2.a. Richardson’s spatial theory of sexual orientation

In The End of Binaries: How Gender and Sexuality Come in Degrees (Richardson, Reference Richardson2025), Kevin Richardson presents a comprehensive analysis of the relationships between gender and sexuality, one which rejects traditional binary views of these things in favor of a theory that models the complexities of gender and sexuality along multiple dimensions.

In the second half of the book, he argues for a view he calls the Spatial Theory of Sexual Orientation. To use his language, this view is both quantitatively and qualitatively nonbinary (Richardson, Reference Richardson2025, pp. 191–196). It is quantitatively nonbinary insofar as it is scalar, and so captures the reality that sexual orientation comes in degrees. And it is qualitatively nonbinary insofar as it does not analyze all sexual orientations as some combination, or degree, of heterosexuality and homosexuality. The Spatial Theory accomplishes this by discarding the traditional way of imagining sexual orientations as points along a line between “absolutes,” and instead envisions sexual orientations as locations in logical space (or: “the gender-sexuality manifold,” Richardson, Reference Richardson2025, Ch. 10).

As locations within this manifold, sexual orientations are characterized by particular profiles made up of four general elements: dimensions, centered dimensions, agent types, and patient types. To briefly explain each element: the dimension of a sexual orientation profile is “what fundamentally makes someone have the sexual orientation they do” (Richardson, Reference Richardson2025, p. 203). It characteristically concerns things like their dispositions to desire and to sexual behavior. Agent types concern qualities of the person with the orientation (characteristically sex/gender qualities), and patient types concern the qualities of the people they are oriented toward (again, characteristically sex/gender qualities). Finally, centered dimensions basically give us a way to account for how one person’s sexual orientation can fluctuate or otherwise vary within certain boundaries and still count as “the same” orientation, both metaphysically and semantically (Richardson, Reference Richardson2025, pp. 202–205).

On this picture, a particular individual’s sexual orientation will be determined by their sexual location, as defined by these elements. But note that, according to the Spatial Theory, there will be much more variation in sexual orientation than what our common taxonomies typically reflect. Richardson argues that this is a good thing (and I agree!), but it also means that we must distinguish between sexual orientations and sexual orientation categories, a distinction which our ordinary use of the term “sexual orientation” often collapses (Richardson, Reference Richardson2025, Ch. 10).

As Richardson explains, in talking about someone’s sexual orientation, sometimes we are referring to a person’s sexual location (as defined by the specifics of their profile), and other times, we are talking about their membership in commonly accepted sexual orientation categories like “heterosexual” and “bisexual.” And while someone’s sexual location will determine their category membership, different people with different sexual locations might be equally well-classified within the same sexual orientation category, depending on how those categories are constructed in the relevant context (Richardson, Reference Richardson2025, pp. 216–218).

In this article, I take issue with none of Richardson’s specific arguments. I agree that previous theories of sexual orientation have been inadequate in the ways he identifies; I think the view he develops succeeds in closing those gaps; and I especially appreciate his careful distinctions between the metaphysics of sexual orientation and the language of sexual orientation categories and taxonomies. I just disagree with the claim that the Spatial Theory amounts to a theory of sexual orientation, and that’s because this view is not really about sex or sexual desire. Instead, it seems to me that Richardson is giving a more general theory of personal orientation—one that takes shape via a formal analysis of the relationships between the metaphysics of gender, dispositions to behavior/desire, identity, and the semantics of the associated terms and taxonomies. This is an insightful and important project, but it’s different from the project of giving a theory of sexual orientation.

