I’ve been trying to help people understand demography on X. It’s not easy.
One of the hardest concepts to convey is the difference between abstraction and concreteness. I have written about this problem many times: Equity and Social Justice: Rationalizing Unjust Enrichment; Awakening to the Problem of the Awokening: Unreasonableness and Quasi-religious Standards; The Myth of Institutional Racism; Why the Rhetoric of “White Privilege” is Anti-White Prejudice and Racism; The Myth of Systemic Racism in Lethal Police-Civilian Encounters; Staying Focused on the Problem with Critical Race Theory; Such a Beautiful Moment—The Self-Flagellating of White People; Policy Presuming “White Privilege” Violates Equal Protection Under the Law; The Behemoth Returns: The Nazis Racialized Everything. So Do CRTs.
As my previous essays make clear, this difficulty is not unique to X users; it extends to academics as well. Many academics, especially those who push the social justice narrative (what Thomas Sowell calls the “unconstrained” vision), routinely commit the fallacy of misplaced concreteness, treating living individuals as though they were nothing more than statistical abstractions (this is how one gets absurd notions like “white privilege”).
Another problem I’ve run into, in trying to clarify basic social science, is the lack of distinction between different kinds of demographic categories: natural, ascribed, achieved, identity-based, and so on. This is not unrelated to the first problem.
When we talk about demographic classifications, it is important to recognize that not all categories are of the same type. The classifications are designed by demographers or pollsters to capture different aspects of human life, some rooted in biology, others in society, and still others in belief or ideology. A clear typology helps us avoid conflating things that are fundamentally distinct, while also highlighting the ways categories are constructed, chosen, or imposed—whether by nature or by law.
Putting on my sociology hat, here’s a brief primer before getting to the point of the present essay:
The most fundamental demographic category is the natural type. This type captures attributes rooted in biological or physical facts about human beings. Age and sex are paradigmatic: they are empirically observable and not subject to belief or choice. One cannot opt out of being a certain age or having been born male or female, even if these realities can be misunderstood or misdescribed. Natural categories are, in this sense, immutable—permanent and unchangeable.
Some traits—such as race—are often treated as natural but depend on social interpretation. For that reason, they blur the line between natural and ascribed classification. Ascribed categories are social positions assigned at birth or imposed involuntarily, regardless of personal agency. Caste, ethnicity, and race (if one subscribes to the social construction thesis) are examples. In many societies, even religion can function as an ascribed category when one’s identity is determined by birth into a family or community (this is typical of Jews and in some communities Muslims). The defining feature of ascribed categories is that they are socially enforced rather than individually chosen or naturally given, although they can often feel natural.
In contrast, achieved categories are those attained through action, attainment, or performance. Education, income, marital status, and occupation all fall under this heading. These categories mark outcomes of agency, though they remain shaped by larger social structural factors. For example, completing a college degree is an achieved rather than an ascribed status. Achieved, unlike natural or ascribed categories, describe what people do rather than what they are.
A fourth type, increasingly prominent in contemporary discourse, is the identity-based or ideological category. These are grounded in belief systems or self-identification/understanding rather than natural or immutable characteristics. Gender identity, political ideology, and religious conversion all illustrate this type. Someone may believe he was “born in the wrong body,” just as another may believe he is “born again” in a spiritual sense. These categories are powerful in shaping behavior (and that’s the point we’re coming to), but they belong to the realm of self-conception rather than the realm of natural fact. They are, in this sense, adopted rather than observed.
There are also relational or situational categories, which only exist in relation to others or within particular contexts. Being a colonizer, an immigrant, or a minority is not a permanent feature of an individual, but positions that depend on the social or political environment. A person may be a minority in one country but part of the majority in another. These categories are thus contingent, arising from the relational dynamics of populations.
Finally, administrative/legal categories are defined and maintained by institutions and states. Citizenship, disability status, residency, and veteran status are classifications dependent on formal recognition and legal authority, even if they are sometimes rooted in natural fact. They may also overlap with achieved or ascribed statuses, but they are distinct in that they exist at the level of policy and enforcement. One is a citizen not because of nature or belief but because a legal system says so, even if citizenship is ultimately natural or organic (such as in jus sanguinis).
These different demographic classifications represent how demographers group and understand human populations. Some categories describe immutable facts, others reflect social impositions, others depend on individual agency, and others rest on belief or institutional authority. Clarity about which type we are dealing with is essential to making sense of demographic data, interpreting social patterns, or asking meaningful questions about human behavior. This is what X users have so much difficulty understanding.
