Pluralism, Opacity, and Values in the Social World: Commentary on Jenkins’s Ontology and Oppression Feminist Philosophy Quarterly Volume 11 | Issue 3 Article 2 Recommended Citation Khalidi, Muhammad Ali. 2025. “Pluralism, Opacity, and Values in the Social World: Commentary on Jenkins’s Ontology and Oppression.” Feminist Philosophy Quarterly 11 (3). Article 2. 2025 Pluralism, Opacity, and Values in the Social World: Commentary on Jenkins’s Ontology and Oppression Muhammad Ali Khalidi Graduate Center, City University of New York makhalidi@gc.cuny.edu Khalidi – Pluralism, Opacity, and Values in the Social World: Commentary on Ontology and Oppression Published by Scholarship@Western, 2025 1 Pluralism, Opacity, and Values in the Social World: Commentary on Jenkins’s Ontology and Oppression Muhammad Ali Khalidi Abstract Katharine Jenkins’s Ontology and Oppression helps bridge the chasm between discussions of social ontology in analytic metaphysics and treatments of injustice and oppression in social and political philosophy. In this commentary on Jenkins’s book, I address three main issues. First, I question a central innovation of the book—its espousal of a new form of pluralism about race and gender kinds, which posits that race and gender are not unitary kinds but can have three different types of manifestation in the social world. Second, I consider how Jenkins’s central notion of “ontic injustice” interacts with a prominent distinction between social kinds that have to be explicitly represented (“transparent”) and those that need not be represented (“opaque”). Third, I take up the issue of evaluative or normative considerations in social theorizing and their place in taxonomizing the social world. I ask whether, in rejecting what she calls the “ontology first” approach to theorizing about trans people, Jenkins stakes a position that opposes scientific categorization on the basis of nonepistemic values. Keywords: social ontology, oppression, ontic injustice, social kinds 1. Introduction Much work in social ontology from the perspective of analytic philosophy has proceeded largely oblivious of normative issues like injustice and oppression, focusing exclusively on the metaphysical underpinnings of social institutions, groups, norms, practices, and similar constructs. Meanwhile, the bulk of social and political philosophy pays scant attention to the ontological basis of social categories and kinds, concentrating primarily on normative issues in the social domain. Katharine Jenkins’s Ontology and Oppression helps bridge the chasm between these two areas of inquiry. Moreover, the book introduces into philosophical discourse new analytic tools, such as “ontic injustice,” that are destined to play a central role in these discussions for a long time to come. In this attempt to engage with Jenkins’s enriching and insightful book, I raise three issues that merit further discussion. In the second section, I tackle a central Feminist Philosophy Quarterly, 2025, Vol. 11, Iss. 3, Article 2 Published by Scholarship@Western, 2025 2 innovation of the book—namely, its espousal of a new form of pluralism about race and gender kinds, which posits that race and gender are not unitary kinds but can have three different types of manifestation in the social world. I raise questions as to the benefits of such a pluralist position over a “hybridist” position that thinks of these kinds as combining various criteria or aspects. Then, in the third section, I consider how Jenkins’s central notion of ontic injustice interacts with a prominent distinction between social kinds that have to be explicitly represented (“transparent”) and those that need not be represented (“opaque”). Does ontic injustice apply to unrepresented opaque social kinds, and if so, does it apply differently? In the fourth section, I take up the issue of evaluative or normative considerations in social theorizing and their place in taxonomizing the social world. In rejecting what she calls the “ontology-first” approach to theorizing about trans people, does Jenkins stake a position that opposes scientific categorization on the basis of nonepistemic values? 2. Race and Gender Pluralism One way to view the social world is through the lens of real or natural kinds. Many of the constructs that populate theorizing about the social world correspond to kinds, which include kinds of persons, institutions, groups, norms, practices, structures, and so on. Gender, race, and class, as well as government, inflation, and marriage, can all be considered natural kinds in the social domain. (In case that last phrase sounds oxymoronic, I should point out that I take “social kinds” to be nothing but natural kinds in the social domain, just as chemical kinds are natural kinds in the chemical domain. That is because I construe the “natural” here to mean real rather than pertaining to the natural world or the natural sciences.1) This perspective has led many philosophers to discuss kinds in the social world, such as gender and race, and think about their distinctive features, how they are grounded, whether they are biological, social, biosocial, and