There may well be several folk theories of race, but many scholars (including me) believe that one of them is far more prevalent, fundamental, and destructive than the others. Scholars call it racial essentialism. Racial essentialism is a particular manifestation of the broader psychological tendency known as psychological essentialism, which is also a large part of the psychological foundation for dehumanization. So, before delving more deeply into the essentialistic folk theory of race, and its connection to dehumanization, it is important to be clear about exactly what psychological essentialism is.1
Psychologists Douglas Medin and Andrew Ortony introduced the term “psychological essentialism” in 1989 as a name for the psychological tendency to attribute essences to certain kinds of things.2 Saying that the word “essence” is ambiguous is an understatement. It has a wide range of meanings, from the quotidian to the esoteric. In everyday speech, something that is described as “essential” is central, important, or indispensable (as in “essential reading” or “essential amino acids”). In the more technical, scholarly arena, philosophers have been thinking about the notion of essence for at least two thousand years. Traditionally, they contrast essential properties with accidental ones. Accidental properties are attributes that a thing could lack while remaining the thing it is. For example, at this moment, your body consists of a specific number of cells, but you could have more or fewer while still remaining the person that you are. Suppose that right now your body consists of thirty-seven trillion cells, and that in the next fifteen seconds sixty-nine of them will die off. The fact that you will then have only 36,999,999,999,931 cells wouldn’t make you any less you. You would retain your essential properties and lose an accidental one. In contrast, essential properties are those that a thing cannot lack and still retain its identity. For example, being a mammal is one of your essential properties. You couldn’t be an oyster, or my left thumb, or the Brooklyn Bridge, and still be you. If a sorcerer were to wave a magic wand and turn you into the Brooklyn Bridge, you would immediately go out of existence.
Beyond this point, the philosophical hairsplitting takes over. There are very many more fine-grained conceptions of essences floating around in the philososphere which need not concern us here. Instead, I want to leave the esoterica behind and concentrate on the distinction between two quite different notions of essences that, if not differentiated, can lead to enormous confusion. One is the notion of “sortal essences.” Sortal essences are the defining properties of a thing—all of those features that make a thing the kind of thing it is. When Plato defined human beings as “featherless bipeds” (and Diogenes the Cynic crashed his seminar brandishing a plucked chicken and blurting out “There is Plato’s man!”), he was trying to specify the sortal essence of being human. Anything can have a sortal essence—even socially constructed things. For example, part of the sortal essence for Thursday is “the day after Wednesday.”
But only members of natural kinds have causal essences. They are supposed to be “deep,” fundamental properties that all and only members of the kind possess. The possession of a causal essence is supposed to be what makes an individual a member of a natural kind. These essences are imagined to permeate the insides of things that have them, mark off sharp boundaries between kinds, and remain constant in the face of superficial changes. Causal essences are theoretical entities. They are supposed to be unobservable but at the same time causally responsible for observable properties that are typical of members of the kind. In particular, essentialistic thinking is associated with the realm of living things, including human beings.3
According to psychologists who have studied our essentialist proclivities, we cannot help being drawn to thinking of the world as populated by sharply demarcated natural kinds, and we cannot help thinking of members of these natural kinds as having essences. This might seem questionable to you. You might sincerely say that you do not believe that species of organisms are natural kinds, or deny that animals have hidden essences that make them the kinds of animals that they are. This is an understandable response, but it does not comport with the usually tacit nature of essentialistic cognition. As the philosopher Sarah-Jane Leslie emphasizes, the essentialist mindset can be suppressed, but it probably cannot be overcome.4 Scientific education can block the tendency to explicitly think of the human genome as a hidden essence or to imagine that species are neatly demarcated kinds of living things. However, these tendencies do not vanish with education. They just go underground.
