Gender as a Self-Conferred Identity Feminist Philosophy Quarterly Volume 8 | Issue 2 Article 3 Recommended Citation Real, Michael. 2022. “Gender as a Self-Conferred Identity.” Feminist Philosophy Quarterly 8 (2). Article 3. 2022 Gender as a Self-Conferred Identity Michael Rea University of Notre Dame michael.rea2@gmail.com Rea – Gender as a Self-Conferred Identity Published by Scholarship@Western, 2022 1 Gender as a Self-Conferred Identity Michael Rea Abstract This paper develops and defends the view that gender is an identity that we confer upon ourselves. The claim that gender is a self-conferred identity is not novel, but its metaphysics is obscure at best. What exactly is an identity, and how do we manage to confer identities upon ourselves? Furthermore, how does the claim that gender is a self-conferred identity comport with the widely accepted notion that gender is also a social identity and that social identities are (at least partly) either conferred upon us by others or constituted by the social positions we occupy? This paper articulates a metaphysics for the self-conferred-identity account that addresses these questions. The most important advantage of the view is that, in contrast to other realist theories about the metaphysics of gender, this one transparently offers a basis for assigning first-person authority to people’s judgments about their own gender. Keywords: gender, identity, narrative, self, conferralism, social identity, social ontology In recent years, there has been an explosion of interest in the metaphysics of gender, and a proliferation of different views about the nature of gender and what (if anything) determines one’s gender attributes—most saliently, the two dominant gender attributes, being a woman and being a man. Broadly speaking, these views can be organized in accord with their answers to the following four questions: (i) Are gender attributes real? (ii) Are gender attributes essences, or natures of anything? (iii) Are gender attributes socially constructed? (iv) By virtue of what does a person belong to a gender category? Is it, for example, a matter of having a certain kind of body, occupying a certain social position, having a particular behavioral and characterological profile, having a certain belief or cluster of beliefs about oneself, or something else? Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 3 Published by Scholarship@Western, 2022 2 The question of what determines the application conditions for gender terms (“man” and “woman” most saliently) is closely related to these and has also received a lot of attention. But as Elizabeth Barnes (2020) has persuasively argued, there is no reason to expect that a metaphysics of gender will necessarily settle the application conditions for gender terms. Accordingly, I will leave the linguistic question aside in what follows. In this paper, I develop and defend the view that gender is an identity (in a sense to be explained later) that we confer upon ourselves. As I shall explain in what follows, this “self-conferred identity” (henceforth, “SCI”) account is a view according to which genders are real but are not essences or natures of anything; genders are socially constructed; and—for the most part, but with some exceptions to be explained later—one belongs to a gender category just if that gender is included in what I shall call a person’s “autobiographical identity.” It is, in other words, a realist, social constructionist, antiessentialist theory of gender.1 Its most important advantage is that, in contrast to other realist theories about the metaphysics of gender (essentialist and antiessentialist alike), this one transparently offers a basis for assigning both epistemic and ethical first-person authority to people’s judgments about their own gender.2 Accordingly, to the extent that it can be adequately 1 Realism is commonly characterized with reference to “mind-independence” in such a way that what is real contrasts almost by definition with what is socially constructed. For this reason, “realist, social constructionist” may sound oxymoronic to some. But especially in the literature on social metaphysics, there has lately been resistance to the idea that realism contrasts with social constructionism. (See, e.g., Mallon [2016, 137–61] for detailed discussion of this issue, and for a helpful distinction between different kinds of “mind-dependence” and correspondingly different kinds of constructionism.) For purposes here, a theory of gender qualifies as realist just in case it treats gender as a real, genuine attribute of things rather than (say) a pretended or in some other way fictional attribute. Thus, social constructionist views of gender can be realist theories of gender, whereas eliminativist and fictionalist theories of gender cannot be. 2 S’s gender judgments have epistemic first-person authority just if S’s judgments about S’s gender are epistemically privileged over third-party judgments. They have ethical first-person authority just if it is ethically inappropriate for third parties to attribute to S a different gender from what S attributes to themselves. (See Bettcher [2009] on this distinction.) To my knowledge, the only other metaphysical theory of gender that manifestly grants epistemic first-person authority to gender judgments is Heather Logue’s (forthcoming) fictionalist theory of gender; but whatever its merits might be, that is (obviously) not a realist theory. Bettcher (2009) claims only ethical first-person authority for gender judgments; and likewise, Katharine Jenkins (2018) Rea – Gender as a Self-Conferred Identity Published by Scholarship@Western, 2022 3 developed and defended, it will have claim to being the most accommodating of trans and nonbinary gender identities. The claim that gender is a self-conferred identity is not novel. It is currently the dominant understanding of gender in women’s, gender, and sexuality studies, and it is rapidly acquiring the status of “common sense” in certain important demographics. But its metaphysics is obscure at best. What exactly is an identity, and how do we manage to confer identities upon ourselves? Furthermore, how does the claim that gender is a self-conferred identity comport with the widely accepted notion that gender is also a social identity and that social identities are (at least partly) either conferred upon us by others or constituted by the social positions we occupy?3 My goal in this paper is to articulate a metaphysics for the SCI account that addresses these questions. In doing this, I take myself not to be providing a novel “ameliorative” account of gender but rather to be fleshing out and lending support to an already widely accepted thesis about gender by providing an adequate metaphysical underpinning for it. My plan is as follows. I will begin by offering an account of what it means to say that gender is an identity, and then I will go on to explain both how we confer it upon ourselves and why we enjoy first-person authority with respect to judgments about our own gender. I will conclude by explaining why, despite appearances to the contrary, I think the SCI account qualifies as a social constructionist view, and by noting points of similarity and contrast between the SCI account and nearby alternative views in the literature. It will emerge in the section that develops the view that there is an important connection between the identities we confer upon ourselves and the narratives we have about ourselves. Although I will not develop this point beyond simply noting it here, I take it that the connection between the SCI account and the idea that identities are narratively constituted offers at least one further advantage to the view—namely, it can explain important aspects of gender oppression in terms of what Hilde Lindemann Nelson (2001) calls “infiltrated consciousness.” Identity and Self-Conferral In metaphysics, the term “identity” is most often used to invoke the concept of numerical identity. But here I want to work with two different notions of identity, one of which is prevalent in the philosophical literature on narrative identity and in the psychological literature on identity and identity development, and the other of acknowledges that the norm-relevancy account of gender identity developed in her work (Jenkins 2016, 2018) accords only ethical first-person authority to gender judgments. 