1. Introduction
The metaphysics of gender and sexual orientation has, in recent years, moved to the forefront of contemporary debates in analytic ontology, prompting the development of increasingly sophisticated and thought-provoking accounts. These contributions have significantly enriched the conceptual resources available to both philosophers and non-philosophers for engaging with such topics. Within this growing body of literature, Matt Andler’s Sexual Orientation and Identity (Andler, Reference Andler2025) and Kevin Richardson’s The End of Binaries (Richardson, Reference Richardson2025) stand out as particularly valuable and insightful contributions.
On the one hand, Andler’s central metaphysical claim is that sexuality categories are socially constructed rather than purely natural. Sexual orientations (e.g., homosexual, heterosexual) are grounded in sexual dispositions, but these dispositions count as orientations only insofar as they are situated within heteropatriarchal kinship structures. Sexual identities, by contrast (e.g., queer, straight), are grounded in socially and politically meaningful ways of life, especially in how they resist or reinforce those structures. On the other hand, Richardson argues that gender and sexual orientation are not best understood as simple binaries, but as multidimensional spaces with socially recognized regions. Categories like man, woman, gay, straight, and bisexual are coarse-grained ways of locating people in these spaces. Richardson emphasizes that gender and sexuality are complex, plural, context-sensitive, and sometimes indeterminate: it can be neither clearly true nor clearly false that someone belongs to a category. Prototypes help explain central cases, while spatial regions explain degrees, fuzzy boundaries, and overlap. The result is a theory that treats non-binary identities as fundamental, not marginal.
My aim in this contribution is to weave together some of Andler’s and Richardson’s themes, especially concerning the ways in which communication about gender and sexual orientation occurs, both within philosophy and beyond. I begin with some observations Andler makes in Chapter 4, where he reflects on possible developments in the dialectic surrounding trans-exclusionary arguments. There, he critiques arguments claiming that trans women cannot be lesbians and therefore ought to be excluded from lesbian spaces. Andler notes that his critique may itself be met with an objection—the “change of subject” charge. I assess his reply to this charge, but I also argue that, given certain features of the exchange under consideration, the broader dialectic (including both the “change of subject” objection and Andler’s response) is defective. At this point, I introduce insights from Richardson’s work, in which disagreements about gender and sexual orientation are reframed as forms of negotiation. I suggest that the exchange in question can likewise be understood as a negotiation, one that includes a negotiation over its own subject matter. I argue that this interpretation makes better sense of the dialectic and sheds light on further moves available to inclusionary theorists who wish to build on Andler’s strategy.
The paper proceeds as follows. Section 2 reconstructs the biology-based view and the trans-exclusionary argument it supports. Sections 3 and 4 examine Andler’s two main critiques: first, that biological sex is not binary; and second, that sexuality categories are essentially sociopolitical. Section 5 introduces the subject-change objection and Andler’s response. Section 6 argues that this exchange generates a defective dialectic due to the indeterminacy of the subject matter. Finally, Section 7 proposes an alternative interpretation, drawing on Richardson’s account of gender discourse as negotiation. On this view, the parties are not only advancing competing answers, but also negotiating the very subject matter of their disagreement. This helps restore coherence to the dialectic around change of subject in this exchange.
2. The Biology-Based View and Trans-Exclusionary Arguments
Stock (Reference Stock2019a, Reference Stock2019b) endorses a biology-based binary view of sex, whereby having an endogenous penis is necessary for being male and having an endogenous vulva is necessary for being female, even though the male and female sexes involve a complex interaction among multiple traits (such as chromosomes, hormones, genital morphology, and gene expression). This biology-based conception can form the basis for further arguments concluding that trans women should be excluded from certain spaces. The starting point is that trans women lack a necessary feature of being female. This could be combined, for example, with a sex-based theory of sexuality categories, such as being gay or lesbian. For instance, a sex-based theory of lesbianism would state that being female is necessary for being lesbian. If trans women are not female because they lack a necessary feature for qualifying as such, then they cannot be lesbian, to the extent that femaleness is required for being lesbian. If trans women cannot be lesbian, then they should not—the argument goes—be admitted to lesbian spaces.