To see this, I want to focus specifically on what Richardson says about dimensions in building the Spatial Theory. If someone’s sexual orientation is determined by their location in the logical space Richardson describes, then these dimensions seem to be what make that location distinctly sexual, given that dimensions are the only elements of these profiles that offer a slot for sexual desires and/or behavior. It is notable, then, that Richardson fills in the details of these dimensions with other theories of sexual desires and/or dispositions. He writes—

This is somewhat surprising, given his earlier objections to these views (Richardson, Reference Richardson2025, pp. 198–202). But more than that, it’s a clear instance of the Independence Assumption. On Richardson’s view, dimensions form the heart of sexual orientation—again, this is what makes for sexual orientation (i.e., this is what makes for sexual locations). But these dimensions are not something the Spatial Theory is concerned with theorizing directly, apart from providing a slot for them in these profiles. In other words, this is a theory of orientation, but it’s only a theory of sexual orientation insofar as we choose to connect it to independent work on sexual desire and behavior.

As further evidence for this, in the same chapter, Richardson also declines to say what might make a desire distinctly sexual, aside from its being goal-directed toward sexual pleasure and/or sexual activity (Richardson, Reference Richardson2025, p. 200). This matters, because it means that when it comes to distinguishing sexual orientations from other ways of being oriented toward people, the Spatial Theory is only as successful as the independent theories of sexual dispositions it’s paired with. This illuminates that the central explanatory tasks here aren’t being accomplished by the Spatial Theory itself—instead, the Spatial Theory is more like a way of modeling the different elements of personal orientation in relation to each other. But the fact that this view remains neutral on the nature of dimensions is telling, because it means the Spatial Theory declines to take a stance on the central question for metaphysical theories of sexual orientation.Footnote ²

These considerations lead me to conclude that the Spatial Theory is not a theory of sexual orientation. Instead, it seems more appropriate to understand this view as a general theory of personal orientation, which offers a careful way to model the many ways that we can be oriented toward other people along various dimensions, and where that orientation is centrally characterized by socially significant features, such as gender and biological sex. But for the theory to be one of sexual orientation specifically, it would need to say something more about sex.

2.b. Andler’s social structural theory

Sexual desires are different from sexual orientations: the former can be very specific and idiosyncratic; the latter are somehow more general. In the literature on sexual orientation, this observation often motivates a shift from talking about specific sexual desires to talking about dispositions to desire as the basis for one’s sexual orientation. But even this is not enough, because it seems that only some, but not all, sexual dispositions ground sexual orientations. What makes the difference? Matthew Andler (Reference Andler2025) calls this the demarcation question of sexual orientation: “In virtue of what do orientation-grounding sexual dispositions, as opposed to freestanding sexual dispositions, ground sexual orientations?” (Andler, Reference Andler2025, p. 7).

In Sexual Orientation and Sexual Identity: A Philosophical Analysis (2025), Andler argues that sexual orientations are sexual dispositions with special properties. Specifically, sexual orientations are sexual dispositions that take on a particular kind of social significance in virtue of their relationship to certain dominant social structures, which Andler calls heteropatriarchal kinship structures. In general, kinship structures are social structures that regulate sex, love, reproduction, and care—where, generally, these regulations are enacted through systems and norms that work to reward people for their compliance, or else to punish them for their failure to do so. Which rewards/punishments you might be subjected to is a matter of your social role, or position, within the social structure? And when it comes to kinship structures in particular, these social positions are constitutively defined by gender.

Importantly, not all kinship structures are the same. Sex, love, reproduction, and care can be socially regulated in many different ways. What makes a kinship structure distinctively heteropatriarchal concerns how it regulates these things in accordance with an ideology which represents

On Andler’s view, sexual dispositions can take on social significance against this backdrop. Consider, for example, a cisgender man who tends to experience sexual attraction exclusively toward women: he desires a monogamous marriage with a woman, and to someday raise children with her. This is the sort of sexual disposition that is considered “natural and normal” within a heteropatriarchal kinship structure, and so it is the sort of thing that will be socially supported (enabled, rewarded) by societies structured in this way—which, just to be clear, is most societies.