3 See Ásta (2018) and Haslanger (2000). Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 3 Published by Scholarship@Western, 2022 4 which is prevalent in social metaphysics and related fields. In the relevant psychological literature, an identity seems just to be a mental construct of some sort—perhaps a representation or concept of oneself, or an internalized narrative about oneself—that brings “a sense of coherence and integration to one’s life, allowing [one] to perceive a sense of continuity through time” (McLean 2015, 19). In social metaphysics and related fields, an identity seems just to be (very roughly) something central about who a person is.4 For example, I am a tennis player, but being a tennis player is not really an “identity” of mine in this latter sense because it is not all that central to who I am. By contrast, the attributes that place me in specific professional, racial, and religious categories are identities of mine. These category- memberships are quite central to who I am (regardless of whether I want them to be), and so likewise for most, even if not necessarily all, people. To help keep these two notions distinct, let us call the first kind of identity a representational identity, and let us call the second kind of identity a social identity. My claim in this paper is that (unlike most other social identities), gender is a self- conferred identity in both senses. Thus, it is a self-conferred representational identity and a self-conferred social identity. But what is self-conferral? And how is it related to mere self-attribution? Most of our social identities are ones we acquire by resembling or differing from other people in salient ways, by being perceived or understood in certain ways, by standing in particular relationships to social structures and institutions, by interacting with our environment in certain ways, or by some complex combination of such factors. For example, having a religious identity is, for most people, at least partly a result of personal choice, but in many cases, it also depends to some extent on one’s relationship to the institutional structure of the religious tradition and perhaps also on the perceptions or beliefs of certain religious authorities. Conversion to Catholicism, for example, requires not only a personal decision but also a willing individual to perform the sacrament of baptism. Likewise, on some of the most influential accounts of disability, being disabled doesn’t depend solely on features of one’s own body, but on various relational facts, such as the statistical differences between one’s own body and the bodies of others in larger society, or the wide variety of social factors that explain why people sharing one’s own particular differences have reduced access to public spaces, certain kinds of jobs, or other social goods. 4 I do not mean to suggest that the facts about what is most central to a person are somehow out there in the world independently of anyone’s subjective interests or of different people’s conceptions of the person. On the contrary, I am inclined to suppose that all facts about what is most central to a person—what count as their most central values, desires, and so on—are relativized to different conceptions (or, as I’ll put it below, interpretations) of the person. Rea – Gender as a Self-Conferred Identity Published by Scholarship@Western, 2022 5 When the beliefs and judgments of others are not only salient in determining whether one has a particular social identity but are more salient than one’s own beliefs and judgments, it is natural to say that the identity in question is conferred by others. Being one of the popular kids, for example, is plausibly like this. Even if one’s own personal choices—like one’s decisions about what to wear, what activities to participate in, and so on—make a significant difference in whether one is popular, ultimately it is the judgments of others that are most salient to enjoying that status. By contrast, where one’s own choices both partly determine whether one has the identity and are more salient than the beliefs and judgments of others, the identity is self-conferred. Being an atheist and being a Star Wars fan are plausibly like this. Obviously, then, self-conferral is different from mere self-attribution. Sadly, no amount of attributing to myself the property of being a guitar player whose skill rivals that of Eddie Van Halen will manage to confer that property upon me. Attributing a property to oneself is simply a matter of representing oneself (by way of belief or some form of nondoxastic mental representation) as having the property. By contrast, conferring a property upon oneself (or upon someone else) is a matter of doing something—perhaps simply holding certain kinds of beliefs, or perhaps doing something other than or in addition to believing certain things—that makes it the case that I have (or someone else has) the property. So this is what I mean when I say that gender is a self-conferred identity: (a) a person’s gender—or, in terminology that will emerge as important later in the paper, a person’s true gender—is at least partly dependent on human beliefs, decisions, or representations of who they are, and (b) one’s own beliefs, decisions, or representations are more salient in determining their gender than anyone else’s. That said, however, none of this tells us very much about what kind of thing an identity (representational or social) is, or about what exactly it takes for an identity to be conferred upon oneself. The remainder of this section focuses on the nature of identities. In the next section, borrowing in significant ways from Ásta’s (2018) more general conferralist account of social properties, I take up the question of what it takes to confer gender upon oneself.5 For purposes here, I propose to understand a representational identity as a certain kind of representational content. In particular, and in accord with the characterization quoted from McLean above, it is the content of a representation that not only identifies salient attributes, social roles, and life events of the person being represented, but also contextualizes them, providing some clue as to their centrality 5 Note, however, that in borrowing from Ásta’s account in the ways that I do, and in characterizing (most) social properties as conferred upon us either by ourselves or by others, I do not mean to signal general agreement with Ásta’s conferralist account. See Rea and VanKammen (2021) for a variety of objections against Ásta’s account. Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 3 Published by Scholarship@Western, 2022 6 and significance to the person being represented, to those doing the representing, or to the various communities to which the represented or representing individual(s) belong. On this way of thinking, a representational identity is a kind of interpretation of a person or of the events known to be part of a person’s life. It is, furthermore, an interpretation that captures something central about who a person is in the eyes of whoever is doing the interpreting, since otherwise it is hard to see how it could play the sort of unifying, integrating, and contextualizing roles described above. We can distinguish (at least) two kinds of representational identity: autobiographical identities and social-representational (SR-) identities. An autobiographical identity captures the central elements of who one is to oneself, as it were, whereas a social-representational identity captures a central part of who a person is to others. An autobiographical identity is most plausibly identified with some portion of a person’s most all-inclusive self-representation—what we think of commonsensically as their “self-conception,” and what I have elsewhere (Rea forthcoming; following Ismael 2016, ch. 3) called their autobiographical point of view. In particular, a person’s autobiographical identity is the content of the most comprehensive portion of their total self-conception that plays exactly the sort of unifying, self-interpretive role described in the preceding paragraph. An SR-identity, by contrast, is not plausibly identified with a portion of their own self-conception. This is because a person’s self-conception can fail— catastrophically, even—to represent who they are to others. The autobiographical identity of an alcoholic, for example, might well represent them as a reliable coworker, a reasonable parent and romantic partner, and a person with excellent self- control, whereas who they are to others might be quite the opposite. Accordingly, an SR-identity is more plausibly identified with the content of what we might call a “collective representation” on the part of others that plays a similarly unifying, integrating role in a (presumably tacit) broader collective representation of who we are. I have no account of collective representation to offer, but I can at least say that the notion should not be understood as necessarily implying the existence of anything like a group mind in which the representation is instantiated. Rather, it should be understood in accord with whatever one takes to be the best account of collective belief or collective intention. Perhaps such collective “mental states” are best understood as emergent properties of a kind of group mind, or perhaps instead they are best understood reductively. I do not want to take a stand on the metaphysics of (alleged) collective mental states; but I do want to assume, as seems quite plausible, that talk of collective representation makes sense and can be understood along the same lines.6 6 A referee points out that the notions of collective belief and collective intention are not unproblematic, and so we can expect an account of collective representation to Rea – Gender as a Self-Conferred Identity Published by Scholarship@Western, 2022 7 As might be guessed from the labeling, there is a close connection between SR-identities as I understand them, and social identities. An attribute is a social identity of X’s in a context C just in case it is among the interpretively significant social attributes included in X’s SR-identity in C. Obviously this allows for variation across contexts not only in what social identities a person has but also in what sorts of attributes count as social identities. This allowance will surely be controversial, but I think it is the right result. To take just one example: in a society that does not recognize racial distinctions, it seems correct to say that race is not a social identity— and this regardless of whether race is in fact a real attribute. One might object that this way of understanding social identities yields the wrong results in some cases. In the Batman fiction, for example, one might well doubt that Bruce Wayne has being a superhero as one of his social identities, and this despite the fact that being a superhero is a central and socially significant attribute in the largest collective representation of the man who is Bruce Wayne, and one that will (apparently) play the sort of unifying, integrative role in the collective understanding of him that a social identity is supposed to play. So it might seem that the present account of social identity is on the wrong track. But in fact I think this objection can be addressed by attending to a distinction between de re and de dicto representational content. In general, it seems we can distinguish between what a representation of an object O directly attributes to O on the one hand, and, on the other hand, what sentences about O the representation represents (explicitly or tacitly) as being true. The former is what I have in mind when I talk about de re representational content; the latter is what I have in mind when I be similarly troublesome. Perhaps, then, we would do better (as the referee suggests) to develop an account in terms of a notion like “common knowledge”? (See Mallon [2016, 258ff], following Lewis’s [1969] understanding of that notion.) Perhaps, but I am doubtful. In my view, the notion of common knowledge as Mallon understands it is not well-suited to the present purpose. What is needed for present purposes is a notion of representation within a group that allows for the attribution of the representation to the entire group even in cases where some members of the group are not themselves hosting that representation. (E.g., perhaps the representation is a belief shared by most members of the group but not all of them.) Absent this allowance, the list of social identities attributable to any given person will be objectionably thin. The concepts of collective representation and belief allow for such possibilities, though it is admittedly difficult to accommodate them in a precise and plausible theory of collective belief or representation. But the concept of common knowledge as Mallon understands it does not make any such allowance. Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 3 Published by Scholarship@Western, 2022 8 talk about de dicto representational content.7 Applied to the Batman/Bruce Wayne case, it seems clear that the attribute of being both Bruce Wayne and a superhero is not included in the de dicto content of the collective representation (in the Batman fiction) of the man who is both Batman and Bruce Wayne. For, after all, virtually nobody in the Batman universe would affirm the sentence, “Bruce Wayne is a superhero”; and this is why it seems that being a superhero is not among his social identities. But being a superhero is clearly part of the de re content of the collective representation of that man, as is being a wealthy nonsuperhero.8 That is, even though nothing in the collective representation of Bruce Wayne represents as true the sentence “Bruce Wayne is a superhero,” the content of that collective representation does represent the man himself as a superhero under the guise of Batman and as a wealthy nonsuperhero under the guise of Bruce Wayne. Thus, the response I propose to the objection is that what contributes to an attribute’s being a social identity of a person is not its being part of the de dicto representational content of a collective representation of the person, but rather it’s being part of the de re content of such a representation. Being a superhero is, on this account, a social identity of the man Bruce Wayne because that attribute is a central part of who he himself is to many people, even if almost none of them know that the Batman persona is one of his own personae. In sum, then, I am offering the following account of what it is to have an attribute as a social identity. Whereas a person’s autobiographical identity is an internalized, self-interpretive representation of themselves that is focused on the attributes that play a central, unifying role in their own total self-conception, a social- representational identity of a person is an external, interpretive representation of them that is focused on one or more of their socially significant attributes. An attribute is a social identity of the person in a context just if it is one of the focal 7 A referee has objected that “representations are inherently de re.” This seems not quite right to me—cf. “the semantically de re/de dicto distinction” described by Michael Nelson (2019). But, be that as it may, I recognize that the notion of “de dicto representational content” might at least sound rather odd. What matters to me, though, is not so much the terminology as the distinction I have just drawn here, which seems not at all odd and, in the present context, is importantly helpful. 