so on. It has been clear for some time that, particularly when it comes to gender and race, surely the most widely discussed social kinds, there can be various ways of characterizing them as social kinds. In this context, Jenkins makes a bold move, elaborating a sophisticated form of pluralism for these two social 1 For further justification of this stance on natural kinds, see Khalidi (2023). Compare also Charles Mills (2014, 92): “Any broader commitment to naturalism implies that the ‘social’ is really a particular zone of the natural—the artifact of a particular ‘natural’ species capable of reshaping itself through culture and technology, but able to do so, of course, precisely because of the ‘natural’ capacities of the specific kind of primate we are and continuous at every step of the way (contra Kant, or more generally, any sharp Christian dichotomization) with the ‘natural’ world and ‘natural’ causality. So a remapping of this contrast would be something like the ‘natural’ unmediated by human causality and the ‘natural’ mediated by human causality.” Khalidi – Pluralism, Opacity, and Values in the Social World: Commentary on Ontology and Oppression Published by Scholarship@Western, 2025 3 kinds. In effect, she proposes splitting gender and race into three different social kinds, which I would summarize briefly as follows: • Hegemonic gender/race: how one is positioned within dominant social structures • Interpersonal gender/race: how other people perceive and respond to one • Identity gender/race: one's experienced relationship to a social category This trifurcation of gender and race captures the fact that they are multifaceted kinds; indeed, it goes further than that by saying that these different facets of gender and race should be understood as distinct kinds in their own right. Focusing for the moment on gender, membership in a gender kind (e.g., woman, man, nonbinary) can be a matter of one’s social positioning, or it can be a matter of how other members of society respond to one, or a matter of how one experiences oneself. These would seem to be three crucial dimensions of gender in many societies and social configurations. Moreover, the same person might be gendered differently depending on whether it is a matter of, say, one’s own self-identification or identification by others. Or rather, if these are really different kinds, then one is not really being gendered differently, at least if that is taken to imply that the different genderings are incompatible. Take the case of an individual A who is gendered as a man within hegemonic social structures, is perceived by others as nonbinary, and identifies as a woman. There is no tension between membership in these different kinds precisely because they belong to orthogonal systems of kinds. To make this more apparent, consider some other individual, B, who is (again) gendered as a man within hegemonic social structures, but who is perceived by others as a man and identifies as a man. These three classifications are not redundant because they concern different taxonomic categories. Strictly speaking, since we are (or ought to be) talking about different kinds, we should have used different subscripts, saying that B is a manH (hegemonic), manP (interpersonal), and manI (identity). But when one puts it this way, an immediate objection comes to mind. Should these really be considered distinct kinds? These purported systems of kinds are not completely independent of one another but are surely linked in intricate ways. For example, the third dimension of gender, the hegemonic structural one, seems to depend in part on the second, the interpersonal dimension of gender. To take a simple example, if I am interviewed for a job and my interviewers perceive and respond to me as a man, and if I am subsequently hired as a man and treated accordingly by my bosses and coworkers in my new place of employment, it is likely that my wages, advancement, and privileges will reflect that fact, and I will consequently be Feminist Philosophy Quarterly, 2025, Vol. 11, Iss. 3, Article 2 Published by Scholarship@Western, 2025 4 positioned as a man in the socioeconomic hierarchy. Thus, interpersonal gender is partly determinative of hegemonic gender. But perhaps this is not a problem for the account, since something similar holds of distinct yet closely related kinds in other domains. In this regard, the relations between the three kinds of gender are comparable to those between different biological species concepts, which also plausibly correspond to different species kinds (see Kitcher 1984; Dupré 1999).2 The different gender and race kinds can be compared to, say, humanL, humanB, humanE, and so on, where membership in humanL has to do with a certain lineage and descent, humanB is a matter of belonging to interbreeding groups, and humanE pertains to occupying a certain ecological niche. Here, too, membership in humanL is intimately related to membership in humanB: it is because of a certain history of descent that an organism is able to interbreed with other members of its species. These kinds are not entirely