3. Andler’s First Critique: Sex Is Non-binary
Andler begins by challenging the central assumptions underlying Stock’s account of biological sex. On Stock’s view, sexuality categories are grounded in a binary classification of sex, where individuals are determinately male or female based on biological features—specifically, possessing an endogenous penis or vulva. Andler argues that this picture is overly simplistic. Biological sex, he notes, is not a single property but rather a cluster of traits, including chromosomes, hormones, genital morphology, and patterns of gene expression (p. 56). Crucially, these traits do not always align in a strictly binary way. Intersex variations, for example, demonstrate that individuals may possess combinations of endogenous traits that do not fit neatly into the categories “male” or “female.” From this, Andler concludes that biological sex cannot be reduced to a simple binary property capable of doing the classificatory work that Stock requires (p. 57). This significantly weakens the plausibility of grounding sexuality categories directly in biological sex. Nevertheless, Andler does not treat this point as decisive. A deeper problem remains concerning the nature of sexuality categories themselves.
4. Andler’s Second and Main Critique: Sexuality Is Essentially Sociopolitical
A biology-based account of sexual orientation categories, such as the one assumed by the trans-exclusionary argument considered above, would hold that being lesbian is grounded in being female and disposed to sexual attraction toward female bodies. Andler rejects this assumption, arguing that sexuality categories are not biological but rather sociopolitical (pp. 60–65).
His argument can be reconstructed as follows. First, if Stock’s theory were correct, sexuality categories (such as lesbian, gay, or bisexual) would be defined purely in terms of one’s biological sex plus sexual attraction to others on account of biological sex traits. Second, however, actual social practices surrounding sexuality suggest that these categories function as sociopolitical identities. They are shaped by and embedded within systems of oppression, marginalization, and collective resistance. The historical development of lesbian and gay communities, the emergence of queer political movements, and the activism of organizations such as ACT UP illustrate that sexuality categories are used to organize social groups, articulate shared identities, and resist heteronormative social structures. In Andler’s view, sexuality categories are ultimately and essentially sociopolitical.
Given this, Andler argues that categories whose primary function is sociopolitical cannot be adequately defined in purely biological terms. Therefore, the biology-based definition of sexuality categories defended by Stock is theoretically inadequate. When it comes to motivating the exclusion of trans women from lesbian spaces, Andler’s argument implies that such exclusion cannot be justified based on purely biological aspects, because being lesbian is not a purely biological matter; it is a matter of belonging to a group that has the function of organizing, protecting, and helping flourish groups of individuals whose biological traits could vary and, in principle, even diverge from the strict requirements imposed by the biological conception. Following Andler, one may hold that sharing certain biological features is less important than sharing a common struggle in a heteropatriarchal system, so much so that individuals who do not display the sex features imposed by the biology-based conception could become part of the community if the right social, cultural, and political conditions obtain. And indeed, Andler seems to imply that nothing suggests the right social, cultural, and political conditions would not obtain. Therefore, Andler concludes, “trans women who are exclusively sexually disposed toward women ought to have (special) access to lesbian spaces” (p. 51).
Andler’s conclusion leaves open whether trans exclusion could be based on other, non-biological but cultural, social, or political reasons. Yet, it still puts significant pressure on trans-exclusionary theorists to the extent that articulating these cultural, social, or political reasons becomes much more difficult once biological reasons are off the table. If the boundaries of sexuality categories are not sensitive to biological considerations but only (or predominantly) to cultural considerations, then finding reasons that are not openly discriminatory for excluding trans women from lesbian spaces (or from women’s spaces in general) becomes harder. The force of Andler’s move lies in its capacity to deprive trans-exclusionary strategies of their main leverage tool—namely, the appeal to binary biological sex distinctions in tracing the boundaries of gender and sexual orientation categories.
5. The Subject-Change Objection from the Trans-Exclusionary Theorist
Still, the dialectic could undergo a twist, as the trans-exclusionary theorist may have access to a counter-move. Andler considers a possible reply to his argument (p. 62)—namely, that, when he rejects the biology-based assumption about the nature of sexuality categories (such as lesbianism), he is simply changing the subject. Footnote 1
To properly examine this objection, I adopt a classical formal model of subject matter inspired by Lewis (Reference Lewis1988a, Reference Lewis1988b)Footnote 2: in this framework, a subject matter (or topic) is a partition of possible worlds. The space of possible worlds is organized into “partitions” T 1…Tn, which establish subject matter in a broad way. In turn, partitions are divided into cells c 1…cn. Suppose, then, you have the partition T for the subject matter of capital cities. Consider the proposition Madrid is the capital of Spain. This proposition is on-topic relative to T iff any two worlds in the same cell where appropriate conditions obtain—for instance, any two worlds in the cell where Madrid exists and is in Spain—“agree” on whether Madrid is the capital of Spain (cf. Kocurek, Reference Kocurek2025; Roberts, Reference Roberts2012).
