One of the reasons that my colleague Jordan Peterson has become such a celebrity is that so many of his critics are so confused. On more than one occasion, he has come out of debates looking like the guy who brought a gun to a knife fight (if one can excuse the metaphor). One area in which this is particularly apparent is in his various discussions of social constructivism, some of which have a “shooting fish in a barrel” quality. This is largely because so many people – both academics and activists – are really confused about what it means to say that something is “socially constructed,” and what the political implications of this are.

As a philosopher and a critical theorist, I feel some responsibility for this, because those of us who trade in these concepts for a living have not done a good enough job at saying what we mean. In particular, we have allowed the impression to develop that, if some characteristic or property, such as gender or race, is “social constructed,” that means that it can be easily changed. This idea is what has created the tight connection, in many people’s minds, between “being committed to social justice” and “thinking that everything is socially constructed” (or “wanting to believe that everything is socially constructed”). This has, in turn, generated considerable distortion in the social sciences, as academics endorse the social constructivist thesis, not on the basis of compelling evidence, but because they want to believe it for political reasons (see my earlier post on “The Problem of Normative Sociology”).

As a matter of fact, a lot of things that are socially constructed cannot easily be changed. To explain this, however, we need to get to the basics of what the social constructivist claim is, and how it became so politically charged. It has commonly been observed that a great deal of misery in the world is caused by how people treat one another. Those who hope to alleviate some of this misery are therefore drawn to the question, whether these behaviour patterns can be changed. In some cases, the answer is obviously positive – if the behaviour is learned from others, for instance, it can be changed just by not teaching it, or if it is imposed by social norms, we can stop enforcing those norms. In other cases, however, the answer may well be negative, since there are aspects of human nature that appear to be constitutional, or “biologically determined.” If your plan for utopia involves people refraining from showing partiality toward their own children, for instance, it is unlikely to succeed.

This is, of course, what the great nature-nurture debate of the 20^th century was about – basically, how much of the “human nature” we see around us is learned, and how much is innate? The ultimate outcome of this dispute was widespread acceptance of the conclusion that “everything is both,” or more precisely, that there is no principled way of drawing a distinction between the “biological” and the “social,” because human development occurs in an environment (which includes other people), in which there is constant feedback between the developmental plan and environmental conditions. So, for instance, humans obviously have certain biological adaptations aimed at language-learning and speech (physical adaptations, such as the position of our larynx, as well as cognitive adaptations, that allow for rapid domain-specific learning). For anyone who interacts with an 18-month old, it is difficult to avoid the impression that human brains are “hardwired” for language learning. And yet languages are not biological, they are cultural artifacts and a part of the social environment. Furthermore, they contain obvious conventional elements (e.g. the vocabulary, aspects of the grammar), such that children growing up in different social environments learn mutually unintelligible languages. And, of course, children raised in isolation, or under conditions of extreme neglect, do not spontaneously recreate language (the way that, for instance, spiders raised in isolation are nevertheless capable of spinning complex webs).

This resolution to the nature-nurture debate unfortunately left both sides a bit too contented, because the “nature” side interpreted “everything is both” to mean, in practice, “we can treat everything as biological,” while the “nurture” side interpreted it to mean “we can treat everything as social.” The correct conclusion, by contrast, would have been that there is a continuum, with some things being more biological and some things more social. Or to put it somewhat more technically, some trajectories of human development are more “canalized” than others, and thus more resistant to social or environmental modification. (For instance, like every other mammal on the planet, our biological endowment gives us certain “innate” food preferences and aversions. These are, however, not as rigid as in many other mammals, which means that they can be modified through social learning and habituation. We have “acquired tastes.” And yet there are some tastes that are more difficult to acquire than others. Coprophagia, for instance, is normally taken as a symptom of mental illness, not self-transcendence. So cultures that invest a great deal of socializing energy on cultivating acquired tastes will wind up with a palette that is more or less “constructed,” whereas those that do not will wind up with one that closely resembles our biological default settings.)

It is, of course, a scientific question how canalized the development of a particular trait is, and thus, how much socializing effort would be required in order to divert that trajectory. Unfortunately, there is a powerful political motive that interferes, for many people, with the impartial adjudication of this question. Whenever one has a trait that is involved, in one way or another, with some injustice, many people would like to believe that it is more “social” than “biological,” precisely because they would like to see it changed. This has had numerous consequences, perhaps the more obvious of which is the outrageous misallocation of the burden of proof that prevails in the social sciences, but also in everyday public debate, where proponents of “biological” explanations, in order to vindicate their claims, must rule out every possible “social” explanation, but where proponents of “social” explanations typically feel no obligation even to present evidence for their claims, much less rule out all possible “biological” explanations. For instance, when it comes to observed gender differences in children, biological explanations are ordinarily taken to be defeated by purely hypothetical accounts of how the observed difference could be the result of socialization. (For an example of this from the philosophical literature, see Jesse Prinz’s book Beyond Human Nature. Philosophers are specially trained to think up weird hypothetical examples, and as a result, find it particularly easy to persuade themselves that our biology is completely irrelevant to every aspect of life. This is, incidentally, seldom a step in the right direction on the path to self-knowledge.)