independent but causally linked in various ways. Still, many biologists and philosophers of biology find it useful to distinguish them for various theoretical purposes. And even though their extensions coincide for many species, there are occasional exceptions, such as different populations of a “ring species” that are counted as members of the same species on grounds of descent but whose members may not be able to produce fertile offspring and so are considered members of different species on that basis. Indeed, there are at least two dozen (and counting) species concepts in biology, according to which we group populations of organisms based on such factors as lineage, reproductive isolation, and occupation of an ecological niche, among other factors (see, e.g., Hey 2001). Although they are regularly referred to as “species concepts,” they can also be thought of as picking out different species kinds. As in the species debate, the different ways of classifying individuals into race and gender kinds often settle on the same extensions, but there are notable exceptions that are caught by these distinctions and that might not be easily discernible without them. These compatible and coexisting classifications also explicitly recognize that there are importantly different causal processes in the social world that can coexist at different scales and be captured by different levels of explanation, despite important links between them. The analogy with species concepts may serve to allay a concern about pluralism by showing that there are other cases in which we distinguish kinds that are 2 There is a debate among philosophers of biology as to whether pluralism about species implies antirealism or eliminativism. Philip Kitcher (1984) and John Dupré (1999) are pluralists about species but not antirealists. Marc Ereshefsky (1998) argues that pluralism leads to antirealism, whereas Ingo Brigandt (2003) responds by defending realism. I will assume that pluralism about social kinds does not lead to antirealism or eliminativism, but that the social world really does contain these different gender and race kinds. Khalidi – Pluralism, Opacity, and Values in the Social World: Commentary on Ontology and Oppression Published by Scholarship@Western, 2025 5 nonetheless intimately connected. But a related worry might be raised regarding pluralism about race and gender—namely, that it would lead to an unnecessary proliferation of kinds in the social world and might actually encumber social scientific theorizing. By way of comparison, even though some biologists find it useful to distinguish species in the ways mentioned, other biologists adopt something like a hybrid category, which comprises two or more sets of criteria. Should one do something similar with gender and race when it comes to both science and the folk? When it comes to scientific theorizing, it should hardly be surprising that in the intricate causal networks of the social world, there are various different types of patterns and configurations that correspond to what we label “gender,” and that there are therefore advantages in keeping them separate. Indeed, in many contexts, it is crucial to distinguish the different kinds that are implicated. Take, for instance, medical research that looks at health conditions in a specific population and collects relevant demographic information, including gender. In many such cases, it is very useful to know whether information about gender is based on self-identification, identification by medical professionals, or some other basis, and it may well make an important difference to the findings.3 There are certain medical conditions or health outcomes that vary primarily with one or the other factor. For example, higher rates of workplace injury among men are likely due to hegemonic gender, which assigns to men certain types of occupation such as construction work, while higher rates of eating disorders among women may be more closely related to interpersonal gender, because of the emphasis that many members of society place on the “ideal” female body shape and the pressures that this places on women. Indeed, we might expect certain scientific disciplines and subdisciplines to be more attuned to some of these gender kinds than others. At least to a first approximation, economists and sociologists are interested mainly in hegemonic gender, while social psychologists investigate interpersonal gender, and psychologists are more concerned with identity gender. The same may go for race.4 For our everyday practices and purposes, also, it might be useful in some contexts to disambiguate. A person might explain that even though they identify as a 3 The correspondence between these means of identifying gender and the gender kinds being discussed is far from perfect. For example, individuals may not self-identify in a clinical context in the way that they would in a more familiar context, and an identification of gender by medical professionals may not always capture interpersonal gender due to differences in the way gender is perceived by different members of society. 