Bentham of Bentham’s Newsletter has an article entitled “What Is a Parent?” in which he argues that Matt Walsh makes some erroneous arguments in his recent documentary “What Is a Woman?.” Bentham rejects the definition that “a woman is a person who identifies as a woman" but provides a defense of what he calls a “social definition” of a woman, namely “a woman is someone who broadly has a large number of traits associated with the female sex.” He uses an analogy to illustrate why this is a good definition:
A clear analogy of the word woman can be seen with the word parent. What is a parent? Well, the reductive biological definition of the word parent would merely be the people whose sperm and egg produced a child. However, this definition is obviously much too narrow. It leaves out step-parents, adoptive parents, and other people who fill the social role of parents, while not being responsible for the fertilization of the egg which became a human. However, a parent is not merely someone who identifies as a parent—that would be regressive and circular. Instead, there are some social roles associated with the strictly biological notion of parent. A parent is someone who fulfills a reasonable number of the social roles associated with a biological parent.
In the movie, Walsh asks a man how he knows that he is a man, to which he replies “I got a dick.” Similarly, one could imagine asking a person how they know that they’re a parent, to which they’d reply, “cause I fucked my wife and a kid popped out.” However, such a definition would be unsatisfactory—after all, adoptive parents are still parents. One could imagine Walsh in an alternative world accusing those who expanded the definition of parent to include adoptive parents of playing a political game. This would, however, be clearly absurd—there are obvious benefits to having the word parent including adoptive and step-parents. The same is true when it comes to gender—the biological definition is explanatorily poor and makes people’s lives worse.
A word’s definition is not determined by the “obvious benefits” of the definition but by how the word is ordinarily used. People ordinarily use the term parent to refer to step-parents, adoptive parents, and biological parents. If someone said “those aren’t your parents” to an adoptee, the adoptee could respond, “yes, they are because adoptive parents are parents.” This would register as true for almost every English speaker because it’s a normal use of the term “parents.” Bentham’s hypothetical Walsh sounds “clearly absurd” because of the ordinary use of the word “parents,” not because of the benefits of having “parent” defined a certain way.
How we currently use a term is separate from how we should use a word. Language as it is and language as it ought to be are different. If Bentham provides a persuasive example about why we should use the social definition of the term woman, that is not a good reason to believe that we do use the social definition of a woman at present. That is a separate moral issue. In defense of the ethical claim that recognizing gender identity makes transpeople’s lives better, he provides links to several articles (Ennis, 2021; Ross et al., 2022; Tordoff et al., 2022). I would like to first deal with the semantic claim and address the ethical claim later.
Bentham defends the accuracy of his social definition by saying, “It would be bizarre to describe Buck Angel as a woman, Blaire White as a man, and this person as a man.” It’s odd because of their outward appearances, but it’s not that bizarre if you believe that manhood and womanhood are determined biologically. I don’t know about Nikkie, but Buck Angel still has a vagina and doesn’t intend to get a penis. Blaire White has been vocal about not wanting bottom surgery and still says she is "avoiding that surgery at all costs." Has she taken on enough “traits associated with the female sex?” This question can only be determined by ordinary usage, and people with penises are not ordinarily regarded as women. I’m not merely suggesting that it is rare to have a penis and be a woman; I am suggesting that most people see this as a disqualifying trait. Granted, White is exceptionally feminine-looking, and most people wouldn’t suspect she has a penis if they saw her on the street. This is why it seems bizarre; White has a highly unusual appearance for a person with a penis.
I asked Bentham, “If Blaire White decided to begin identifying as a man tomorrow, would it be bizarre to describe (him) as a man? How should we address Blaire in this hypothetical?” Bentham responded, “I'd classify Blaire as a woman for the reasons described in the post, but I'd use whatever pronouns Blaire preferred.” This indicates something significant about his definition: identification is insufficient. For most trans rights advocates, identification is necessary and sufficient. For Bentham, you must take on enough traits, and taking on enough traits without identifying could land you as the opposite gender.
Why is this relevant? It’s important because, under this definition, it’s not the case that “all transwomen are women.” You could identify as a woman but not have taken on enough traits to be a woman. For this reason, I think Bentham shouldn’t claim credit for including tranwomen when he says: “We have a very good reason to include trans women as women and trans men as man [sic], in our definitions of the words.” He includes some, but not all, transwomen. Bentham addresses this when he responds:
Now one might at this point worry that the definition provided would entail that transgender people’s preferred pronouns shouldn’t be used across the board. After all, there are many transgender people who still fill the social role of the gender they were assigned at birth, such that one would ordinarily mistake them for the gender they were assigned at birth, even though they currently don’t identify as that gender. This is not the correct implication.