The resulting state of affairs is one in which the typical belief among educated persons dramatically overestimates the importance of social influences (learning, institutions, etc.) on human behaviour, based on an elaborate form of politically motivated wishful thinking. This gives rise to “politically correct” science, which Peterson has had something of field day pointing out and debunking. For instance, he first drew widespread attention by getting into a tussle over the extension of the Canadian Human Rights code to prohibit, not just sex discrimination, but also discrimination based on “gender identity.” Although people routinely say things like, “gender identity is different from, and is not determined by, biological sex,” it is worth pausing to ask what the term “determined” means in this sentence.

Social scientists are often impressed by correlations in the neighbourhood of .3 or .4. In the case of gender, over 99% of the population has a gender identity that maps directly onto their biological sex. If that isn’t what it means for something to be “determined by” biology, Peterson asks, then what is? Biological determinism cannot mean that a particular genotype invariably, in 100% percent of cases, produces a given phenotype. For instance, the fact that we have five fingers on each hand is something that we take to be “biologically determined,” despite the fact that there is a developmental aspect to it, such that some people wind up with six fingers, while others get one or more cut off, and so wind up with fewer. So the implicit definition of biological determination being used in the sentence above (which requires a correlation of 1) is one according to which nothing is biologically determined. This serves to distract attention from the obvious fact that there is an extremely powerful relationship between sex and gender.

At the same time, there is a perfectly coherent sense in which “gender” is a social construct, while “sex” is not. When Simone de Beauvoir said “on ne nait pas femme, on le devient,” what she was observing was that our popular understanding of what “women” and “men” are like involves a great deal of learned behavior. Typically, what happens in childhood is that biological sex is used as a way of partitioning the human population into two classes (“male” and “female”), who are then socialized in different ways, to occupy two distinct social roles (“men” and “women”). Girls may be taught to cross their legs when they sit, boys will not, as a result of which men and women wind up behaving differently, but this is properly explained as a consequence of the gender role, not their biological sex. The fact that the behaviour can easily be changed or unlearned strongly suggests that it is more due to social role than biological endowment.

The distinction between sex and gender is worth making in part because the very strong correlation between the two categories leads many people to assume that aspects of the social role are biologically fixed. (In other words, gender becomes invisible, as people appeal to biological sex to explain various aspects of social behavior that are actually due to gender roles. Precisely because of the near-perfect correlation between sex and gender, the error is not detected, because one so seldom encounters any counterexamples.) It is also worth observing that some people are born with a biology that makes it difficult to classify them as “male” or “female,” and so, historically, have been somewhat arbitrarily assigned to one gender or the other, while other people experience dissatisfaction with their assigned gender, and so have sought to switch roles, despite lack of ambiguity in the underlying biology.

The big question, of course, is whether, or to what extent, the content of the roles can be changed. Can we change the expectations? Can we create new roles? Can we abolish them entirely? Many people mistakenly assume that, if the social constructivist thesis is correct, then the answer is “yes.” Proving that something is not “developmentally canalized” shows that we can do whatever we want – and if there’s any resistance to change, it must not be coming from human nature, but rather from “conservative” or “reactionary” elements in society.

To see the problem with this, consider an analogy to religion. Religion is obviously cultural, and it is also, obviously, made up. (If there is one thing that everyone – both theists and atheists – can agree upon, it is that religious beliefs are generally false. The only difference between theists and atheists on this point is that theists allow one exception to this generalization (in the case of their own beliefs) while atheists make none.) In any case, from the fact that religions are made up, one might conclude that it would be easy to make up a new one. And indeed, many people have tried. What they discover, however, is that while working up a new “revelation,” or developing a set of new ritual practices, is not that difficult, securing their widespread acceptance is almost impossible.

Social diffusion, in other words, is the tricky part. This is partly because human attention and memory are all subject to powerful biases, of biological origin, that make certain kinds of stories more compelling than others. (Pascal Boyer, in his Religion Explained, describes some fascinating experiments, in which they tried making up new “superstitions” and propagating them, then sought to figure out which qualities made some of them “stick,” while others were ignored or quickly forgotten.)

There is a similar dynamic in children’s books. There are literally thousands of these published every year (people spend so much time reading these simple books to their kids, everyone at some point thinks “I could write one of these.”) But it’s a competitive market out there. Only a handful are actually successful. The system that picks them out is essentially one of natural selection. What is the environment? Human cognition, and in particular, the protean biases of the unsocialized human brain. That’s why “Freudian” readings of children’s books are so persuasive. It’s not because this is what the authors secretly had in mind. It’s because these themes are what cause certain stories to resonate, to be remembered, and to be retold.