4 Even within the same broad discipline, it is often important to disambiguate which type of gender or race is implicated, as in the examples from the medical sciences just cited. Feminist Philosophy Quarterly, 2025, Vol. 11, Iss. 3, Article 2 Published by Scholarship@Western, 2025 6 man, they are often perceived and treated as nonbinary, they are situated as a woman in the social hierarchy, and there is no outright contradiction between these classifications and kind memberships.5 But does this way of proceeding encourage a quietist position toward oppressive practices of misgendering, since it acquiesces in perceptions and classifications that may be at odds with one’s gender identity? Jenkins might say that in order to understand oppression and to resist misgendering, we need to acknowledge the way that people are categorized and the kinds that they effectively belong to. There is no point denying that someone who identifies as a man can yet be discriminated against as a woman and have an employment history that reflects that fact. That is not a matter of acquiescence in discrimination and prejudice but rather a matter of adopting a clear-eyed understanding of social reality (see section 4 for more on this point). Notwithstanding these benefits of splitting, the virtues of lumping should also not be ignored as an alternative in both science and everyday practice. Rather than embrace pluralism about race and gender kinds, we might adopt a position whereby these different aspects or dimensions of race and gender are combined or folded into the same category. Those categories may identify a hybrid kind in the world that preserves elements of each of these dimensions. (Note that these ways of talking are meant to reflect a broadly realist stance, whereby kinds are aspects of the world itself and categories are discursive devices that are designed to ascertain and distinguish them.) This type of lumping may be useful, and indeed more efficient, in some contexts, even though it may obscure different dimensions of race and gender in others, as intimated by the above examples from medical science. True enough, lumping makes no dedicated room for those persons who are gendered differently in terms of self-identification, interpersonal relations, and hegemonic social structure, but there may be good grounds for amalgamating gender and race kinds for some purposes. The advantages of hybrid gender and race kinds are worth considering, especially give one additional source of pluralism about these kinds, to be considered next. A different type of pluralism about race and gender has to do with diverse spatiotemporal contexts. When it comes to race, there is a debate as to whether it “travels,” which is to say whether race is localized to specific contexts, or whether it is 5 Compare the following picture evoked by Ásta (2018, 73): “Consider this scenario: you work as a coder in San Francisco. You go into your office where you are one of the guys. After work, you tag along with some friends at work to a bar. It is a very heteronormative space, and you are neither a guy nor a gal. You are an other. You walk up the street to another bar where you are a butch and expected to buy drinks for the femmes. Then you head home to your grandmother’s eightieth birthday party, where you help out in the kitchen with the other women while the men smoke cigars.” Khalidi – Pluralism, Opacity, and Values in the Social World: Commentary on Ontology and Oppression Published by Scholarship@Western, 2025 7 a universal or global phenomenon (see, e.g., Root 2000, S631; Mallon 2004, 658–60; Ludwig 2019). Jenkins refers to this debate briefly and endorses the idea that since race does not travel, there are actually different race kinds in different societies at different times. As she puts it, someone can be “a member of one race kind in one location and another race kind in another location” (Jenkins 2023, 146). This conclusion would seem to follow directly from Jenkins’s views on the nature of social kinds like race and gender. To see this, note first that Jenkins considers social kinds to be constituted in some sense by the constraints and enablements that pertain to them. As she puts it, “What it is to be a member of a certain social kind is, at least in part, to be subject to certain social constraints and enablements” (17). In addition, she observes that these constraints and enablements can vary along three dimensions: scope, breadth, and granularity. Scope has to do with the specific time and place in which the constraints and enablements associated with a social kind operate—for example, Europe during the Middle Ages, or a specific meeting in a workplace (102–3). Breadth concerns the varieties of constraints and enablements that pertain to a social kind. And granularity is the level of generality at which constraints and enablements are specified. It follows that if different settings for these parameters are established, this will result in distinct kinds.6 Consider scope, which is the dimension most pertinent to the question of “traveling” race and gender kinds. If we restrict ourselves to hegemonic