A good heuristic for identifying the topic of a conversation is identifying the conversation’s “question under discussion,” or QUD. What question Q are the speakers trying to answer? This can be used as a guide for determining which subject matter the speakers are addressing. In turn, the question Q can be modeled as the set of its complete and exhaustive answers (cf. Groenendijk & Stokhof, Reference Groenendijk and Stokhof1984; Hamblin, Reference Hamblin1958, Reference Hamblin1973). Thus, if we identify the question as “What is the capital of Spain?”, the topic can be represented as the set of worlds (in relevant cells of a partition T) that exhaustively answer this question. For instance, the topic might be the set of worlds in cells where Spain exists and has a relevantly similar history and geography to the actual world, where Madrid is the capital.
With this framework in place, we can start to reflect on the subject matter of the hypothetical Stock-Andler exchange. To begin with, what is the question the trans-exclusionary theorist is trying to answer with their initial argument? I propose that the trans-exclusionary theorist’s argument is about a very specifically constructed question, namely whether trans women should be excluded from lesbian spaces provided a biology-based conception of lesbianism. One might observe that this is a rather loaded question, because it includes a very specific conception (or theory) of sexual orientation. While this might seem uncharitable at first sight, I believe it serves my broader purpose, for reconstructing the trans-exclusionist’s question as theory-laden in this way helps explain why the same theorist would, in some sense, “appropriately” press the change-of-subject charge against Andler’s argument, as Andler himself seems to implicitly acknowledge by taking the charge seriously enough to deserve a reply.
The following explanation might be offered: Andler’s contention that lesbianism (together with other sexuality categories) is a sociopolitical category seems partly to change “what we are talking about,” or “the question under discussion,” seen from the trans-exclusionist’s point of view, as it seems to shift to another specifically constructed question: whether trans women should be excluded from lesbian spaces provided a sociopolitical conception of lesbianism. Since the two questions, and hence subject matters, differ, Andler’s opponent seems positioned to reject his argument on account of a lack of relevance to the initial QUD.
Andler himself frames the subject-change challenge in ways compatible with this reconstruction, with additional details that illuminate the semantics of terms. First, Andler tells us, one might see the opponent as objecting that the ordinary meaning of “lesbian”—the meaning people “have in mind”—is biology-based (p. 62). If ordinary language meanings are to be taken seriously, then following the template from the previous paragraph, we should focus on the question whether trans women should be excluded from lesbian spaces provided a biology-based conception of lesbianism (or provided a biology-based interpretation of the word “lesbianism”) and not on a different question. Second, the opponent might insist that in certain contexts—such as medical contexts—the biological meaning is relevant (p. 62). If this meaning is more fundamental or relevant to the present discussion, then following the same template as above, this would again seem to justify why we should focus on the question that presupposes a biology-based conception of lesbianism (or a biology-based interpretation of the word “lesbianism”), rather than a different question. So, Andler’s presentation of the challenge fits with a reconstruction whereby the trans-exclusionist addresses a theory-laden question.
Andler resists both replies, maintaining that his strategy does not change the subject. If anything, Andler maintains, it better captures the subject matter or QUD. For instance, Andler questions the assumption that we should take the ordinary, sex-based meaning of “lesbian” seriously, emphasizing that this neglects the meaning already operative within queer communities, where the term is understood in sociopolitical terms (pp. 59, 63). If the queer community should have the final say about what terms like “lesbianism” mean, then we should focus on the question whether trans women should be excluded from lesbian spaces provided a sociopolitical conception of lesbianism (or provided a sociopolitical interpretation of the word “lesbianism”), rather than a biological one. Furthermore, he questions the fundamentality or priority of medical considerations, conceding that the biology-based interpretation of “lesbian” or “gay” matters in circumscribed contexts, such as the medical ones, while in other contexts more relevant to the present discussion, the sociopolitical interpretation takes over (pp. 63–65). This would again justify why Andler’s preferred subject matter is (in his perspective) the correct subject matter for the discussion; that is why we should focus on the question whether trans women should be excluded from lesbian spaces provided a sociopolitical conception of lesbianism (or provided a sociopolitical interpretation of the word “lesbianism”), rather than a biological one.