So, with all that in mind, here’s why I’m talking about this: I suggested on X, since there have been trans identifying people killing children and a man sympathetic to the trans cause assassinating a youth leader, we should ask why a trans-identifying individual—or an ally of that community—might commit an act of violence, such as killing children or assassinating a political leader. (See my most recent essays: Charlie Kirk’s Killer is in Custody and the Specter of Antifa; The Fool Has Come Down Off The Hill. But Who Called on Antifa to Terrorize the Village?)
This suggestion was met with a familiar objection: “Most mass shootings are committed by white heterosexual males.” I could have simply replied, “So?” But I didn’t, because I saw a teachable moment (though in hindsight I assumed too much from my audience). Not only is the claim factually misleading (in fact, black heterosexual males commit more mass shootings), but it also reflects a deeper conceptual confusion. The categories being compared—race, sexuality, gender, and transgender identity—are not of the same type.
As I noted above, gender, race, and sexuality are natural categories—although perhaps race is better treated as ascribed (although I am leaning towards the opinion that it’s both, since the attributes are immutable). Gender and sexuality, however, are not “assigned.” They are natural facts, empirically observed. This is why the popular phrase “assigned male at birth” is so grating: it suggests arbitrariness where there is, in reality, none. Gender is observed, sometimes mistakenly, but observed nonetheless. Upon closer examination, one will always be found to be either a male or a female. There is no in between. Nor does one stand outside the binary.
By contrast, gender identity (“cis” vs. “trans”) is a qualitatively different kind of category. It is not a natural or immutable characteristic but an ideological, belief-based identity claim, closer to political or religious ideology than to gender or race (really, gender identity is both those things). No one is literally “born in the wrong body.” Ideology can lead a boy to believe he is a girl in a boy’s body, just as a religion can lead someone to believe there is an immortal soul or that he will reach paradise after death. These are beliefs—powerful and consequential, to be sure—not natural facts. Can we treat them as demographic categories? Sure. But we cannot change their type. They necessarily fall into the identity-based or ideological classification.
This distinction matters when we ask causal questions—that is, when we ask why something happened. If a white heterosexual male who embraces fascism commits political violence, the explanation is not that he is white, heterosexual, or male. It is his fascist beliefs that motivated his actions. Categories like gender and race do not possess agency; people do. And people act because of ideas, commitments, and cultural conditioning. Here, identity and ideology are decisive.
The same logic applies to disparities in violence between groups. Black heterosexual males are disproportionately represented among perpetrators of mass shootings. But the explanation cannot be reduced to “being black” or “being male.” Instead, we must ask about the cultural, ideological, and social environments shaping behavior: What beliefs, values, and experiences make violence seem like an available or meaningful option? The demographic disparity compels us to explain why disparities are statistically significant, but the category itself explains nothing. It possesses no agency.
The same applies when examining trans-identifying individuals. If some commit acts of violence, the root cause cannot be reduced to their observed gender. Most trans shooters are male; they can be lumped in with other male mass shooters progressives like to talk about—but that’s not an explanation. The relevant causal domain lies in belief: identity claims, ideology, and cultural narratives. Here, the distinction between natural demographic categories (immutable facts such as gender or race) and constructed demographic categories (belief-based identities such as gender identity or religion) becomes crucial.
In short, natural categories describe what people are, whereas constructed categories describe what people believe. The former are observed; the latter are adopted. The former have no agency. Personalities have agency, and personalities are moved by their beliefs. And if we want to understand violence—or any form of human behavior—our explanations must look not to what people are but to what they believe, what they do, and how they do it.
The problem of trans shooters cannot be sidestepped by pointing to the fact of white heterosexual male shooters. The trans shooter, like the Islamic terrorist, is moved by the same ideology that determines his identity—an identity self or other imposed but not natural. The white heterosexual male, by contrast, is simply an intersection of demographic classes organized around observed attributes. Reference to the latter cannot be a causal explanation.
For the record, the US Census does not (presently) include a question about gender identity. It only asks about gender, with response options “Male” or “Female.” Makes sense, since that is all there is. So, sure, gender identity can be a demographic category, and there is polling on the question (which indicates a social contagion; see Why Aren’t We Talking More About Social Contagion? Luring Children to the Edge: The Panic Over Lost Opportunities). But so also can those who live in reality, with such attributes as “Roundearthers” or “Flatearthers.” There’s polling on that question, too. It doesn’t mean it should be a demographic category in a government census. And so it isn’t.
freedom and reason 