8 A referee raises an objection that seems to suggest that those who represent Batman in this way—as both being a superhero and not being a superhero—will have inconsistent beliefs about the man. But this doesn’t follow, and for familiar reasons. Although they believe of Batman/Bruce Wayne that he is a superhero and that he is not a superhero, they do not believe of anyone that they are both a superhero and a nonsuperhero, nor would they affirm any sentence that implies as much. Rea – Gender as a Self-Conferred Identity Published by Scholarship@Western, 2022 9 attributes their SR-identity in that context.9 People share a social identity in a context just when there is overlap in their socially significant attributes and when their SR- identities in that context reflect the same interpretation of those attributes. For the most part, social identities will be relatively simple, socially salient attributes like gender, race, professional identity, religious identity, and the like; but in some cases, they may be highly complex attributes of the sort we are inclined to describe as full- blown personae (as in the case of the Batman persona). One welcome consequence of this way of thinking about social identity is that it accommodates the commonly held view that at least some social identities change from context to context. On social-position accounts of race, gender, and disability, for example, whether one counts as disabled, Black, or a woman might well vary from context to context. The present account of social identities allows for this by virtue of allowing that, among the many different collective representations of a person that exist in the world, the one most salient for determining a particular social identity is just the one tied to the context that our preferred account of the identity in question specifies as most salient. However, an initial concern with the account I have offered is that it is singularly ill-suited to accomplish the task I have set for it in this paper—namely, to support the idea that gender is a self-conferred identity. The reason is that, as I said at the outset of this section, our self-conception can diverge from who we are to others; and so it would seem that, on the present account, if our gender is a social identity, and is therefore a part of who we are to others, there is no particular reason to think that our gender would have to conform to our own conception of our gender. I take up this concern in the next section. Gender as Self-Conferred We do not unilaterally confer social identities upon ourselves (although we might contribute to their conferral by being part of the group whose collective representation of us is salient for conferring one or more of our social identities). This is both intrinsically plausible and a straightforward consequence of the account of social identity offered in the previous section. However, I maintain that gender is a social identity; gender is self-conferred; and furthermore, we have first-person authority over our gender. How can this be coherent? The answer is to draw a distinction between social gender, which is a social identity conferred upon a person in a social context by their SR-identity in that 9 This view bears some affinity with Ásta's (2018, ch. 6) account of social identities as locations on a social map. See also my recent article, “The Metaphysics of the Narrative Self” (Rea forthcoming) for fuller treatment of the metaphysics of narrative identities and selves. Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 3 Published by Scholarship@Western, 2022 10 context,10 and autobiographical gender, which is both self-conferred and part of one’s total autobiographical identity.11 Obviously this view allows that one’s social gender might differ from one’s autobiographical gender, and this, in turn, raises the question of which (if either) identity is “privileged”—or which counts as one’s real or true gender identity—when there is conflict between the two. One possibility is that everyone has (at least) two gender identities, their social gender and their autobiographical gender, and neither is any more “true” to who they really are than the other one. Given my interest in developing a realist account that preserves first- person authority over judgments about one’s own gender, however, it should come as no surprise that I reject this possibility, arguing instead that autobiographical gender is privileged in the sense that one’s true gender is one’s autobiographical gender.12 Accordingly, on the view that I shall defend, gender simpliciter is to be identified with autobiographical gender and a person’s social gender in a context reduces to facts about the (perhaps accurate, perhaps inaccurate) collective representation of the person in that context. Before turning to that defense, however, I want to address two preliminary objections. First, one might object that if what I am calling “social gender” simply reduces to facts about collective representation, then it doesn’t really deserve to be described as a kind of gender. We should instead just say that it is a collective attribution, or misattribution, of gender to an individual by others. Second, one might worry that what I say below could successfully be adapted to establish that one has 10 I have no theory about how, or under what conditions, a group of people with divergent individual representations of a person’s gender will manage to collectively represent them as being of one particular gender, but presumably this will work in much the same way as collective belief in cases where there is a lack of unanimity at the level of individual belief. 11 The distinction between social gender and autobiographical gender bears some resemblance to Katharine Jenkins’s (2016) distinction between “gender as class” and “gender as identity.” However, insofar as her conception of gender as class is modeled on Haslanger’s (2000) social-position account of gender, it is both narrower and less determined by collective representations than social gender as I conceive of it. For discussion of how her account of gender identity differs from my own, see section 3 below. 12 I take it that one way to be an antirealist about gender is to suppose that there is no fact of the matter about what gender attribute(s) a person “really” has. Thus, if it turns out that neither autobiographical gender nor social gender specifies a person’s true gender—and if those are the only candidates for being a person’s true gender— then realism about gender is false. Rea – Gender as a Self-Conferred Identity Published by Scholarship@Western, 2022 11 first-person authority over judgments about all of their social identities, including, for example, their racial identity. What shall I say about these concerns? I am not wedded to the label “social gender,” and I am already committed to the idea that real, true gender is not social gender; so there is little cost in simply conceding the first objection. Doing so would require me to give up the claim that some real kind of gender is a social identity. But that is no big loss given that what I am calling “social gender” would remain a social identity, albeit under a different label. But insofar as what I am calling “social gender” very often aligns with real gender and, furthermore, typically has important gender-like social consequences, describing it, as I have, as a kind of gender seems entirely apt. As for the second concern, I think the worry is misplaced. On the view that I am developing, gender is not only a social identity but a representational identity; and, as shall emerge, it is its status as a representational identity that allows me to defend the claim that autobiographical gender is more authoritative than social gender. By contrast, race is not a representational identity, at least not according to either folk belief or the most prominent theories of race. Both folk belief and the most prominent theories of race seem to agree that if race is real at all, it is determined by some combination of non-self-conferred attributes: facts about social position, ancestry, culture, or some combination of these or other non-self-conferred and nonrepresentational traits.13 Nobody that I am aware of thinks that there