6. A Defective Dialectic
In the dialectic considered by Andler, both the party who initiates the subject-change objection and the party who responds to it seem to presuppose that the exchange has a determinate subject matter that the opponent is missing. This is a legitimate presupposition, if only because it would seem prima facie irrational to enter a conversation believing it lacks a determinate subject matter. So, we could concede that it is justified for the trans-exclusionary theorist and their opponent to believe that their conversation has a subject matter, and, on this basis, debate whether the subject matter has changed.
All this notwithstanding, it should be noticed that identifying the subject matter in this exchange is tricky; on closer inspection, there seems to be a lot of indeterminacy regarding what the QUD is for the parties, which strongly suggests that the exchange’s subject matter is unclear.
Let’s start with a simple option. One could point out: “Look, but aren’t they simply debating whether trans women should be excluded from lesbian spaces, full stop?” At a coarse-grained level, the subject matter may be identified with the plain question whether trans women should be excluded from lesbian spaces. This broad subject matter seems shared by Stock and Andler—it seems to be the broadly construed question they are both addressing. Yet I argue this is too coarse-grained, as the conception of the sexuality category lesbian—or the interpretation of the term “lesbian”—seems crucial to properly addressing the issue under discussion. If the answer to the trans-exclusionist’s question is “yes,” then this answer can only obtain in a specific set of possible worlds where the metaphysics of sexuality categories is entirely biology-based. By contrast, if the answer to Andler’s trans-inclusionist question is “no,” then this answer can only obtain in a specific set of possible worlds where the metaphysics of sexuality categories is determined by sociopolitical facts. If the answers each contender provides require selections of possible world sets where incompatible metaphysical conditions obtain, one could argue that they come to the table attending to diverging partitions of possible worlds from the outset. One theorist attends to the possible worlds of partition T 1, whose metaphysical structure makes sexuality categories fully biology-based, while the other attends to the possible worlds of partition T 2, whose metaphysical structure makes sexuality categories sociopolitically-based.
Now one could rejoin: “What if we obtain the union of the two partitions? Wouldn’t that count as the exchange’s overall subject matter?” The problem with this move is that the resulting “macro-partition” would lack uniformity in some fundamental, structural metaphysical conditions of the worlds that belong to it. It would be a disjunctive, gerrymandered macro-partition. Ceteris paribus, I take it, it would be better for a partition to feature worlds that are metaphysically uniform—or at least where the structure of the relevant portion of reality under discussion is uniform. In conclusion, identifying the subject matter of the Stock-Andler exchange broadly is unsatisfactory, because it would either leave out fine-grained metaphysical characteristics of world partitions (which seem crucial for the answers each contender supports), or force the obtainment of a metaphysically gerrymandered macro-partition.
Restricting the subject matter so that it is more fine-grained seems problematic too, once again for metaphysical reasons. This is because, at least prima facie, it seems reasonable to restrict the subject matter in accordance with how the actual world is—and, in this case, in accordance with the nature of sexuality categories, or the semantics of their terminology. Yet this appears unsettled. Empirical facts do not seem to reveal anything determinate about the biology-based or sociopolitical nature of gender and sexual orientation, so adopting a descriptive approach to the identification of subject matter will not help. No fact of the matter points to its determination one way or another.
If these considerations are plausible, there are reasons to think that subject matter is indeterminate. This provides the basis for my main argument, which draws on previous work of mine (Belleri, Reference Belleri2025). There, I argue that subject matter is generally indeterminate and shifty, and that this has implications regarding the legitimacy of subject-change challenges, as well as their responses—in other words, the whole dialectic surrounding subject change. My arguments applied especially to conceptual engineering projects, where a theorist strives to change a concept and introduce it in communication and inquiry. Someone who (like Haslanger, Reference Haslanger2000, for instance) is promoting a change in gender concepts, so these are seen as referring to socially constructed categories rather than biological categories, may be accused of changing the subject with respect to ordinary discourse (in which these categories appear to be widely thought of as biological). This, of course, resembles the challenge considered by Andler, to the extent that the sociopolitical conception he uses could be seen as excessively “altering the question under discussion,” as presented in the trans-exclusionary argument, which seems to appeal to a commonly held, biology-based understanding of these categories. The point I defended in the referenced article is that challenges like the one illustrated above (and the attendant replies) risk generating an overall defective dialectic.
In continuity with my previous position, I wish to argue that the dialectic is similarly defective—on both sides. First, the dialectic is defective for the party who presses the subject-change charge. If the subject matter of the discussion is ultimately indeterminate, then appealing to a change of subject is not an effective move. Recall that, as we have argued so far, there are good reasons to believe that the subject matter under discussion between the trans-exclusionary theorist and their opponent (in this case, Andler himself) is indeterminate: as we saw, the coarse-grained reconstruction neglects important metaphysical details or generates a gerrymandered macro-partition; furthermore, attempting a restriction to a more fine-grained subject matter based on how the actual world is appears metaphysically unprincipled.