Causal essentialism might sound scientifically respectable, because theoretical science traffics in the “deep” properties of things that determine their outward form and behavior. Causal essentialism does make sense in the domains of physics and chemistry. Chemical elements have causal essences. The causal essence of oxygen is located in its microstructure. Everything that is oxygen possesses this kind of microstructure, and everything that possesses this kind of microstructure is oxygen. And oxygen’s microstructure accounts for its observable properties—its solubility in water, the temperature at which it condenses, its role in combustion, and so on. Essentialism is somewhat plausible in the chemical and physical domains, but it does not play out so nicely in biology.5
It is worthwhile to pause to consider just why it is that essentialism about species is inconsistent with what biological science tells us, because essentialistic biases have such a powerful and pervasive influence on how we view the world of living things. The essentialistic mindset is so difficult to resist that even those of us who are aware that it is a psychological bias that has no scientific support are likely to slip into it unless we are vigilant. The philosopher Paul Griffiths is one of a number of scholars who have explained why an essentialist picture of species is incompatible with the scientific perspective. He points out that the essentialist picture
is precisely the … perspective on species that Darwin had to displace in order to establish the gradual transformation of one species into another. Species are not types to which individual organisms more or less imperfectly conform, but abstractions from the pools of overlapping variations that constitute the actual populations of that species.… The limitations of folk taxonomy become apparent when working on larger geographical and temporal scales. Many species grade into one another spatially, and all do temporally. When individuals exist who are intermediate between two species due to hybridization or incomplete speciation it is senseless to ask whether these individuals are “really” of one species or the other. That question presumes that the species is more than an abstraction from the varied individuals that compose it.6
To illustrate: from an essentialist perspective, what makes an animal a North American porcupine is its possession of the porcupine essence. That essence is supposed to be something that all porcupines possess and no nonporcupines possess. Furthermore, the porcupine essence is supposed to be causally responsible for the observable features that are typical of porcupines, such as their greyish color, having four legs, being covered in sharp quills. The idea that there are sharply demarcated species, each with a unique essence, is supposed to explain why there are many different kinds of living things: porcupines are different from bluebirds because porcupines have the porcupine essence and bluebirds have the bluebird essence.
On the face of it, if the observable traits of members of a biological species are caused by their essence, and if every member of a species shares the same essence, shouldn’t every member of the species be exactly the same as all the others? How can the essentialism account for the fact that no two porcupines are exactly alike, and that some (for example, albino porcupines) depart dramatically from the species norm? It is here that a second component of the folk theory kicks in—one that is not much emphasized in the psychological literature but which is nevertheless very important for understanding both racial cognition and dehumanization. It is the notion of development. According to the essentialist folk theory, the development of any individual is internally driven. Development is just the gradual unfolding of an individual’s essence over time. From this perspective, if development always proceeded perfectly, without obstruction or deviation, then every member of a biological kind would fully embody all of attributes that are typical for members of that kind. If this were the case, then every porcupine would be a paradigmatic porcupine, possessing only and all of the attributes that porcupines ought to have. The point can be put this way: an individual is true to its kind to the extent that its development is unimpeded, and within-species deviations from the ideal developmental trajectory produce within-species variation.7 In very deviant cases, the appearance of an individual may depart so dramatically from what is characteristic of its kind that it is unrecognizable as a member of that kind. The folk theory allows that an animal might look and behave in ways that are unlike the appearance and behavior that is characteristic of porcupines generally, and yet still be a porcupine by virtue of possessing a porcupine essence.
There are three important additional aspects of the folk biological mindset that should be noted before we move on.8 The first is that essences press forward to express themselves, so even if an essence is not manifest, it will become manifest if the obstacles to its expression are weakened or removed. Cognitive anthropologist Dan Sperber aptly captures the idea as follows: “If an animal does not actually possess a feature ascribed to it by its definition, then it possesses it virtually: not in its appearance but in its nature.”9 The second concerns the normative dimension of essentialist assumptions: the belief that it is “natural” for individuals to embody their essence, because the essence fixes how an individual should be, and thus that “individuals who deviate from their natural state are malformed.”10 And third, possessing the essence of a certain natural kind is supposed to be absolute rather than incremental. To possess an essence is to possesses it fully. As psychologists Gil Diesendruck and Susan Gelman put the point,
All members of a category are believed to possess the category’s essential properties to the same degree and are therefore considered members of the category to the same extent. Members of a category may differ, however, in the typicality of their nonessential features (e.g., physical appearance) and therefore may vary in how good an example of the category they are. The essentialist account, then, attempts to capture the intuition that, for instance, although a Chihuahua and a German shepherd differ in how representative they are of the category dog, the former is as much a dog as the latter. More generally, the essentialist account argues that categorization is all-or-none: Items are judged absolutely as either members of their category or not members of their category.11
So, according to the folk theory, kind membership is absolute. An animal cannot be more or less of a porcupine: it must be wholly a porcupine or not a porcupine at all. Later on, in Chapter 12, I will show why this feature of essentialist thinking is crucial for explaining the distinctive phenomenology of dehumanization.