What is a person named John? Well, it’s not really a person who identifies as a person named John—that definition would be circular. Really a person named John is someone who is generally described as named John, by most people. If a person who was named Ralph at birth decides they want to be called John, prior to them telling anyone, it wouldn’t quite be accurate to describe them as a person named John.
Similarly, what is a former addict? Well, it’s someone who was addicted at one point, but who has put their addiction firmly behind them. In the most reductionist sense, it wouldn’t be quite accurate to call a person who has been sober for two weeks a formal alcohol addict.
Finally, what is an American? It’s not just a person who identifies as an American—if the Taliban called themselves American, they’d be wrong. However, if a person recently immigrated to the U.S. and valued American culture, but wasn’t yet really part of American social life, I’d call them an American if they wanted me to.
Similarly, if a person wanted to be called John or a former addict, I would do so. I’d especially do so if there was solid evidence that calling John by his birth name would make his life a lot worse, and that calling people who have been sober for a week former addicts makes them less likely to go back to their addiction.
This is similar to what Aella does in her article “What A Woman Is.” She recognizes that her definition excludes some transwomen, but she decides to address people how they want to be addressed for ethical reasons. She analogizes, “If your grandmother with severe dementia keeps thinking you’re her long-dead brother, you’d probably have no problem playing along with it.” Likewise, Bentham is willing to call people things even if it’s not entirely true, especially if there is evidence to support the idea that it makes their lives better. Both value the ethical benefit of recognizing self-identification. It seems they believe that it’s not true that “all transwomen are women,” but it is true that “we should act like all transwomen are women,” at least concerning how we address them.
Bentham correctly recognizes that his definition doesn’t give a straightforward way of determining if someone is a woman. He says that the “same is true of nearly all words.” I agree! He says that the biological definition is “tricky and imprecise.” He interprets the biological definition to be chromosomal. I don’t believe the chromosomal definition is accurate despite many conservatives accepting this definition, but I do believe in a biological definition. I think that the ordinary definition of a woman is an adult human female. This is usually the first dictionary definition. While dictionaries are not perfect, Merriam-Webster says that a woman is an “adult female person,” and Cambridge says a woman is “an adult female human being.”
I’m not sure if he does, but if he thinks that the biological definition of adult human female is “tricky and imprecise,” his definition must be even more tricky and imprecise because it is both more complex and includes the words “female sex.” If he is confused about what I mean when I say the “female” part of “adult human female,” then he has to also be confused about what “female” means in his provided definition “a woman is someone who broadly has a large number of traits associated with the female sex.”
On Merriam-Webster, the definition of female is “of, relating to, or being the sex that typically has the capacity to bear young or produce eggs.” And an additional definition is “having a gender identity that is the opposite of male.” This additional definition is confusing because people don’t identify as the opposite of males; they identify as either women or females, but it looks like Merriam-Webster wants to avoid the issue of recursiveness. The ordinary use of the term female is the biological one.
When discussing an animal’s sex, we do not discuss gender identity; we think of sex organs and gametes. We also have terms for the different sexes of the same species. For example, we call adult male cattle “bulls” and adult female cattle “cows.” That is what we have done with humans. Adult human females are called “women,” and adult human males are called “men” in the ordinary use of language. Even gender scholars agree that this is what people ordinarily mean, at least maybe until extremely recently:
Few terms can be perfectly defined without any exceptions. We can imagine slight ambiguity in the terms adult, human, and female. Most people don’t care to dispute the adult and human parts. Even if I were to concede that perfectly defining a female is tricky, it’s reasonable to say that we know what a female is not. If a human produces sperm, they are not a female according to the ordinary usage of the term. The biological definition does not have to perfectly categorize all living humans to be the more commonly used definition. Besides, the social definition seems less capable of classifying all people perfectly. We could even regard true hermaphrodites as both male and female, but we could also categorize them by which gametes they produce. If someone were to exist who produced both eggs and sperm, we could say they are both male and female. This doesn’t seem like a bastardization of these terms as they are typically used, although these are extremely unusual cases.
It would be odd to say that people don’t know what they’re talking about when they say Lia Thomas is a man swimming in a women’s league. These people are not confused. In ordinary usage, womanhood is not a bundle of traits because certain traits are immediately disqualifying. If you have a penis, you are not a woman according to the ordinary use of the term. If you asked people in 2000, 1900, 1800, 1700, and 1600 if a woman could have a penis, they would tell you no, and they would not be ignorant of the definition of “woman.”
In 2022, a large portion of the population uses the identity definition, which is not the “social definition” that Bentham prefers. Perhaps, in the next decade, the ordinary use of the term woman will be non-biological. If that is the case, I will no longer defend the claim that the ordinary use of the term is an adult human female. All of this is not to say that we should adopt a different use of language; it is just to note that the definition of words does not depend on their ethicality or utility but their everyday usage.