On the other hand, consider a trans woman who tends to experience attraction exclusively toward other women: she wants to raise children with someone someday, but she’s not particularly interested in getting married, and her relationship practices tend toward non-monogamy. This is the sort of sexual disposition that is not considered “natural and normal” within a heteropatriarchal kinship structure and is therefore socially discouraged (punished, constrained) by societies structured in this way.

Importantly, both of these individuals have sexual dispositions that society cares about in some sense, and in each case, there are consequences for that person. We might compare these sexual dispositions to cases where someone has desires exclusively for blond people, or for tall people. These are ways of being attracted to other people, but these dispositions are not regulated by heteropatriarchal kinship structures: there are no public policies that discourage or encourage relationships with blond people, there are no tax breaks for individuals who marry tall people, and so on.

In other words, in contexts characterized by heteropatriarchal kinship structures, only those sexual dispositions that interact with what is considered “normal and natural” take on social significance—and it’s only when people seem to have/lack those sorts of dispositions that they are rewarded or sanctioned (marginalized or included; privileged or subordinated) as a consequence. This is ultimately what makes for the difference between “freestanding” sexual dispositions and those dispositions that ground sexual orientations (Andler, Reference Andler2025, p. 23). Andler calls this the Social Structural Theory of Orientation Demarcation, and it forms the basis for his discussion of the relationships and differences between sexual orientation, sexual identity, and sexuality cultures throughout the book.

Again, in this article, I take issue with none of Andler’s specific arguments. I agree that the demarcation question presents a challenge; I think the theory Andler develops offers an answer that is both metaphysically interesting and politically useful; and I especially appreciate his careful treatment of the relationships between queer sexual identities and different sexuality cultures, later in the book (Andler, Reference Andler2025, Chs. 3 and 4). I just disagree with the claim that the Social Structural Theory is a theory of sexual orientation, and that’s because this view is not really about sex or sexual desire. Instead, it seems to me that Andler is giving an analysis of the relationships and differences between sexual dispositions, orientations, identities, and cultures. This is an insightful and important project, but it’s different from the project of giving a theory of sexual orientation.

To see this, I want to focus specifically on what Andler says about the relationship between sexual dispositions and sexual orientations. On Andler’s view, sexual orientations are socially significant sexual dispositions. As stated, it is therefore a theory of what it takes for a sexual disposition to be special, but it does not tell us what it is for something to be a sexual disposition.

Now, one way to go here is to say that the Social Structural Theory accomplishes this anyway, insofar as other ways of being oriented toward people will not be socially significant in the same manner. But I think that’s unlikely. Consider an older woman who tends to feel maternal toward men, in particular: she wants to take care of them when they are sick, she wants to make meals for them, and iron their clothes, and so on. This is a way of being oriented toward other people, and it is strongly gendered (plausibly, even sexualized), and it is the sort of disposition that is considered “natural and normal” within heteropatriarchal kinship structures. And insofar as that’s true, women who are judged to have this disposition are rewarded for it. But as described, this is not a sexual orientation.

So, I do not think we can say that the Social Structural Theory distinguishes between sexual orientations and other ways of being oriented toward people, simply because the theory singles out sexual orientations as dispositions that are special in this way. Instead, it seems we need to say something about the difference between sexual dispositions and the rest. This gives us a different sort of demarcation question, one which specifically has to do with dispositions.

Andler does have a view about this. At the beginning of the book, he expresses skepticism about the project of analyzing a property like being sexual, but suggests that, when it comes to distinguishing sexual dispositions from the rest, we might not need to take that approach. Instead, he proposes that “sexual dispositions are dispositions sufficiently similar to dispositions that could ground sexual orientations,” and more specifically, that “an individual instantiates such-and-such sexual orientation property in virtue of instantiating such-and-such sexual disposition property” (Andler, Reference Andler2025, p. 6). In other words, his suggestion is that we answer the question of what makes something a sexual disposition by first addressing the questions of what it takes for something to be a sexual orientation. This is an interesting strategy, but it’s awkward in application to the Social Structural Theory, which analyzes sexual orientations as special sexual dispositions. That is, we seem to be traveling in a circle. Moreover, this strategy does not give us a way to explain why “feeling maternal toward” is not a sexual orientation—in fact, according to the suggested strategy, it seems this is both a sexual orientation and a sexual disposition.Footnote ³