is nothing more to race than the various ways in which a person is racially interpreted, which is, in effect, what I am claiming about gender. I turn now to defending the claim that autobiographical gender trumps social gender when it comes to determining a person’s true gender. Philosophers and psychologists engaged in research on “the” self often distinguish between a person’s various superficial selves (one’s “public” and “private” selves, for example) and one’s true or real self, the self that corresponds to who the person really is (as contrasted with who they take themselves to be, who they present themselves as, or who others take them to be). Prima facie, there is much to be said for identifying a person’s true self with their autobiographical self. Ordinarily, each of us will have greater and deeper access than anyone else not only to facts about what we have done and experienced but also to facts about our inner life—for example, our beliefs, motivations, desires, and so on. For this reason, our autobiographical identity—the identity that captures who we are to ourselves—seems, in ordinary cases, to be a much better candidate for capturing who we really are than any other representation of who we are; and so the self that is characterized by our autobiographical identity will generally be the best candidate for being our true self. If this is right, then there 13 See, e.g., Glasgow et al. 2019. Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 3 Published by Scholarship@Western, 2022 12 is a clear sense in which our autobiographical identity is privileged over other representations of who we are. That said, however, it would be hasty to conclude simply on the basis of this that, whenever our autobiographical identity conflicts with other representations of us, the representation included in the autobiographical identity is authoritative. Factual error, self-deception, lack of social awareness, persistent tendencies to overestimate or underestimate one’s own abilities or influence upon others, and so on are all common, widespread phenomena; and to the extent that we are susceptible to these, we will be unlikely to have a fully accurate conception of ourselves. Narcissism, excessive humility, having a drug or alcohol problem, and being an abuser or predator of some kind or other, to name just a few examples, are all traits that may well characterize who a person really is, even if the person doesn’t explicitly believe such things about themselves. Similarly, depending on what the correct accounts of identities like race or disability turn out to be, a person may have important social identities that they do not believe they have. Cases like these provide at least prima facie reason for thinking that the authority of a person’s autobiographical identity is defeasible. The question is what kinds of conflicts with other representations of us, or with real world facts, result in such defeat. There are three main points I want to make on this topic. The first is that it doesn’t follow from the fact that a person has false beliefs about who they are that their autobiographical identity inaccurately represents who they are; and so it will not automatically follow from the fact that (say) a narcissist believes she is not a narcissist that her autobiographical identity does not (authoritatively) represent her as a narcissist. This is most easily seen by attending to fiction. A novel written in the first- person can easily portray its protagonist as a narcissist, as subject to self-deception, as excessively humble, and so forth while at the same time portraying the protagonist as being unaware that they have such attributes. Likewise for nonfiction autobiographies. In the first case, it is true in the fiction, even though it is not explicit in the fiction, that the protagonist is a narcissist, self-deceived, or whatever; and we can cash this out in terms of whatever theory of truth in fiction ultimately seems best to us. Although autobiography is (typically) not fiction, it seems that truth in autobiography will work in much the same way as truth in fiction. So just as what is true in a fiction need not coincide either with what is in fact true or with what the author believes to be true, so likewise what is true in someone’s autobiography need not coincide with what is in fact true or with what the author believes to be true. Thus, there is no reason to doubt that one’s true self can be characterized by attributes that one does not recognize in oneself or might even explicitly deny of oneself. Consider again the alcoholic in denial: In the typical case, such a person possesses evidence that defeats the rationality of their explicit belief that they are not an alcoholic—this is why we say they are in denial rather than simply saying they are Rea – Gender as a Self-Conferred Identity Published by Scholarship@Western, 2022 13 unaware of their alcoholism. They know, on some level, how much they drink; they are aware, on some level, of their own struggles with self-control, followed by outright failures or implausible rationalizations; they are aware on some level, of negative comments and nonverbal social cues from friends and family, and of their own defensive reactions or rationalizing explanations that prop up their belief that they are somehow being misunderstood or persecuted by their friends and family; and so on. And, of course, in the typical case, all of this defeating evidence will be included in their autobiographical identity. Thus, we can say that the identity remains authoritative with respect to who they are, even if their explicit self-affirmations lack such authority. Alternatively, imagine someone who has a deviant concept of race and, on that basis, sincerely identifies as Black even though virtually everyone else would identify them as White. Suppose, for example, they think that it is sufficient for being Black that a person have at least one ancestor, however distant, who was born in Africa and had dark skin and, furthermore, they think that the first human beings emerged in Africa and had dark skin (so that literally everyone now counts as Black). Imagine also that their understanding of race is developed well enough to be at least coherent, and that, due to social isolation or other factors, they have never encountered evidence against it or evidence that it is deviant. This is, to be sure, a fanciful case; but it is meant just to be a hypothetical case where a person’s autobiographical identity (a) represents them as having a social identity they clearly lack, (b) includes no clearly defeating counterevidence (as in the case of the alcoholic), and (c) clearly lacks authority with respect to its representation of that social identity. The question is, why would this person’s autobiographical identity lack authority with respect to its representation of their race, and why wouldn’t that reason carry over to cases where one attributes to themselves a gender that most others in some particular context wouldn’t attribute to them? The answer, on the view of gender that I am developing, is that race, unlike gender, is not a representational identity. In other words, whatever exactly race amounts to, a person’s having a particular race involves more than someone’s (or some group’s) simply having a particular kind of interpretive representation of them. By contrast, a person’s gender, on the view I am developing, is fundamentally a matter of how they are interpreted—and, more fully, a matter of how their inner experience of themselves is interpreted. Social gender is, in effect, a collective external interpretation of what the hosts of that interpretation presume on the basis of outward signs (behavior, anatomy, etc.) to be the person’s inner experience;14 14 Is it plausible to say that those with transphobic views about gender are, in making their gender judgments, interpreting the inner self-experience of the trans people whose gender self-attributions they deny? Insofar as they are talking about Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 