In such a situation, the trans-exclusionary theorist may go ahead and press the change of subject objection, yet the objection would be ineffective. Why? On the one hand, the trans-exclusionary theorist may be working with the broader, more coarse-grained subject matter. Yet if the coarse-grained subject matter suffers from the metaphysical limitations just mentioned and is therefore incomplete or gerrymandered, the subject-change objection fails to point at sharp boundaries the opponent would be violating; plus, it may also be unfair—and thus dialectically ineffective—if it fails to recognize such metaphysical limitations, insisting that the subject matter is determinate when there are good reasons to think it is not. On the other hand, the trans-exclusionary theorist may be restricting the subject matter so that the QUD presupposes a biology-based conception of gender and sexuality categories. If the nature of gender and sexual orientation is indeterminate in the actual world, this is unfair—and therefore dialectically ineffective—to the extent that the restriction is not metaphysically justified, apart from the strategic contextual convenience of the contender.
Second, the dialectic is defective for the party who responds to the charge too. Andler could easily respond to the subject-change objection that since subject matter is indeterminate, there is no subject matter to change, so the objection misfires. Still, this is not his response: the point he seems to makeFootnote 3 (pp. 62–63) is that his strategy helps better capture the subject matter, to the extent that it illuminates sociopolitical categories that perform a valuable explanatory task and are altogether absent in Stock’s framework. This, however, may seem confusing and ultimately misguided if we accept that the subject matter is indeterminate, both in its coarse-grained and fine-grained versions. It may also be considered an unwarranted move, since if the subject matter is indeterminate, Andler may be seen as strategically restricting or shaping the subject matter in the way that best serves his strategy. Overall, Andler’s response seems ineffective too, since if the subject matter is indeterminate, then it is either factually wrong (as there is no determinate subject matter to be captured) or introduces some unwarranted restriction or structuring.
Before closing this section, I should devote a few more lines to different ways in which Andler’s reply (“I am better capturing the subject”) could be read, to clarify why I do not think they would be viable.Footnote 4
First, one could read Andler in the following way: “I am not changing the subject; rather, I am offering a better theory about some key features of the phenomenon under discussion.” Indeed, one could see Andler as using the concepts Woman and Lesbian with the same functional role as the trans-exclusionist, and also as addressing the same broadly construed question (whether trans women should be excluded from lesbian spaces), which would seem to suggest that he is staying on the same subject matter. Meanwhile, he would be providing reasons to think that the true theory of gender and sexuality categories is sociopolitical, thus restricting the partition of logical space to worlds where these categories are sociopolitically grounded. Yet if we agree that there is no descriptively adequate theory of said categories because the evidence suggests that the subject matter is indeterminate, this move would be descriptively unwarranted. Of course, it could be a proposal for construing gender and sexuality categories as sociopolitical. I have a lot of sympathy for this interpretation. However, it requires embracing a different constructionist conception of gender and sexuality categories, which the opponent might resist. It is not clear to me that someone who endorses such a constructionist conception would still stay “on topic” in a conversation with a trans-exclusionist, who might well endorse a non-constructionist conception. The two parties might occupy positions so distant from one another, even methodologically or metaphilosophically, that their chances of recognizing the other’s move as legitimate and topically relevant may be practically nil.
Second, one could read Andler in the following way: “I am indeed changing the subject; plus, I am focusing on a different and better subject.” If Andler were to regard the trans-exclusionist’s question as a loaded question, such as whether trans women should be excluded from lesbian spaces provided a biology-based conception of lesbianism, this would make sense. That loaded question would require countenancing a partition of possible worlds that someone like Andler would not include in their ontology, so to speak, since the set of worlds in which one could answer that loaded question would, from his perspective, be empty. In this scenario, it would make sense for him intentionally to change the subject and introduce a partition and a related QUD linked to possible worlds in which the metaphysics of gender and sexuality is sociopolitical. Again, I think this move would not be descriptively warranted if we concede the indeterminacy that surrounds gender and sexuality categories. It could be interpreted as a proposal to construe gender and sexuality categories in this way, and once again I have a great deal of sympathy for this interpretation, as Section 7 will make clear. Yet if this is a perspicuous way of reading Andler’s reply, it seems confusing to articulate it by claiming that “I am better capturing the subject matter,” since this statement would suggest a common subject matter between him and the trans-exclusionist—which simply cannot obtain in this scenario. I will have more to say about this second reading and the broader framework within which it should be situated, in Section 7.