These considerations lead me to conclude that the Social Structural Theory is not a theory of sexual orientation. Instead, it seems more appropriate to understand this project as an analysis of the relationships that hold between sexual dispositions and heteropatriarchal kinship structures, and the consequences of those relationships for sexual identity, politics, and culture. But for the theory to be one about sexual orientation specifically, it would need to say something more about sex.

4.a. The attraction view of sexual orientation

At the beginning of this article, I proposed a framework for evaluating theories of sexual orientation. Again, at a minimum, a descriptively adequate metaphysical theory of sexual orientation will:

1. (i) Theorize what (if anything) is generally true of the sexual desires/behavior of a given individual;

2. (ii) Theorize the similarities and differences between the sexual profiles of different individuals;

3. (iii) Respect existing terminology and taxonomies of sexual orientation; and

4. (iv) Account for the difference between things that are sexual orientations, and things that are not sexual orientations.

On what I will call the Attraction View of sexual orientation, someone’s sexual orientation fundamentally concerns what sort(s) of sexual interactions they want to have. These are not dispositions, but general truths about what the person wants. These facts, in turn, ground certain sexual habituals about that individual, such as “Mary desires sex with men,” and they also ground certain sexual generics about groups of people, such as “heterosexual women desire sex with men.” These elements of the view address the first criterion on the list above (i) and also provide a frame with which to theorize similarities and differences between the sexual profiles of different individuals (ii).

Some but not all of these sexual habituals and generics will inform how we classify a person within the sexual orientation taxonomies at play in a given context. Among other things, this means that the metaphysical reality of sexual orientation will often be much more complicated and nuanced than the language of sexual orientation categories and taxonomies. Sexual habituals and generics—and specifically: socially significant sexual habituals and generics—do the work of gathering up the underlying metaphysical complexity and connecting it to everyday language. This element of the theory addresses the third criterion above (iii), regarding the relationship between the metaphysics of sexuality and the language of sexual orientation terms and classificatory schemas.

Finally, as is the case with many generics and habituals, the underlying facts here may be much more complex than the general truths they ground, and importantly, they typically won’t directly reference gender/biological sex category membership. As I’ll argue, we can gain a window into what these facts are like by reflecting on how people experience sexual attraction in real time: “You seem like someone I could do _______ with.” This element of the theory addresses the fourth criterion above (iv), regarding the challenge of distinguishing sexual orientations from other ways of being oriented toward people: paradigmatically, your sexual orientation is about the sort(s) of sexual interactions you want to have.

4.b. The attraction view: sexual generics and habituals

On the Attraction View, sexual orientations aren’t dispositions.Footnote ⁴ Instead, true claims about sexual orientation are based on true sexual generics and habituals. The following are examples of sexual generics and habituals:

1. a) Heterosexual women desire sex with men.

2. b) Lesbian women desire sex with women.

3. c) John desires sex with women.

4. d) Mary desires sex with women and men.

The first two examples on this list are sexual generics; the second two examples are sexual habituals. Standardly, generics are general truths about kinds, whereas habituals are general truths about individuals (Fara, Reference Fara2005). Something can be generically true of a kind, and yet not true of every member of that kind; similarly, a given habitual can be true of an individual even if they do not satisfy the associated description all the time (Leslie & Lerner, Reference Leslie, Lerner and Zalta2026). Standard examples in the larger literature include claims like “mosquitoes carry West Nile virus” (a generic claim about mosquitoes) and “Martha smokes after work” (a habitual claim about Martha). Both of these claims are true, despite the fact that only a minority of mosquitoes carry West Nile virus, and sometimes, Martha skips her cigarette and decides to go for drinks with her colleagues instead.