3 Published by Scholarship@Western, 2022 14 autobiographical gender is a person’s own internal interpretation of their inner experience. But, of course, as a general rule, by virtue of our own privileged access to our own inner experience, all of us are prima facie in a better position to interpret our own inner experience than anyone else will be. This is why it makes sense to insist that, when it comes to determining a person’s real gender, their autobiographical identity is authoritative. At any rate, one’s autobiographical identity is prima facie authoritative with respect to their gender. But is that authority indefeasible? No, I don’t think that quite follows. For example, it is possible for a person to be mistaken about their possession of attributes that they themselves take to be relevant to gender assignment. Suppose Sam has congenital adrenal hyperplasia, has male-appearing genitalia, but is chromosomally XX. Suppose further that Sam both believes that chromosomal profile is determinative of gender and, being unaware of their own chromosomal profile, believes on the basis of their visible external genitalia that they are XY and this means they are necessarily a man. It is also in principle possible for a person to be confused about what genders are. Borrowing an example from Heather Logue (forthcoming), suppose Sam believes that pizza is a gender and declares their gender to be pizza.15 It seems to me that in both of these scenarios we should say that Sam is genuinely mistaken or confused as to their gender. In the one case, they are mistaken (until puberty) about their possession of attributes they themselves take to be relevant to gender; in the other case, they are (probably) confused about what genders are.16 In gender, and not simply confusing gender with (presumed) “biological sex,” I think this is plausible. It is entirely common for such people to accuse trans persons of “pretending” or of “self-deception,” and both of these judgments reflect opinions about the inner self-experience of the people they are mistreating in this way. 15 Logue herself borrows the example from Reilly-Cooper (2016). 16 I say “probably” because I want to leave room for conceptual change and corresponding new coinages. There are, in English, a bewildering variety of names for genders—for example, “astral gender,” “earth gender,” and “lunarian.” (See “List of Uncommon Nonbinary Identities,” Nonbinary Wiki, https://nonbinary.miraheze .org/wiki/List_of_uncommon_nonbinary_identities, accessed on March 26, 2021.) I have no view about whether the people who have introduced these terms have successfully given them meaning as genuine conceptions of gender; but I am, at any rate, in no position to deny that they have. Accordingly, I am in no position to deny that one could introduce “pizza gender” into the mix. In doing so, however, one would have to do something very different from simply asserting (again, presumably in confusion or in jest) that pizza is their gender. One would have to take a concept that is genuinely a concept of gender and somehow modify it so that it continues to be a concept of gender while coming to encompass a meaningful conception of Rea – Gender as a Self-Conferred Identity Published by Scholarship@Western, 2022 15 such cases, then, Sam’s explicit affirmations about their own gender will lack first- person authority. Moreover, in the second case, where Sam has a radically deviant concept of gender, what I want to say is not just that Sam’s judgments about their own gender lack first-person authority, but that Sam has failed to confer gender upon themselves at all. The reason is that even if our gender attributes are not socially determined, gender concepts are socially constructed. That is, the content of our gender concepts and their constitution as concepts of gender, rather than of something else, is partly determined by the nature and use of gender concepts in the broader communities of which we are a part. I think that this understanding of gender concepts is fairly intuitive and widely accepted; but for those who disagree, I am content just to leave it as an unargued-for stipulation of the SCI account. Let us return now to the question of what self-conferral involves. Earlier I said that the self-conferral of gender is a matter of doing something that makes it the case that one has the gender that one has. In light of the foregoing, we can add the further claim that what one does to confer gender upon oneself is just to interpretively represent oneself as having that gender. We might flesh this out along roughly the same lines as Ásta’s (2018) account of the conferral of social properties in general. According to Ásta, a social property is conferred under certain conditions by attitudes or actions that consciously or unconsciously aim to track some underlying “base property.” For example, in the case of a strike in baseball, Ásta says that being a strike is conferred by the declared beliefs of the umpire in the context of a baseball game; and the umpire’s beliefs aim to track the ball’s trajectory, the base property. In the case of gender, Ásta thinks that gender properties are conferred in different contexts by the beliefs and attitudes of people who have what you might call the relevant kind of “social authority,” or (as Ásta puts it) “standing” in that context—people who enjoy social dominance within the context, for example, or people who have a certain kind of institutional authority (e.g., doctors, judges, etc.).17 These beliefs and attitudes aim to track certain base properties of the individuals upon whom gender is conferred.18 pizza. I don’t know how this could be done, but neither have I any proof that it cannot be done. 17 Note, then, that Ásta would identify gender with what I am calling “social gender” only if the following turns out to be true: the collective representation of a person P in a context C attributes gender G to P just in the case that the people with “standing” to determine gender in C collectively represent P as G. 18 Furthermore, on Ásta’s account, which base properties are tracked varies from context to context. For example, in some contexts, gender is conferred on the basis of one’s reproductive role; in other contexts, it is conferred on the basis of one’s social Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 3 Published by Scholarship@Western, 2022 16 Self-conferral works in much the same way, except that where it is typically the attitudes and actions of a particular social group that are relevant for the conferral of a social property, it is primarily one’s own attitudes and actions that are relevant for the conferral of a self-conferred attribute. So, on this view, as far as autobiographical gender is concerned, everyone is their own “gender umpire” as it were. By virtue of what gender is, and the epistemic positioning we have with respect to the experiences that underlie gender, each of us has standing to confer autobiographical gender upon ourselves, and our standing in this regard trumps anyone else’s such standing. Note, however, that I do not say that no one else has standing at all. I think it is plausible to say that, in the case of infants and others who fail to confer gender upon themselves simply by virtue of lacking concepts of gender or in some other way failing to represent themselves as gendered, they nonetheless have gender in some contexts, and their gender is simply their social gender in the relevant context. Saying this is not essential to the account I am offering, but it is consistent with it and affords what seems to me a plausible way of accommodating the common belief that babies and relevant others do indeed have gender. Thus, because one’s autobiographical identity is prima facie authoritative with respect to who a person really is, where autobiographical gender and social gender diverge, it is autobiographical gender that is prima facie authoritative with respect to the question of what gender someone really has. Again, that authority can be defeated in