This is, incidentally, Peterson’s reason for thinking that the great world religions, and religious texts, contain deep truths, which we have not yet succeeded in fully articulating. It’s that certain stories have resonance, they appeal to us at the deep biological level, which is what caused them to become successful, and to be reproduced over time. We don’t actually know what it is about these stories that causes them to resonate so powerfully, so there is still much that we can learn about ourselves by exploring them. (I’m inclined rather to think they involve ‘misfiring’ learning heuristics adapted to other contexts, rather than esoteric wisdom. Just goes to show that one man’s deep truth is another man’s arbitrary cognitive bias.)

All of this creates some rather significant constraints on how easily social institutions can be changed. For instance, the “family” is in some respects an extremely flexible social institution. It exists in very different configurations in different societies, it can be extended quite arbitrarily through practices like adoption, and so on. And yet, social reformers have been trying, off and on, to abolish the family for literally thousands of years, without success. That is because the institution organizes and channels a set of emotional and behaviour dispositions that are deep feature of mammalian psychology, which are extremely difficult to override. The most that anyone has been able to accomplish is to create a “sterile caste” (such the Catholic priesthood, or the Chinese imperial eunuchs), and these also tend to be unstable. By contrast, the “guardian” arrangement that one finds in Plato’s Republic has been tried innumerable times, and has never been made to work.

The commune movement of the 1960’s counterculture provides some instructive lessons in this regard. At one time, there were more than 5,000 communes active in the United States, engaged in a variety of more or less extreme “experiments in living.” A friend of mine spent several years in a “child-rearing collective,” a group of five men and five women who decided to have children and raise them collectively, without any distinctions of biological parenthood. (The plan involved not telling the children who their biological parents were, but for all ten to be treated as the parents, and all to be equally involved in each child’s care and development.) The entire thing was an ignominious failure, and collapsed in acrimony within a few years. Many of the people involved wound up in family court, suing each other for custody and visitation rights.

What is perhaps most striking about the commune movement is the fact that almost everything that was tried failed. Of the literally thousands of experiments that were undertaken – different ways of organization social relations, in practically every dimension of life – almost everything fell apart. Depressingly, there were only two types of communes that proved capable of surviving for more than a few years: those that had a powerful religious orientation (e.g. cults), and those that chose to bureaucratize, adopting written laws and formal decision-making procedures. In other words, what the commune movement wound up proving was that the only ways people have of reliably getting along with one another are ones that are already being used in mainstream society.

So what are we to conclude from all this? It’s that humans are in many respects an extremely flexible species, heavily reliant upon social learning, and able to exercise a great deal of control over our social interactions by enforcing norms of behavior, and yet we remain at the same time heavily constrained – sometimes in rather surprising ways – by our evolved psychology, and a set of biologically inherited dispositions, which bias our social relations in particular directions. So a great deal of what we naively think of as “human nature,” is in fact social in origin, not natural at all, and yet is also, in many cases, remarkably resistant to modification.

One other point bears repeating. This analysis helps us to understand both why and how the “progressive left” has the capacity to become dangerous. Most people who identify with the left think of themselves as such morally upstanding people, that they have difficulty imagining themselves as capable of doing anything bad. This is why is it worth remembering that the left, or at least governments that enjoyed widespread support on the left, killed somewhere between 70 and 100 million of their own people over the course of the 20^th century. A large part of this was due to the fact that they were pursuing utopian political projects, which required individuals to exhibit levels of prosociality that were beyond the capability of most people. When these projects began to fail, or to encounter resistance, instead of acknowledging that they were encountering the limits of “human nature,” governments instead looked for scapegoats, “counterrevolutionary,” “bourgeois,” or “reactionary” elements, who were then held responsible for sabotaging the project. The results were tragic.

This is why all the “social constructivism” talk makes some people nervous. The problem is that we have, on the one hand, the requirements of justice, and on the other hand, the constraints of human nature. We still have not formulated any principled way of thinking about the relationship between the two, or how we might reasonably accommodate the former to the latter. The problem was identified quite clearly by Sigmund Freud, who in his conclusion to Civilization and its Discontents, offered the following as a critique of what he called the “cultural superego”:

It, too, does not trouble itself enough about the facts of the mental constitution of human beings. It issues a command and does not ask whether it is possible for people to obey it. On the contrary, it assumes that a man’s ego is psychologically capable of anything that is required of it, that his ego has unlimited mastery over his id. This is a mistake; and even in what are known as normal people the id cannot be controlled beyond certain limits. If more is demanded of a man, a revolt will be produced in him or a neurosis, or he will be made unhappy. The commandment, ‘Love thy neighbour as thyself,’ the strongest defence against human aggressiveness and an excellent example of the unpsychological proceedings of the cultural super-ego. The commandment is impossible to fulfil; such an enormous inflation of love can only lower its value, not get rid of the difficulty. Civilization pays no attention to all this; it merely admonishes us that the harder it is to obey the precept the more meritorious it is to do so.

It does not seem to me that we have made any progress in our thinking on this problem, over the course of the past century.