Now, consider (b) on the list above: lesbians desire sex with women. The truth of (b) is consistent with some lesbians desiring sex with nonbinary people; it’s also consistent with some lesbians not desiring sex at all. Equally, a given habitual can be true of an individual, even if that person sometimes behaves in ways that fall outside of that claim, or else seem to contradict it. Consider (c) on the list above: John desires sex with women. That can be true, even if sometimes John does not want this.

In this way, talking about sexual orientation in terms of sexual generics and habituals gets us the same things as does talk of dispositions. In addition, this approach also avoids some of the traditional problems with dispositional analyses of desire, insofar as these dispositions are often understood in terms of conditionals: A is disposed to desire B under conditions C just in case, were A to encounter B in C, A would desire B. The problems here include how to specify the relevant conditions (Dembroff, Reference Dembroff2016), how to talk about degrees of desire (Richardson, Reference Richardson2025), and how to compare and measure different dispositions against each other (Wasserman, Reference Wasserman2011). Of course, talk of generics and habituals is not without its own puzzles and problems (Leslie & Lerner, Reference Leslie, Lerner and Zalta2026)! But at least we avoid the issues listed here.

It will also matter here that there are lots of sexual generics and habituals that we do not normally talk about, and which do not wind up (directly) informing our sexual orientation taxonomies. For example, consider claims like the following:

1. e) Alex desires to sexually dominate feminine people.

2. f) Mary does not desire to receive any kind of penetrative sex.

3. g) Alan desires to receive anal penetration.

4. h) Moira desires vaginal fisting.

As general truths about what these individuals want, these are sexual habituals. However, they do not clearly place any of these individuals into one or another sexual orientation category (e.g., “heterosexual” and “bisexual”). Going back to the introduction (Section 1), it seems to me that the literature on sexual orientation has focused largely on claims like “John desires sex with women,” which are sexual habituals that largely concern the biological sex and/or gender of the people we are attracted to, often in relation to our own sex/gender. And again, it seems this happens because these claims are the ones singled out by our sexual orientation terms and taxonomies. But if we are serious about drawing a distinction between the metaphysics of sexual orientation on the one hand, and the language of these taxonomies on the other (as happens in both Richardson and Andler’s projects), then I do not see why claims like (e)–(h) should not also track facts about someone’s sexual orientation.

Now, perhaps someone will object: if we grant that habituals like (e)–(h) can track facts about sexual orientation, then things get too messy. Specifically: if we allow for this, then the complexity of sexual orientation will outpace our available terms and taxonomies and so make it impossible to communicate about this stuff in meaningful, practically efficient ways. That is a legitimate concern, and my response to it is as follows: acknowledging the diversity and complexity of human sexuality—and so, of true, sexual generics and habituals—does not entail that we need to radically revise our current ways of talking about sexual orientation. Now, perhaps there are other reasons to do this, for example, reasons that concern sexual inequality, discrimination, and marginalization (Dembroff, Reference Dembroff2016; Srinivasan, Reference Srinivasan2021). But when it comes to the project of explaining how the complexity I’m describing here might align with the categories we already have, there are a number of options.

Both Andler and Richardson discuss how certain features of the world can become socially significant in various ways and for various reasons, and that this social significance explains the unique explanatory roles played by those aspects of reality. This is how Andler answers the demarcation question, for example: there are lots of sexual dispositions, but only some of them ground sexual orientations in virtue of the social significance that attaches to them. Richardson says something similar: the diversity of sexual orientation space far outpaces the nuances of everyday language, but again, we can appeal to the tools of social construction to bridge the gap—something is a sexual orientation insofar as it is socially recognized as such, in the context (Richardson, Reference Richardson2025, pp. 216–217).