Is the social definition actually the ordinary definition? I don’t think so. Recall Bentham’s social definition: “a woman is someone who broadly has a large number of traits associated with the female sex.” We don’t really have any insight into which traits are more important than others, so we are left guessing. One implication of this definition is that almost all transwomen in the closet would not be women. For example, if someone still presented as a man and told people they were a man but internally felt as though they were truly a woman, this person would not be a woman. This is in contrast to the more typical identification definition, which would grant this person the status of a woman.
Under Bentham’s definition, someone could inadvertently become a woman. Imagine a feminine homosexual who enjoys wearing makeup and many things typically associated with being a woman. This person could perhaps be regarded as a woman, even if they didn’t want to be. Perhaps a woman who has short hair, likes lifting weights, and wants to be a CEO should be regarded as a man. A person could teeter on the edge of manhood and womanhood, switching genders depending on what they watch on TV or what color shirt they choose to wear. Does this sound descriptive of how the word “woman” is ordinarily used?
An important question is who is creating the associations in the social definition. Suppose it is society specific; under Bentham’s definition, someone could cease being a woman if they left their culture and moved to a different country where different things are associated with the female sex. Perhaps on a trip through North Africa, the Middle East, and Europe, a person could switch from man to woman and back by crossing borders. Also, someone could be a “woman” but not a “mujer” if Spanish-speaking people continue to use the regressive definition of “persona adulta de sexo femenino.”
Are the associations time-specific? In this case, someone might cease being a woman because the culture changes as she grows older. If we are asked to “imagine a woman in the year 2500,” it would be challenging because we don’t know the traits associated with womanhood in this year. Indeed, many of the women of today would be regarded as highly manly by seventeenth-century English speakers. The shift to a more relativistic definition opens the door for strange conclusions that would seem to contradict people’s ordinary understanding of a woman. It’s hard to find the social definition more accurate compared to the adult human female definition.
Bentham desires to adopt the social definition on account of ethical considerations. This is a more defensible position. I do not think the social definition could accurately describe ordinary language usage. Still, I don’t have an issue with entertaining the idea that we should intentionally try to change our use of language. I think transgender advocates should abandon the semantic argument that the average person’s idea of the defining traits of womanhood is incorrect. Instead, they should concede the semantic point and work toward getting people to adopt a new definition. This project is sometimes called ameliorative inquiry. Jenkins (2016) explains:
In a series of papers published together in the collection Resisting Reality, Sally Haslanger develops a distinctive form of conceptual analysis that, she claims, can be applied to the concept of woman to offer a unique way of avoiding the inclusion problem. This method of analysis is termed ‘ameliorative inquiry’. According to Haslanger, an ameliorative inquiry into a concept F is the project of arriving at the concept of F-ness that a particular group should aim to get people to use, given a particular set of goals that the group holds. Ameliorative inquiries thus make use of normative inputs. The concept of F-ness that is generated by an ameliorative inquiry is the target concept of F. Ameliorative analysis is not bound to comply with our ordinary understanding or use of a concept: the target concept may be revisionary, provided that it furthers the goals guiding the analysis.
The newly constructed definition should not be Bentham’s social definition. In addition to being inaccurate, as I argued above, Bentham’s definition lacks utility because it is vague. More importantly for gender activists, it very much fails to solve what is sometimes referred to as the “inclusion problem” or “The Trans Inclusion Problem”: “the problem of conceptually engineering our concept of womanhood so as to respect everyone’s self-identification” (Bogardus, 2022). The social definition could potentially exclude large portions of the transgender population and unintentionally make a bunch of people trans. Granted, this is a complex problem. The philosopher Tomas Bogardus—who inspired many of the above arguments—believes that this is impossible to solve primarily because “no matter what it means to be a woman, it’s one thing to be a woman, and another thing to identify as a woman” (Bogardus, 2022).
Bentham recognizes that some people are men but identify as women. Fellow substack writer Aella also recognized this (Aella, 2022). For a rebuttal of her article, see mine entitled “Contra Aella on What a Woman Is.” They both advise going along with it for ethical reasons. But if we construct a new definition without claiming it is the ordinary definition, we can be significantly more inclusive. I think gender activists should argue like this:
A woman ordinarily means an adult human female. We propose an additional definition which is “An adult human who genuinely calls themselves a ‘woman.’” Under this additional definition, ‘woman’ need not have a meaning. The result is that this doesn’t convey any information about someone other than that they call themselves a ‘woman.’ This allows us to say that “transwomen are women.”
By design, this definition will convey no information about what a woman is. If self-identity is not sufficient, then it’s possible that some tranwomen are not women. Rather than feeling it is necessary to lie or go along with what we deem false narratives, we can accept a meaningless definition. I think that a major goal of trans advocates is to conclude that “transwomen are women.” I have provided a means of doing so.
