue that there is not enough of it—and that it is not distributed in such a way—to justify racial classification (see Hochman 2016). No anti-realist about race denies, for instance, that people racialized as Black have darker skin, on average, than those racialized as White. One can accept this while maintaining that race fails as a scientific category.

Perhaps the orthodoxy that Reich is reaching for—an orthodoxy that actually has an orthodox—has to do with the “consensus in most fields (e.g., philosophy of biology, evolutionary biology, psychology and behavioral genetics) that heritability measures…only have a very limited use” (Downes 2017). However, Reich does not engage with this literature, instead reinforcing the reductionist and empirically refuted idea that “the human genome…provides all the information that a fertilized human egg needs to develop” (Reich 2018b, xxiv). As Horsburgh explains, Reich’s “metaphor of genes as blueprints is powerfully seductive but wholly inaccurate. Genes do not specify organismal morphology or behavior. They are merely one factor in a dynamic developmental system in which feedback loops both affect and are affected by the rest of the system” (Horsburgh 2018, 656; see also Oyama et al. 2001; Griffiths and Hochman 2015).

Reich’s failed defense of race science

Racial hereditarianism assumes biological racial realism. If there are no biological races, then it is unscientific to ask whether “biological races” differ in psychological traits. In this section, I show that when Reich tries to offer positive support for race science, he fails to offer any relevant evidence. Without biological racial realism, racial hereditarianism suffers from a lack of construct validity.

Reich insists that “as a geneticist I….know that it is simply no longer possible to ignore average genetic differences among ‘races’” (Reich 2018a). Speaking “as a geneticist” is like speaking “as a mother”: it is intended to lend the speaker a certain air of authority. The problem, of course, is that just as not all mothers think alike, not all geneticists agree about the validity of human racial classification (Lieberman et al. 2003; Morning 2011).

Reich attempts to support his view about “average genetic differences among people grouped according to today’s racial terms” using examples of groups that are not grouped according to today’s racial terms at all (Reich 2018a). For example, he mentions that “northern Europeans are taller on average than southern Europeans.” He also discusses a study about Icelanders and his own work, in which he and his team found that a certain region of the genome “contained at least seven independent risk factors for prostate cancer, all more common in West Africans.” Northern and southern European, West African, Icelandic: none of these are racialized groups. If Reich believes that we ought to conceptualize these as racial categories, he fails to make this belief explicit, let alone defend it.

Not only does Reich fail to offer evidence for race science, Reich actually provides some resources for refuting the racial hereditarian position that he promotes to the public. Consider Reich on the reconstruction of human evolutionary history. One way to define race biologically is as a synonym for ‘subspecies’ in the lineage sense of the term (Andreasen 1998). On this view, races—if they exist—would be branches on the tree of life. However, as Reich argues:

while a tree is a good analogy for the relationships among species—because species rarely interbreed and so like real tree limbs are not expected to grow back together after they branch—it is a dangerous analogy for human populations. The genome revolution has taught us that great mixtures of highly divergent populations have occurred repeatedly. Instead of a tree, a better metaphor may be a trellis, branching and remixing far back into the past. (2018b, 81)

This is exactly what Alan Templeton (2013) argues in his empirical refutation of the lineage approach to race (see also Spencer 2018).

The closest Reich comes to endorsing biological racial realism in the book is his discussion of twenty-first century genetic clustering studies, which show that it is possible to cluster genotypes from certain populations around the world into groups that roughly resemble conventional racial categories (Rosenberg et al. 2002). Many have interpreted this discovery as support for biological racial realism (e.g. Sesardic 2013; Spencer 2014; Hardimon 2017). According to Reich, “nonrandom sampling could account for some of the effects…observed [in clustering studies]. However, later work proved that nonrandom sampling could not account for most of the structure, as substantial clustering of human populations is observed even when repeating analyses on geographically more evenly distributed sets of samples” (2018b, 252).

This is a major misunderstanding. Reich is referring to an article by Noah Rosenberg and colleagues, in which it is argued that “allele frequency differences generally increase gradually with geographic distance [but] small discontinuities occur as geographic barriers are crossed, allowing clusters to be produced” (2005, 661). A later study showed that around 75% of human genetic diversity (from isolated populations) is gradually distributed across geographical space and that, “Adding information on genetic clusters to this model captures only an extra ∼2% of the variance” (Handley et al. 2007, 435). So, when Reich writes that “most of the structure” is clustered, he gets things the wrong way around: almost all of the structure is smooth.

Despite this misunderstanding in favor of biological racial realism, Reich is still skeptical of race in the book. Responding to Duana Fullwiley’s (2008) critique that geneticists such as Reich use terms like “ancestry” as a euphemism for “race”, Reich writes that:

“ancestry” is not a euphemism, nor is it synonymous with “race.” Instead, the term is born of an urgent need to come up with a precise language to discuss genetic differences among people at a time when scientific developments have finally provided the tools to detect them. It is now undeniable that there are nontrivial average genetic differences across populations in multiple traits, and the race vocabulary is too ill-defined and too loaded with historical baggage to be helpful. (2018b, 253)

Reich is right that while terms like ‘ancestry’ can be used as euphemistic synonyms for ‘race’, they should be used, instead, as genuine alternatives. But while in his book Reich defends himself against the charge that he uses terms such as ‘ancestry’ and ‘population’ as euphemistic synonyms for ‘race’, in his opinion pieces he seems guilty as charged.