To sum up, I started this section arguing that we have good reasons to deem the subject matter of the exchange considered by Andler indeterminate. Against this backdrop, I have argued that the dialectic generated by the change of subject objection and Andler’s response is defective. The subject-change allegation may be seen as ineffective or even unfair, and Andler’s actual reply may be seen as either factually wrong or unwarranted too.
7. Setting the Dialectic Back on Track: The Negotiation Option
So far, I have argued that the dialectic engendered by the subject-change objection considered by Andler would be defective, provided that the subject matter in the exchange is indeterminate and the parties assume there is a determinate subject matter to be captured.
Yet the dialectic can be set back on track if we switch to a different way of reading the exchange. It may be that the subject matter itself is indeterminate and that the parties are, in part, negotiating how to build its structure and boundaries alongside other theoretical and ethical issues (each on the basis of their own background, methodological, and metaphilosophical approaches). Indeed, Kevin Richardson has argued that discussions of gender and sexual orientation often, if not always, involve ethical and metaphysical negotiation. My proposal is that this negotiation can extend to the subject matter itself.
7.a. Richardson on negotiating perspectives
In Chapter 7 of his The End of Binaries, Kevin Richardson argues that disputes about gender are not disagreements about how things are in the world out there, but complex negotiations that intertwine metaphysics, language, and ethics. He challenges the idea that gender terms simply describe a fixed reality. Instead, he maintains that the way we use gender language actively shapes social structures, norms, and expectations. As a result, debates over gender categories (such as whether trans or non-binary identities “fit” within the categories of man or woman) are not just about truth and capturing how things are, but about how things should be in our way of collectively organizing these matters.
Richardson’s key claim is that when we argue about gender, we are not just debating facts; we are negotiating between competing perspectives (pp. 143–151). Gender statements express underlying bundles of social norms—for instance, as to what counts as a woman, how people should be treated, and so on—so disagreement reflects a clash of these normative frameworks. When one person says “trans women are women” and another denies it, they are not simply describing reality differently, as if there were an underlying fact of the matter to be mirrored. Rather, they are expressing different preferences about how gender categories should be structured and how society should operate. These perspectives function like proposals or plans for organizing social life. Through language, people try to get others to adopt their perspective, and when enough people accept one, it helps shape actual gender categories and norms. In other words, negotiating gender is ultimately tantamount to negotiating which perspective—and which set of related social norms—should guide our shared social world.
7.b. Negotiating subject matter, too
Seen through the lens offered by Richardson’s account, a disagreement such as whether trans women can be lesbians (and similar exchanges) may simultaneously constitute a negotiation over metaphysics, ethics, and semantics in the trans-related discourse. My claim is that, under a broad understanding of semantics, this may include a negotiation of the exchange’s subject matter itself.
My argument in support of the latter thesis draws from the dialectical difficulties about subject matter ascertained in Section 6. As we have seen, there are reasons to think that, in the present case, the subject matter shared between the trans-exclusionary argument and Andler’s argument is indeterminate, as it is too coarse-grained and attempts to make it more determinate fail. Still, the trans-exclusionary theorist seems justified (from their own point of view) in maintaining that the opponent is changing the subject, while the opponent seems within their rights to maintain that the subject is fine as it is or it is better captured by their own proposal. Each party’s contention seems justified, but it is not clear that the back-and-forth generated by the ensuing dialectic can find a resolution, because indeterminacy prevents settling the descriptive question of what the exchange’s subject matter is. Is it whether trans-women should be excluded from lesbian spaces tout-court, whether they should be excluded from lesbian spaces under a biology-based understanding of lesbianism, or whether they should be excluded from lesbian spaces under a sociopolitical understanding of lesbianism? The result is a tension between (i) indeterminacy of subject matter on the one hand, and (ii) apparent justification by the parties to believe that the exchange has a subject matter the opponent is missing. How to reconcile (i) and (ii)?
A way out of this predicament consists in seeing the exchange between the trans-exclusionary theorist and Andler as a negotiation not only about the metaphysical and ethical question of whether trans women are women, or whether they can be lesbians, or whether they should be excluded from certain spaces, but also about what the very subject matter of the exchange should be.