In past centuries, the essence of a being was often identified with its “soul.” The type of soul that a being has was supposed to determine the kind of being that it is. According to the Aristotelian framework, which was a dominant influence on Christian, Muslim, and Jewish philosophical and scientific thought, even plants and animals were thought to be ensouled. Plants have “vegetative” souls insofar as they grow and reproduce, and animals have this plus a “sensitive” soul that allows them to have sense perceptions. Human beings were distinguished by having a “rational” soul that plants and animals lack. Aristotelians believed that it is this that sets us apart from “lower” organisms and therefore is what makes us human. It is important to bear in mind that the Aristotelian conception of soul is not some sort of metaphysical organ—it is a form of life. From this perspective, saying that a being “has” a rational soul is just to say that it has access to rationality.
Although the Aristotelian theory of the soul might seem to be little more than navel-gazing by a Greek aristocrat with too much time on his hands, it had vast and terrible ramifications. Aristotle used his theory to justify the institution of slavery. Some people, he argued, are “slaves by nature” (a category that included virtually everyone other than high-born Greek males) because they possessed the capacity for rationality to only a very imperfect degree. Aristotle did not go as far as to say that barbarians lack rationality entirely—which would have excluded them from the category of the human and ranked them with nonhuman animals. Barbarians were human because they had rationality, albeit a defective, third-rate grade of rationality. Unlike nonhuman animals, which are utterly incapable of rationality, the barbarian “shares in reason to the extent of understanding it, but does not have it himself.”12 In other words, these quasi-brutes have it in them to be responsive to reason, but they do not have the wherewithal to initiate it—they lack the “deliberative” component of rationality.13 Aristotle also claimed that barbarians could participate, albeit indirectly, in a higher form of life by subordinating themselves to their Greek masters. As this is a good thing, it follows that Greeks should enslave barbarians, and that barbarians owe it to themselves to be enslaved by Greeks, because this allows them to participate in a higher form of life. As Malcolm Heath aptly puts it, Aristotle believed that the master is “a kind of cognitive prosthesis” for the slave.14
Aristotle distinguished natural slavery from slavery as a social institution. In his view, only those who are natural slaves should be enslaved. Similarly, he believed women are stunted, chronically underdeveloped human beings who should be subordinated to Greek men.15
It is clear from this summary that Aristotle viewed the social and biological world as having a hierarchical organization. Greek men were above Greek women, who were above barbarian men, who were (presumably) above barbarian women, who were above nonhuman animals.16 Aristotle’s opinions about natural slavery were hugely influential for more than two thousand years after his demise. The idea that whole groups of people are defective, and therefore benefit from being taken in hand by their superiors was very attractive to apologists for colonialism and slavery. Perhaps the best-known example is the exchange between the Dominican friar Bartolomé de las Casas and the jurist Juan Ginés de Sepúlveda in 1550, over the question of whether or not the indigenous peoples of the Spanish Empire in the New World were natural slaves, and therefore whether or not it was legitimate to enslave them. Echoing Aristotle but going one step further in excluding Indians from membership in the human family, Sepúlveda claimed that the “barbarians of the New World … are as inferior to the Spaniards as are children to adults and women to men. The difference between them is as great as … I am tempted to say, between men and monkeys.”17 North American apologists for slavery, such as South Carolina senator William Harper (1790–1847), also often cited Aristotle’s thesis. Harper, who was but one of many who enlisted Aristotle’s authority in support of this institution, wrote of Black people that “they approach nearer to the nature of the brute creation, than perhaps any other people on the face of the globe. Let me ask if this people do not furnish the very material out of which slaves ought to be made, and whether it be not an improving of their condition to make them the slaves of civilized masters?”18
For Aristotle, then, the soul was a form of life rather than a thing that people possess. But over time, the Aristotelian rational soul was replaced with the Christian conception of an immortal soul, an immaterial thing the presence or absence of which determined whether a being was human or not. Writings by seventeenth-century Anglican clergyman Morgan Godwyn make it clear that British colonists in the New World considered enslaved Africans to lack souls in this sense and therefore to be less than human. Godwyn wrote that he had been told “privately (and as it were in the dark) … That [sic] the Negros, though in their Figure they carry some resemblances of Manhood, yet are indeed no Men” and that they are “Unman’d and Unsoul’d; accounted and even ranked with Brutes”—“Creatures destitute of Souls, to be ranked among Brute Beasts, and treated accordingly.”19 Even during the nineteenth century, enslaved people in the United States were often told that they did not have souls. One former slave, interviewed in the late 1930s, recalled being told by a preacher that “only White people had souls and went to heaven” and that “niggers had no more soul than dogs.”20