unusual cases, But in typical cases—that is, ones where a person has a basic grasp of what genders are and how gender terms are commonly used, and where a person is not in serious error about their own physical and mental attributes—the authority will be indefeasible for the simple reason that they have privileged access to their own inner experience, and so they are in the best position to provide the sort of interpretation of that experience that makes it the case that they have the gender that they have. The SCI Account and Other Theories of Gender I want to close with some brief comments about the relationship between the SCI account and other available theories about the metaphysics of gender. In particular, I want to comment on whether it is properly considered a social constructionist account of gender, on whether it is ultimately just a variation on Ásta’s conferralist account of gender, and on whether it is equivalent to the “self- role, of perceptions about one’s body, etc. But this complication is not especially relevant for the discussion of autobiographical gender. Rea – Gender as a Self-Conferred Identity Published by Scholarship@Western, 2022 17 identification” account that Katharine Jenkins (2018) attributes to Talia Mae Bettcher (2009, 2017).19 Given that gender is self-conferred on the SCI account, one might think there is reason to doubt that it is a social constructionist theory. After all, the theory implies that no individual’s gender is constructed by social ideologies, institutions, or conventions; thus, it would seem that gender attributes lack some of the most important hallmarks of being socially constructed. But even so, the account does respect the key social constructionist idea that what it is to be of a certain gender, and whether some specific person counts as belonging to that gender, will be at least partly determined by the gender concepts operative in some salient group of concept- users, and on the beliefs, attitudes, or other mental states of the members of that group.20 The difference is just that, for the SCI theorist, the salient “group” is always just the one-membered group consisting of the individual whose gender is in question. Moreover, the very idea of gender—as well as the social norms, stereotypes, and so on in accord with (or in resistance to) which each person conceptualizes, performs, and thereby constructs their gender—seems clearly to be not only constructed but constructed by society. So I think there is ample reason to classify the SCI account as a social constructionist view. One might also wonder whether the SCI account might be appropriately treated as a mere variant on Ásta’s conferralist theory of gender. Granted, the SCI account shares with Ásta’s account the basic idea that gender is conferred rather than being constituted by a particular base property or by the responses that the gendered individual induces in others. But to my mind, the differences are more important than the similarities. On Ásta’s view, gender is a social property, conferred upon us by other people; and because of this, it comes with objectionable consequences that the SCI 19 Bettcher does not formulate a view that she explicitly describes as the “self- identification account”; but Jenkins does formulate such a view, attributing it to Bettcher on the strength of a quotation from Bettcher 2017, informed by attention to Bettcher 2009. 20 One might wonder how similar one’s gender concepts have to be to those of others in order for one to count as having the same concepts of gender. Why think, for example, that if S confers (what she thinks of as) womanhood upon herself, her concept of womanhood has anything whatsoever to do with the gender concepts operative in any other group? I have no theory to offer in response to this question; but I take it that it is commonplace to suppose that (a) people with very different understandings of what (say) womanhood or gender amount to often nonetheless both count as using the concept of womanhood or gender, and (b) this fact has something to do with the way in which the concepts of womanhood and gender are communally used and shaped. Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 3 Published by Scholarship@Western, 2022 18 account manages to avoid. On Ásta’s view, one is not always one’s own “gender umpire.” Accordingly, sincere, linguistically and conceptually competent, and reflective self-identification as a woman or man in a context is neither necessary nor sufficient for being a woman or a man in that context. Moreover, and for the same reason, Ásta’s account allows for the possibility that one can be mistaken about whether one is a man, a woman, or a member of some other gender category; and therefore one will not necessarily be misgendered by people who disagree with one’s own gender self-assessment, nor will what we normally describe as misgendering someone necessarily count as a harm or any other sort of violation. Finally, one might wonder whether the SCI account is just an alternative expression of the so-called “self-identification” account of gender (identity). According to Jenkins (2018, 727), this account maintains that to have a female gender identity is to be someone who is disposed to reply to the question, “What is your gender?” with the statement, “I am a woman” (or words to that effect) and to be acting in good faith in doing so. Here, then, self-identification refers to the act of expressing or claiming certain identity, or, at least, being disposed to make such expressions or claims. This is not quite what Heather Logue (forthcoming) calls the “belief account” of gender, according to which having a particular gender is just a matter of believing that one has that gender; but it is very close to that. How does the present account differ? As an interpretive mental representation of oneself, an autobiographical identity is very much like a belief. If it is not identical to a dispositional belief, it is at least quite plausibly the ground of a cluster of dispositions to believe. So it might seem that autobiographical gender is determined, in effect, by a person’s own dispositional beliefs or dispositions to form beliefs about their gender; and, of course, this isn’t far off from the “disposition to claim” that figures in Jenkins’s formulation of the self- identification account. One important difference, however, is that the SCI account can accommodate the in-principle possibility of confusion or delusion about gender in a way that the self-identification account cannot. If being disposed to claim that one has a certain gender is sufficient for having that gender, then it seems that someone who believes they are a woman in part because of mistaken beliefs about their own physical attributes, or someone who believes that their gender is pizza, has to be regarded as correct in those beliefs. But this seems like exactly the wrong result. Furthermore, at least as it is developed by Heather Logue (forthcoming), the best way of making sense of the belief account of gender, and so likewise, presumably, for the self-identification account, is to embrace fictionalism about gender. If one counts as (say) a woman if, and only if, one believes oneself to be a Rea – Gender as a Self-Conferred Identity Published by Scholarship@Western, 2022 19 woman, it is hard to see what the content of the belief (or disposition to affirm) that one is a woman could possibly be. After all, if the belief that one is a woman amounts to a belief that one satisfies some condition C (C ≠ believing that one is a woman), then it would seem that satisfying C, rather than simply believing that one is a woman, provides the criterion for womanhood. But, Logue argues, fictionalism about gender lets us avoid the demand for an account of the contents of our gender beliefs because there is no general demand that every fictional attribute be analyzable. In other words (to put it in terms that she herself does not) fictional primitives, in contrast to “real” primitives, are wholly unproblematic. In