By producing their argument in favor of the exclusion of trans women from lesbian spaces, the trans-exclusionary theorist is not just recommending that being trans be incompatible with being a woman or with being lesbian, and that trans people be treated in certain ways; they are also recommending that the very question about the exclusion of trans women from lesbian spaces be framed in a certain way—namely, presupposing a biology-based conception of lesbianism. This is a proposal about how to structure the very QUD, which is a straightforward way of identifying the exchange’s subject matter. According to the Lewisian framework introduced in Section 5, this would be tantamount to shaping the partition T relevant for assessing topicality so that, in all its possible worlds, the metaphysics of gender and sexuality is biology-based, thus proposing that the space of possibilities under consideration be metaphysically structured in a certain way. Conversely, Andler’s rebuttal, whereby his own argument better captures the subject, may also be read as a proposal that the QUD should be framed differently—namely, presupposing a sociopolitical conception of lesbianism and of sexuality concepts generally. Translated into Lewisian terms, this would be tantamount to shaping the partition T relevant for assessing topicality in such a way that, in all its possible worlds, the metaphysics of gender and sexuality is determined at least in part by sociopolitical facts, thus proposing that the space of possibilities under consideration be metaphysically structured in a different, incompatible way.
Based on this reading, points (i) and (ii) can be reconciled in the following way. The Richardson-style view I suggest, whereby the exchange is not only a negotiation of metaphysics and ethics, but also a negotiation of its own subject matter, would nicely explain why (in accordance with (ii)) both parties seem justified in their contention that the other is “changing the subject,” despite indeterminacy of subject matter (point (i)). Such an appearance of justification would obtain on account of the normative force of each contender’s contribution to the debate. It is because the trans-exclusionary theorist is proposing a certain subject matter, shaped by the wider perspective they are putting forward, that they appear justified in their complaint that the opponent is deviating from that subject matter. And it is because the opponent is also making a normative move, proposing a different subject matter, that they appear justified in defending the viability of their move. If each party is advocating a different way of shaping the QUD (and the attendant partition of logical space), an allegation such as “You are changing the subject” makes sense to the extent that it can be paraphrased as “You are switching to a different subject than the one I am proposing.” The rebuttal, phrased by Andler as “I am better capturing the subject,” also makes sense to the extent that it can be paraphrased as “The subject matter I am proposing is better.”
One may object that it is absurd to think that a conversation may occur without a determinate subject matter and that the subject matter itself may be under negotiation in parallel with the unfolding of the conversation. Yet arguably this is something one may just have to accept, as this may perfectly well happen if the very metaphysics of a certain domain of discourse is partially or wholly dependent on linguistic or conceptual negotiation for its construction. Indeed, I would argue this rather straightforwardly follows from Richardson’s theory: if we negotiate the very metaphysics of the things we talk about when we discuss gender and sexual orientation, there is a relevant sense in which we simultaneously negotiate what it is that we are talking about—and that is none other than the subject matter of our linguistic exchanges. Thus, perhaps we should revise our idea that the subject matter of a conversation cannot be constructed “on the go,” as it were, for this result seems to follow clearly from an intelligible and insightful view such as Richardson’s on the metaphysics of gender and sexual orientation.
Going back to the dialectic generated by the change of subject, we can therefore say that the dialectic can be set back on track. We can read the change of subject objection as the trans-exclusionary theorist pointing out that the opponent is deviating from the subject matter initially established by the trans-exclusionary argument by means of an implicit act of proposal or recommendation of a plan. Conversely, we can read Andler’s reply as a counter-proposal for a different (better) subject matter or question to be discussed.
One may applaud the new reading but still express a worry: if the structure of the conversation is such that even its subject matter is under negotiation, then the “changing the subject” charge loses much of its force. There is no fully determinate subject to begin with, and if part of the aim of the dispute is precisely to establish the structure, extension, and boundaries of that subject matter, then it seems that changing the subject (a) is not really possible because there is no unique, stable subject and (b) is something that will happen at every turn in the debate, at least until the negotiation is over and one party’s preference (if any) has prevailed. This is partly true: changing the subject stops being a fully objectionable move if the subject matter itself is under negotiation and each party is trying to get their perspective accepted by the other.