A similar strategy is available to the Attraction View: there are lots of sexual generics and habituals, but only some of them guide our classificatory practices when it comes to sexual orientation categories. For example, a claim like “John desires sex with women” might lead us to classify John as heterosexual, whereas a claim like “John desires to be sexually submissive” will not similarly inform how we classify him. And that’s fine: not every truth has to directly correspond to a label. The important question now concerns what might make these sexual generics and habituals true, in the first place.

4.c. The attraction view: Sexual desires and fantasies

What makes sexual generics and habituals true? This is the million-dollar question, because this is ultimately what makes the Attraction View a theory of sexual orientation, and not personal orientation more generally. I’m going to argue that we can gain a window into what these desire facts are like by reflecting on how people experience sexual attraction, “in real time.”

Again, much of the existing literature on sexual orientation makes the Gendered Orientation Assumption: in theorizing sexual orientation, the only relevant generalizations and comparisons to be made across sexual desires and/or behavior are those pertaining to gender and/or biological sex. This assumption suggests that, when we are sexually attracted to other people, the primary locus of our attraction somehow concerns their (apparent) sex/gender. My basic point here will be that (most) people do not go to the bar, order a drink, see someone across the room and, in finding them attractive, think something like: “Wow, that person seems as though they are a woman!” There’s a lot more to this than that.

Specifically, I suspect this has to do with how sexual desire often involves fantasy. I think this is especially true of the desires we have for people we do not know, but I think we also fantasize about people we do know quite well and have been in relationships with for a long time. In talking about fantasies, I mean to focus on how we imagine the fulfillment of our desires. And, paradigmatically, a fantasy is sexual when it either involves imagining a sexual interaction, and/or when it leads us to feel sexual/erotic pleasure (even if we were not expecting to feel that way).Footnote ⁵

A few clarifications here. First, while some people have very elaborate sexual fantasies, the view I’m describing does not require this. Something as straightforward as seeing someone with nice biceps and imagining what it would be like to feel them can be a fantasy. On the other side of things, some of us have extremely elaborate, (almost) metaphysically impossible fantasies. If someone falls asleep every night to thoughts of their kinky, multi-gender mermaid polycule, does that make them bisexual or pansexual? I do not know, but this also is not the right question to ask (for a few reasons). I’m arguing that truths about our fantasies ground sexual generics and habituals—but again, not every sexual generic or habitual will inform how we are classified within common taxonomies. If you fantasize about sex with mermaids, then some interesting and unique claims are true of you, but that does not make you bisexual. We’d need a bit more information about you and your context to figure that out.Footnote ⁶

Additionally, it may not be the case that every instance of sexual attraction or desire involves fantasy. Earlier I talked about how, when someone asks about our sexual orientation, each of us automatically sorts out some experiences of attraction and focuses specifically on others. How do we do this? Where do we draw the line? I suspect the answer will be different for different people, but what I’m proposing is that sexual fantasies play a central role here, in identifying which desires and modes of attraction inform our sexual orientation and which do not.

Again, I think that paradigmatic sexual fantasies involve imagining sexual interactions, and/or imagining things that lead us to feel sexual/erotic pleasure. This means that there are infinitely many sexual fantasies, and even if we try to sort them into categories based on their similarities, there’s going to be tremendous diversity here. But for the sake of argument, I’ll focus on just one kind of fantasy which I think is especially common and which fits into the following general schema: “You seem like someone I could do _______ with.”

For example:

1. i) You seem feminine and submissive, like the sort of person I would enjoy dominating during penetrative sex.

2. j) You seem dominant and masculine, like the sort of person who would kiss me by putting your hand under my chin, and who would make me feel like a princess.

3. k) You seem like someone chaste and demure, like the sort of person who would be a good homemaker and a good mother to our biological children.

4. l) You seem confident and intelligent, like the sort of person who would provide for me and bring my life stability and safety.