contrast, however, the SCI account does not need to invoke fictionalism in order to skirt a demand for content for gender concepts. The SCI theorist can simply say that gender concepts get their content in the way that other social concepts do—presumably by way of communal patterns of usage, or (in the case of new coinages) imaginative and deliberate acts of conceptual engineering. Unlike the belief account, the SCI account is not committed to anything like the idea that there is nothing more or less to being a woman than representing oneself as a woman. It is fully consistent with the SCI account to say, for example, that S’s representing herself as a woman consists in her representing herself as satisfying condition C (C ≠ being a woman), where being someone who satisfies C is, according to her concept of womanhood, sufficient for being a woman. And so there is no need for the SCI account to take genders as primitive and to resort to fictionalism in order to justify doing so. As noted earlier, the SCI account is a realist account of gender, and this is an advantage because antirealist accounts risk minimizing the importance of our gender identities by treating them as (at best) attributes we merely pretend to have for one reason or another rather than as real attributes of ours. A final difference between the SCI account and the self-identification account: According to Jenkins, the self-identification account fails with respect to two of what she takes to be the six desiderata that must be satisfied by any adequate account of gender identity. First, the account “fares very poorly at showing that gender identity is important and deserves respect” because there is no obvious reason why we should “care about dispositions to utter certain sentences” (Jenkins 2018, 728). As she rightly notes, to the extent that we care about gender identities, what we most care about is “whatever it is that makes people want to utter those sentences, or whatever it is that they express when they do utter them” (728; italics in original). Second, the account struggles with . . . compatibility with possible need for transition- related healthcare: it is difficult to perceive any relationship at all between a linguistic disposition and the sort of felt need for one’s body to be different that would prompt the desire to access transition- related healthcare. (728) Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 3 Published by Scholarship@Western, 2022 20 I have no strong opinion about whether proponents of the self-identification account can adequately reply to these objections. The main point I want to make is just that, whereas they are at least superficially plausible objections to the self-identification account, they are clear nonstarters with respect to the SCI account. If gender is most fundamentally an interpretation of one’s own inner experience, it is crystal clear why gender would be important and why a person would want to declare themselves to be of one gender rather than another, and it is also clear why gender would be connected with a felt need to access transition-related health care. I note in passing, furthermore, that the SCI account also satisfies the four other desiderata that Jenkins (2018, 723–24) takes as adequacy constraints on accounts of gender: it is compatible with a norm of first-person authority, it is clear and noncircular, it applies equally well to binary and nonbinary identities, and (so far as I can tell) it combines well with broader critiques of current gender norms and social structures. Furthermore, although it is different from the “norm-relevancy account” that Jenkins favors, the SCI account has the resources to explain the intuitive appeal of the norm-relevancy account. On that view, one’s gender identity is determined by an internal “map”—an “internalised sense of the norms operating in social spaces that they regularly navigate, and the implications of those norms for the status of their own behaviour as norm-compliant or norm-violating” (729). In short, on Jenkins’s view, one has a gender identity of X in a context C just if one’s internal map is formed to guide someone who is socially positioned as a member of X through the various realities that are, in C, characteristic of (the social positioning of) Xs (730). Obviously, the SCI account makes no appeal to internal maps in the sense Jenkins has in mind; and likewise obviously, the SCI account does not invoke social positioning or context sensitivity in the way that her account does. But if the SCI account is correct, it is easy to see why members of a given gender would have internal maps of just the sort that Jenkins describes; and it is also easy to see why it might generally be true that the people who have (say) an internal woman-guiding map would be just the people whose gender is, by the lights of the SCI account, women.21 References Ásta. 2018. Categories We Live By: The Construction of Sex, Gender, Race, and Other Social Categories. New York: Oxford University Press. 21 For helpful conversations about the ideas in this paper, helpful comments on earlier drafts, or both, I am grateful to Sara Bernstein, Laura Callahan, Robin Dembroff, Heather Logue, two anonymous referees for this journal, and, subsequent to the paper’s acceptance, Kathryn Norlock. Rea – Gender as a Self-Conferred Identity Published by Scholarship@Western, 2022 21 Barnes, Elizabeth. 2020. “Gender and Gender Terms.” Noûs 54, no. 3 (September): 704–30. Bettcher, Talia Mae. 2009. “Trans Identities and First-Person Authority.” In You’ve Changed: Sex Reassignment and Personal Identity, edited by Laurie J. Shrage, 98–120. Oxford: Oxford University Press. ———. 2017. “Through the Looking Glass: Trans Theory Meets Feminist Philosophy.” In The Routledge Companion to Feminist Philosophy, edited by Ann Garry, Serene J. Khader, and Alison Stone, 393–404. New York: Routledge. Glasgow, Joshua, Sally Haslanger, Chike Jeffers, and Quayshawn Spencer. 2019. What Is Race? Four Philosophical Views. New York: Oxford University Press. Haslanger, Sally. 2000. “Gender and Race: (What) Are They? (What) Do We Want Them to Be?” Noûs 34, no. 1 (March): 31–55. Ismael, J. T. 2016. How Physics Makes Us Free. New York: Oxford University Press. Jenkins, Katharine. 2016. “Amelioration and Inclusion: Gender Identity and the Concept of Woman.” Ethics 126, no. 2 (January): 394–421. ———. 2018. “Toward an Account of Gender Identity.” Ergo 5 (27): 713–44. Lewis, David K. 1969. Convention: A Philosophical Study. Cambridge: Harvard University Press. Logue, Heather. Forthcoming. “Gender Fictionalism.” Ergo. Mallon, Ron. 2016. The Construction of Human Kinds. Oxford: Oxford University Press. McLean, Kate C. 2015. The Co-authored Self: Family Stories and the Construction of Personal Identity. New York: Oxford University Press. Nelson, Hilde Lindemann. 2001. Damaged Identities, Narrative Repair. Ithaca, NY: Cornell University Press. Nelson, Michael. 2019. “The De Re/De Dicto Distinction.” Supplement to “Propositional Attitude Reports.” In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta, Spring 2019 edition. https://plato.stanford.edu /entries/prop-attitude-reports/dere.html. Rea, Michael. Forthcoming. “The Metaphysics of the Narrative Self.” Journal of the American Philosophical Association. Published ahead of print, March 11, 2022. https://doi.org/10.1017/apa.2021.28. Rea, Michael, and Elizabeth VanKammen. 2021. “Troubling the Metaphysics of Social Properties.” Unpublished MS, last modified December 6, 2021. Reilly-Cooper, Rebecca. 2016. “Gender Is Not a Spectrum.” Aeon, June 28, 2016. https://aeon.co/essays/the-idea-that-gender-is-a-spectrum-is-a-new-gender- prison. Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 2, Article 3 Published by Scholarship@Western, 2022 22 MICHAEL REA is Rev. John A. O’Brien Professor of Philosophy and concurrent faculty in Gender Studies at the University of Notre Dame. His research focuses on topics in philosophy of religion, metaphysics, and feminist philosophy. 13959 Rea title page 13959 Rea final format