Still, it may be objectionable to change the subject (or even not to change the subject!) for other reasons, especially those intertwined with ethical and practical considerations. There may be a question that is “more worthy of our discussion” because it frames the issue in a way that is, all things considered, better for our society, so one subject matter may prevail over another because of its superiority in broadly normative terms. My own intuition, for example, as a philosopher of language who does not work directly on gender-related issues and is not especially engaged in activism, and who identifies as a cisgender, heterosexual woman, is that framing a question such as whether trans women should be excluded from lesbian spaces in rigid, black-and-white terms risks prejudging the range of possible answers. In particular, it may foreclose responses that expand our conceptual horizons and challenge entrenched ways of thinking, thereby inhibiting a more open attitude toward forms of nonconformity with respect to fixed and stereotyped roles. I take it that similar motivations could be articulated more rigorously and in a more nuanced way by scholars who specialize in gender studies or feminist analytic philosophy and who are more deeply engaged with these debates.
Regardless of my own (admittedly somewhat inchoate) position, the central takeaway I wish to emphasize is this: the dialectic between the trans-exclusionary theorist and the sociopolitical theorist (such as Andler) may be seen as a competition between the ethical, practical, social, cultural, and political viability of one “question-to-be-under-discussion” (the question that presupposes a biology-based account of sexuality categories) and another “question-to-be-under-discussion” (the question that presupposes a sociopolitical account of sexuality categories), alongside being a competition between perspectives or plans about how to organize our shared social life—at least according to Richardson’s view. The debate is complex and multi-layered, requiring one to take a position along several dimensions—not only metaphysical and ethical, but also broadly semantic and conversational.
8. Conclusion
This paper has examined the dialectic between Andler’s critique of trans-exclusionary arguments and the subject-change objection it seems to invite. In this dialectic, both parties seem to presuppose (not at all irrationally) that there is a determinate subject matter that can be straightforwardly identified and either preserved or altered. In fact, as I have endeavored to show, there are strong reasons to think the subject matter of the exchange is indeterminate: while a coarse-grained QUD is shared, it is metaphysically problematic, and attempts to render it more fine-grained are metaphysically unwarranted.
Against this backdrop, subject-change objections and replies to them generate a defective dialectic. If there is no independently fixed subject matter, then accusing one’s opponent of changing it is either ineffective (because it fails to identify any clear boundary crossed) or unfair (because it surreptitiously imposes an unwarranted restriction). Similarly, responses that claim to “better capture” the subject matter are problematic if there is no determinate subject matter to capture in the first place.
To recover the tenability of the dialectic, I have proposed a reinterpretation of the exchange as a negotiation, drawing on Richardson’s account. Richardson tells us that disputes about gender and sexual orientation are not merely about describing a pre-existing reality, but about shaping the metaphysics, ethics, and semantics of the relevant domains of discourse. Extending this idea, I have argued that such negotiations can include the subject matter of the discourse itself. On this view, the competing positions would not simply disagree based on a fixed QUD (or partition of logical space); rather, they may simultaneously put forward rival proposals for what that very question or partition should be (as well as rival views as to how to answer it).
This perspective allows us to make sense of the apparent legitimacy of both the subject-change objection and the response to it, despite the underlying indeterminacy. Each party can be seen as advocating a different way of structuring the QUD or the attendant partition of logical space, and thus the exchange’s subject matter. In this scenario, changing the subject may not be an overly problematic move, because the subject matter itself is unstable after all. Yet changing (or preserving) the subject matter may become a normatively relevant question. Some “questions under discussion” may turn out to be ethically or practically more worthy of consideration than others.
Acknowledgments
I am grateful to Matt Andler, Kevin Richardson, and Ben Caplan for involving me in this project. I also greatly benefited from feedback provided by the participants at the CJP/Madrid Symposium (4–5 June 2026, at SLU-Madrid). Work on this paper was supported by grant RYC2022-035705-I, funded by MCIN/AEI and “NextGenerationEU”/PRTR (FSE+), and by grant PID2024-157224NB-I00, funded by MICIU/AEI and FEDER, UE.
Competing interest
I confirm that I have no affiliation with any organization that has a direct or indirect financial interest in the subject matter discussed in the manuscript. I have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
AI statement
AI tools were used during the preparation of this manuscript for language editing and stylistic refinement. No AI tools were used to generate original research findings or substantive arguments.
Delia Belleri is a Ramón y Cajal Fellow at the Institute of Philosophy of the Spanish National Research Council (CSIC) in Madrid. Her research focuses on philosophy of language (especially semantics vs. pragmatics, conceptual engineering, and metalinguistic negotiation) as well as metaphilosophy (especially metaontology). She is currently co-PI of the project “CECODISP: Conceptual Engineering and Communicative Disruptions: between Semantics and Pragmatics” (PID2024-157224NB-I00).
Open access