The possession of a soul is less often invoked as the criterion for humanness in today’s secular societies. Instead, we have delegated this role to the genome, which has been charged with a pop-metaphysical significance that extends far beyond anything that science tells us about the biological role of DNA.21 As sociologists Dorothy Nelkin and M. Susan Lindee observe,
Spiritual imagery sets the tone for popular accounts of DNA, fueling narratives of genetic essentialism and giving mystical powers to a molecular structure. Indeed, DNA has assumed a cultural meaning similar to that of the Biblical soul. It has become a sacred entity, a way to explore fundamental questions about human life, to define the essence of human existence, and to imagine immortality.… Just as the Christian soul has provided an archetypal concept through which to understand the person and the continuity of self, so DNA appears in popular culture as a soul-like entity.… The genome appears as a “solid” and immutable structure that can mark the borders and police the boundaries between humans and animals, man and machine, self and other, “them” and “us.”22
In light of its conceptual role as an essence-bearer, it is not surprising that DNA is often called upon to validate everyday notions of race. If genes are what make people members of one race or another, could it be that racial essentialism has been vindicated by science?
People who do not know the basic science, but who have the vague conviction that race must somehow reside in our DNA, tend to think that there are genes “for” race. This is the idea that there are sharp discontinuities in our genetic makeup that map perfectly or nearly perfectly onto the conventional racial partitions of the human family. To put the point as crudely as it deserves to be put, it is the notion that there are “White” genes, “Black” genes, “Asian” genes, and so on, that make it the case that one belongs to one or another of these races. This assumption, which simply transposes racial essentialism onto the genome, has zero scientific credibility.
Although the idea that racial divisions correspond to real, biological categories had begun to fade away earlier in the twentieth century, most notably in response to the horrors resulting from the Nazis’ brand of scientific racism, the decisive event in its decline was work by Harvard geneticist Richard Lewontin. In his 1972 paper “The Apportionment of Human Diversity,” Lewontin argued on empirical grounds that the genetic variation that exists between individual members of a given race far outstrips, on average, the genetic variation between members of different races. In other words, racial differences represent only a small proportion of the differences between people and are therefore not very significant biologically.
Let me flesh this argument out a bit. All human beings are practically identical at the genetic level. However, there are some places or “loci” on our genome that can be occupied by different “versions” of a gene. These variants are called “alleles.” Variation among alleles accounts for many of our individual differences. For example, the fact that one person has straight hair and another person has curly hair results from the presence of different alleles at the loci that control hair texture. Obviously, some larger-scale differences between whole groups of people are likely to correspond to differences in allele distribution: alleles that produce curly hair will be far more frequent in populations where almost everyone has hair of that sort than in populations where few people have curly hair (this should not be taken to imply that variations in hair texture are always brought about by the same alleles, or that the relevant alleles do not depend on the presence of other alleles for producing this effect).
Lewontin’s famous study considered allelic variation at only seventeen loci. However, with improved technology it has become possible to perform much more ambitious analyses of between-group genetic variation that consider much larger numbers of loci. When several alleles frequently occur together in a population, this is called a “cluster.” Genetic researchers use a computer program called STRUCTURE (because it analyzes what biologists call “population structure”) to discover such clusters in aggregated genetic data. This involves getting STRUCTURE to divide the data into the number of clusters specified beforehand by the researchers. STRUCTURE does not search for alleles that are unique to any group of people (as in the naïve essentialist picture of genes for race). Rather, it uses statistical facts about allele frequencies to distinguish between population clusters. You can think of clusters as defining the average genotype of the whole group rather than the actual genotype of individual members of the group.