Racial hereditarianism relies on the existence of human biological races. But Reich does much to throw biological racial realism in question, and when he attempts to justify biological racial realism he appeals to his own authority “as a geneticist” and to the existence of population differences between groups that are not “grouped according to today’s racial terms” or argued to be races. So, he does not offer a convincing defense of biological racial realism. As a consequence, he does not offer a convincing defense of racial hereditarianism.

Reich’s “rational framework” for hereditarian research

It is little wonder that scientists are generally wary of hereditarianism, a research tradition that has been used to justify racism, sexism, homophobia, slavery, colonialism, and genocide (Smith 2020). Reich is sensitive to ethical concerns about hereditarian research, but he believes that heritable population differences in cognitive and behavioral traits exist and that we need to prepare for their discovery:

Even if we do not yet know what the differences are, we should prepare our science and our society to be able to deal with the reality of differences instead of sticking our heads in the sand and pretending that differences cannot be discovered. If as scientists we willfully abstain from laying out a rational framework for discussing human differences, we will leave a vacuum that will be filled by pseudoscience, an outcome that is far worse than anything we could achieve by talking openly. (Reich 2018b, 258)

So, Reich accuses scholars of denying evolved psychological differences between populations but at the same time admits that “we do not yet know what the differences are.” In this section, I will briefly discuss Reich’s “rational framework for discussing human differences” and argue that it renders hereditarian research redundant.

What is Reich’s recommended framework? “The right way to deal with the inevitable discovery of substantial differences across populations”, he explains,

is to realize that their existence should not affect the way we conduct ourselves. As a society we should commit to according everyone equal rights despite the differences that exist among individuals. If we aspire to treat all individuals with respect regardless of the extraordinary differences that exist among individuals within a population, it should not be so much more of an effort to accommodate the smaller but still significant average differences across populations. (Reich 2018b, 265)

It is unclear whether Reich believes that research into these topics should be done or whether well-meaning scientists should be prepared for when it is inevitably done. He does, after all, write about “the onslaught of science” (2018b, 254). But what would be the point of research into evolved population differences in cognitive traits if “their existence should not affect the way we conduct ourselves”? Perhaps Reich believes that such research should be done because “racist pictures of the world” are “in conflict with the lessons of the genetic data” (Reich 2018b, xxvi). But, yet again, if we follow Reich’s framework, according to which population differences should not matter, there is no reason to ask whether racist pictures of the world are accurate. Some may disagree with Reich’s framework, but this is a logical outcome of its application. In trying to lay out a “rational framework” for racial hereditarianism Reich ends up giving a reason for abandoning the research program altogether.

Conclusion

I have argued that the idea that there is a meaningful constructionist consensus about race is an illusion. There are so many views that can legitimately be called social constructionism about race that the label is almost meaningless. Moreover, I’ve argued that one version of social constructionism is overly broad, that another version is better described as anti-realism about biological race, and that the other two versions are false.

Social constructionism about race is ill-suited to the useful roles that it used to play. Contrary to popular belief, most versions of social constructionism are consistent with both biological racial realism and racial hereditarianism. If we distinguish anti-realism about biological race from social constructionism about race—as I have argued we should—social constructionism is perfectly compatible with biological racial realism and racial hereditarianism.

Reich’s defense of racial hereditarianism fails. However, he is right to assume that social constructionism about race and racial hereditarianism are compatible. And when we consider the ambiguity of the social constructionist metaphor, this is not so surprising. Given the respectability of social constructionism about race and its looseness of meaning, I would be unsurprised if more racial hereditarians began aligning themselves with the constructionist metaphor.

Indeed, since the publication of Reich’s opinion piece in The New York Times, another scholar has endorsed both social constructionism and racial hereditarianism. Noah Carl, who lost his Toby Jackman Newton Trust Research Fellowship at the University of Cambridge after he was accused of conducting racist pseudo-science, writes that “just because the meaning of ‘race’ has varied over time and across societies, this does not mean that it is a wholly social construct. Indeed, there is a strong case to be made for treating race as a partly biological construct too” (2019, 265). Because of the ambiguity of the constructionist metaphor, it was always a possibility that racial hereditarians could endorse social constructionism about race. This possibility is now being exploited.

I would like to see a new consensus position emerge: that while race is an illusion, racialized groups are real. This view, which is a version of what Joshua Glasgow (2009) calls anti-realist reconstructionism about race, is appealing for a number of reasons. Unlike social constructionism about race, it is at odds with racial hereditarianism, which suffers from a range of conceptual and empirical problems and, as Reich himself recognizes, is risky because it has been “used in the past to try to justify the slave trade, the eugenics movement to sterilize the disabled as biologically defective, and the Nazis’ murder of six million Jews” (2018b, 250). And unlike unreconstructed anti-realism about race, it offers a basis for group recognition and solidarity, while at the same time highlighting the historicity of racial classification. In other words, it does all of the things that most social constructionists want their view to do, but unlike social constructionism, it can fulfil its promise.