After adopting my definition and conceding the ordinary use of the term woman, the job of the gender activists would then be to demonstrate that using this constructed additional definition is socially beneficial and improves the lives of trans people. Bentham already provided some research on why it is emotionally beneficial to have gender-affirming care and use inclusive language (Ennis, 2021; Ross et al., 2022; Tordoff et al., 2022). This seems plausible to me, but I have serious concerns about transitioning, especially in the case of children.
Affirming a child’s gender and letting them socially transition may not be best for their overall wellbeing. Trans researchers Michael Bailey and Ray Blanchard provide some recommendations for what to do if your child is experiencing childhood-onset gender dysphoria: “If you want your childhood-onset gender dysphoric child to desist, and if your child is still well below the age of puberty, you should firmly (but kindly and patiently) insist that your child is a member of his/her birth sex” and “above all else, you should not let your child socially transition to the other sex” (Bailey and Blanchard, 2017). Childhood role transition increases the likelihood of gender dysphoria persisting (McGuire et al., 2013). And the literature suggests that most gender dysphoria desisted in the past (Cantor, 2016). If we could prevent the dysphoria from existing, there is no need for social role transition, hormone injections, and invasive surgeries.
There may be similar issues with affirming the gender of adult transwomen and encouraging social role transition. Some even argue that the very phenomenon of gender dysphoria is rooted in the idea that changing sex is possible (Imouto, 2021). Perhaps reducing the portion of the population that believes males can become women could curtail dysphoria.
This is a challenging ethical issue in which people concerned about the welfare of gender dysphoric and potentially gender dysphoric people could find themselves unsure about what to do. We can even see this sentiment expressed by someone who has written in support of the idea that transwomen are women (see: Alexander, 2014); Scott Alexander explains the complex difficulties of this issue in his article “In Partial, Grudging Defense Of The Hearing Voices Movement”:
I’ve been following the conflict between pro-transgender activists who want to celebrate transgender people and fight stigma against them, vs. anti-transgender activists who want to prevent a bunch of kids from hearing that being trans is cool and so transitioning. This is a really sensitive issue but I am going to do my best to talk about it, and I’m sorry for probably offending both sides.
My starting point for any discussion of this, which I feel like it’s really hard for a well-informed and well-intentioned person to disagree with, is that at least some large subset of transgender people aren’t consciously faking it. That is, they genuinely have the experience of feeling like they are the other gender, they’ll be absolutely utterly miserable if forced to live life as their birth gender, and telling them “no, just snap out of it” will not work, at all. I think it’s hard to know transgender people closely without coming to this conclusion, unless you have some kind of really galaxy-brained take beyond my ability to even imagine .
So what’s left for the people who believe transgender is a “social contagion” or about “special snowflakeness”? If I had to steelman their position it would be something like: there’s some switch that can be flipped by social pressure and wanting to look cool. Once the switch is flipped, you’re transgender in some pretty real way: you’re not faking it, and you’ll be miserable until you’re allowed to gender transition. Still, being transgender makes people worse off on net, so society should try to avoid flipping that switch.
(if you read this blog often, you might notice similarities to my theory of anorexia: yes, lots of people start dieting because they want to be a ballerina or something, but the extreme dieting seems to flip a switch, the switch turns it biological, and you can’t make anorexics go back to healthy eating just by convincing them not to want to be a ballerina anymore)
If this were true, a maximally compassionate policy would involve both trying to support people who are already transgender, and trying to prevent the switch from being flipped in people who aren’t transgender yet. I’ve never heard anyone explicitly advocate this policy, probably because it’s really hard to get right - the harder you try to avoid talking about it in front of impressionable young minds, the more you risk stigmatizing existing transgender people, and vice versa. There’s enough bad blood around this issue that I’m sure neither side would trust the other to respect a compromise like this. Probably they would be right not to trust them. Still, when I try to figure out how I personally should behave, I give some weight to considerations like these.
I think a plausible case can be made for the idea that the proliferation of the view that male humans can become girls and women may be socially harmful; however, this is beyond what I wanted to do here. This is a debate for another time. As I discussed in my article “Contra Aella on What a Woman Is,” the fundamental debate is not semantic but ethical. I don’t think the trans activists have the semantic argument on their side; the ordinary use of the term woman is “adult human female.” Even if that is not the case, when “women’s sports” were created, the intended meaning was “adult human female sports.” The same goes for prisons, scholarships, parking spaces, bathrooms, shelters, and other spaces for “women.” Trans rights activists should concede this point, adopt my constructed definition, and reject the idea that the constructed definition has to convey meaning about what a woman is. They should then make the case from an ethical perspective, as many do, that we should affirm transwomen's identities and fully integrate them into society. In other words, spend less time arguing about words and focus more on ethics.
This was well argued. I may write a response at some point, but I think you represented your position pretty well.
> I think gender activists should argue like this:
You probably know why this isn't happening.