Race apologists have been heartened by the fact that if you sample genetic data from all over the world and instruct STRUCTURE to segment the data into five clusters, it geographically groups individuals in a way that sort of matches the division of humankind into five races: Africa, the Americas, Oceania, East Asia, and Europe-Middle East-Central Asia. However, the claim that STRUCTURE scientifically vindicates the biological reality of race is questionable. Scientists and scholars have raised several telling objections to this interpretation of what STRUCTURE reveals about genetic diversity among human populations, and have argued against the view that population clusters are races.23
These arguments turn on empirical or methodological issues. But the claim that statistical facts about populations can underwrite the reality of race can also be faulted on conceptual grounds.
There are some characteristics that only individuals can have. An example is baldness. Individual people can be bald, but groups of people cannot be bald. To be bald, a thing has to have a head. Individuals have heads, but groups do not. Suppose that a bunch of bald people created a club called the Slaphead Club. Although it would be true to say that every member of the Slaphead Club is bald, it would be false to say that the Slaphead Club is bald. Because it does not have a head, a group that is composed entirely of bald people cannot itself be bald. Speaking loosely, one might refer to this group as “the bald club,” but it is really the members of the club that are bald, not the club. To attribute the properties of the parts (baldness) to the whole (the club) is to commit what is known as the fallacy of composition: the fallacy of thinking that what is true of the parts of a thing must also be true of the thing that they are parts of.
There are also characteristics of groups that are not features of the individuals that constitute those groups. To attribute these sorts of properties to individual members is to commit what is called the fallacy of division—the mirror image of the fallacy of composition. Here is an example. According to the United States Census Bureau, the median age for American women in the year 2012 was 38.1 years. But it would be wrong to conclude from this that any individual woman has the median age of 38.1 years, because individual women do not have median ages. Constructs like “median age” give us information about aggregates of individuals, but they do not apply to the individuals that make up the aggregate.
To claim that a club is bald, or that an individual woman has a median age, is to commit a category mistake, as described in Chapter 1. Likewise, saying that measurements of allele frequencies vindicate folk conceptions of race rests on a category mistake, albeit a subtle one. To see why, the place to start is to consider whether people think of race as a property of individuals, a property of groups, or a property of both. Obviously, race is supposed to be a feature of individuals, otherwise it wouldn’t make any sense to say that Barack Obama was the first Black president of the United States.24 “Races,” in the collective sense, are groups consisting of all and only those individual people who are of a certain race, but the races themselves—the aggregates of individuals that constitute the race—do not have racial properties (the White race is the sum of individuals with the property of being White, but the White race does not have the property of being White any more than the Slaphead Club has the property of being bald).
Allele frequencies are, like the median age of American women, statistical properties of whole populations rather than properties of individuals. So, although facts about different allele frequencies in different populations might provide the basis for a scientific conception of race (I say “might” because there are other reasons to doubt that this can be done), genetic clusters cannot be equated with ordinary conceptions of race, and racial essentialists cannot look to genetic studies of population structure for support.
Racial essentialism is a special case of psychological essentialism and conforms to the same general pattern as essentialism about species. To the extent that one is a racial essentialist, one believes that there is a hidden racial essence, the possession of which is what makes a person belong to a certain race and is also responsible for producing observable, race-related features of human beings. The racial essence imagined to be possessed by all and only Black people is supposed to be what makes such people Black, and is also supposed to cause such people to have the physical features that are typically used to identify members of the Black race—skin color, hair texture, facial morphology, and so on. However, according to essentialist thinking, it is possible for a person to be raced without ever manifesting the appearance or behavior that is associated with that race. Their racial essence is latent. They “have it in them” even though they do not express it, or express it fully. For example, Jews are supposed by anti-Semites to be essentially greedy, deceptive, and exploitative. Jews who do not behave in these ways are nevertheless imagined to be disposed to do so. It does not matter that a Jew behaves generously, honestly, and compassionately. This is merely a façade that conceals the Jewish person’s true nature. From this perspective, the less a Jew conforms to anti-Semitic stereotypes, the more suspicious one should be of them, because this might lead you to falsely believe that he really is a decent human being. This way of thinking makes racist stereotypes irrefutable. Once person becomes entrenched in such beliefs, there is no possible evidence to convince them that Black people are not violent, Latinos are not feckless, and Roma are not thieves because any evidence that one might offer to show that these assumptions are false pertains only to their manifest behavior and not to their (by definition, unobservable and unalterable) essential nature.