My claim is not that people literally think these things upon experiencing attraction to one another. I think (sexual) desire is often much more immediate and embodied. What I am suggesting is that (i)–(l) are fantasies, insofar as they are descriptions of imagined interactions which fulfill someone’s desires.

What’s more, once we start thinking about the details here, it becomes less plausible that we desire other people largely because of their membership in gender and/or biological sex categories.Footnote ⁷ Of course, some of the items on the list above explicitly mention sex characteristics, and many (if not all) mention qualities and characteristics of people that are strongly gendered in many contexts. For example: if we stipulate that the person thinking (i) is a man named John, then (i) reads like a heterosexual desire in many contexts—the sort of desire that might ground a habitual claim like “John desires sex with women.”

Notably, the description given in (i) is not exclusively satisfied by women: there are plenty of feminine and submissive people who are not women (and there are plenty of women who are neither feminine nor submissive). Now, John might not consider this and so take his desires to be exclusively satisfiable by women, but in that case, he’d be wrong. Nevertheless, a desire like (i) might ground a claim like “John desires sex with women” if it is true enough of the time that this description picks out women in his context. This is where it matters that generics have weird truth conditions: “mosquitoes carry West Nile Virus” is true, even though only a minority of mosquitoes satisfy that description (Leslie & Lerner, Reference Leslie, Lerner and Zalta2026). Similarly, “John desires sex with women” might be true based on desires like (i), even if the description associated with the underlying desire does not single out women.

Among other things, this presents a new way to analyze the manner in which people can be mistaken about their sexual orientation. Cases like this have traditionally presented a challenge for the literature. For example: consider a woman who takes herself to be heterosexual largely on account of things like, for example, external social pressure or internalized homophobia, but if she were to move to a different city, she would have sexual relationships with other women. What is her sexual orientation? Perhaps the truth of this counterfactual means that she is mistaken about being heterosexual—she’s really a lesbian, or perhaps bisexual; it’s just that her restrictive social environment presents barriers to realizing her “true” sexuality. Alternatively, perhaps what matters here is just what she actually does (not what she would do if things were different), and since she actually has sex with men (and nobody else), she’s heterosexual.

On the Attraction View, the question of sexual orientation category membership is separate from the metaphysics of sexual desire. Whether the person in this example is heterosexual, bisexual, lesbian, or something else will depend on which sexual habituals are true of her and on how those claims inform her membership in one or another sexual orientation category (which is largely a question of social construction). And again, which sexual habituals are true of her is a matter of her desires. This presents at least two ways in which someone can be “mistaken” about their sexual orientation. First, they can be mistaken about which sexual orientation labels apply to them, in a given context. And second, they can be mistaken about which sexual habituals are true of them, given their desires.

The literature to date has focused largely on the first sort of mistake: does it ever happen that someone thinks they are heterosexual, but they are actually bisexual? If so, how do we analyze what’s going on there? I think this is an interesting question, but I also think it’s a question that has largely to do with the (meta)semantics of sexual orientation terms and taxonomies—that is, it’s ultimately a question about the meanings of sexual orientation terms, why those terms mean those things, and the conditions under which those terms apply to people.

I’m much more interested in the second mode of analysis here, regarding whether we can be mistaken about which sexual habituals are true about us, given our desires. For example, consider someone like John, who desires to have penetrative sex with feminine and submissive people. He might think that, in every context, this means it’s true that he sexually desires cisgender women. But there will be some contexts in which this desire is possibly fulfilled by men, nonbinary people, or trans women. I think this sort of “mistake” about which people actually satisfy the descriptions associated with our fantasies can do a lot to explain the fluctuating nature of sexual identity over time (Andler, Reference Andler2025). And I think it can also do a lot to explain transphobic and homophobic violence, as people realize not only that they were mistaken about someone else’s gender/sex, but that more fundamentally, they were mistaken about the nature of their own desires (Bettcher, Reference Bettcher2007).