race, it will certainly function differently and have different constraints and enablements associated with it in, say, the Jim Crow United States and contemporary Brazil. Though hegemonic race figures in both spatiotemporal contexts, the restrictions that it imposes on people and the entitlements that it affords them are quite different. Hence, it will have a distinct identity as a social kind in these different contexts.7 These divisions (scope, breadth, and granularity) would seem to cut across the three already mentioned—identity, interpersonal, and hegemonic—since all three race kinds can have either wide or narrow scope, and so on. In the face of all this pluralism, one might ask what unifies race and gender (see also Ludwig 2019; Mason 2020). Jenkins (2023, 183–88) tackles the “unification question” head-on when it 6 In earlier work, Jenkins (2019) is more explicit on this point, since she argues there for “ultra radical racial pluralism,” which holds that there may be different kinds of entities to which race terms refer even in the context of a single society like the United States. 7 Jenkins thinks that scope differences are less prominent or apparent for hegemonic race kinds than they are for interpersonal race kinds. She writes that hegemonic race kinds “allow for extensive traveling,” whereas interpersonal race kinds “are well able to capture the sense in which race does not travel” (2023, 146), but it seems to me that hegemonic race kinds also vary across spatiotemporal context. Feminist Philosophy Quarterly, 2025, Vol. 11, Iss. 3, Article 2 Published by Scholarship@Western, 2025 8 comes to pluralism about both gender and race kinds and responds convincingly to it. Gender kinds are ones that can be explained by the binary social structure that attributes social significance to perceived biological sex (see Barnes 2020) and kinds that are appropriately historically related to those social structures (see Bach 2012). Similarly, race kinds can be explained by a social structure that attributes social significance to “perceived biological ancestral links to different geographic regions” (Jenkins 2023, 186). Therefore, while the gender and race kinds that Jenkins identifies (identity, interpersonal, and hegemonic) can all be regarded as gender and race kinds, respectively, they can also be considered distinct kinds in the social world.8 The same would also seem to apply to the differently scoped gender and race kinds, with different kinds in distinct spatiotemporal contexts; likewise for various combinations and intersections of these kinds—for example, interpersonal race during the Jim Crow era in the United States or hegemonic gender in medieval Europe. Thus, one can make sense of the unity of gender and race despite these multiple sources of pluralism. Still, this proliferation of kinds may exacerbate the previous worry about multiplying kinds beyond necessity. For some purposes, there may be benefits to theorizing about race and gender using hybrid categories that amalgamate them across diverse contexts, a possibility that deserves further consideration. 3. Opacity and Transparency A distinction can be drawn between social kinds that are transparent, in the sense that they must be represented or conceptualized in order to exist, and those that are opaque, in the sense that they need not be represented or conceptualized to exist (see, e.g., Thomasson 2003; see also Ruben 1989). One can illustrate this distinction by comparing the social kinds permanent resident and refugee: while someone cannot be a legal permanent resident in a certain jurisdiction without being represented as such, one can be a refugee without anyone thinking of them as such or, indeed, anyone having the concept REFUGEE. Indeed, as this example suggests, when it comes to the social kind permanent resident, not only does the type need to be represented, but each token of that type needs to have the status of permanent resident explicitly conferred upon them. This suggests that one can introduce an intermediate category: those social kinds whose types must be represented to exist but all of whose tokens need not be represented as such to be members of the relevant kinds. Consider the legal category senior citizen, which in some jurisdictions 8 These gender kinds should not be thought of as subordinates of a superordinate gender kind, since a member of a subordinate kind is automatically a member of its superordinate kinds, but someone might be a manI but womanH or womanP, so not a man or woman simpliciter, even though the former kinds are somehow related to an overarching gender kind. Khalidi – Pluralism, Opacity, and Values in the Social World: Commentary on Ontology and Oppression Published by Scholarship@Western, 2025 9 includes all and only those citizens who are 65 years old and older. The legal category needs to be represented for the kind to exist, but an individual might belong to the kind senior citizen without being recognized to be one, perhaps because they and everyone else have forgotten their age. We can therefore posit three kinds of social kinds: (1) the kind need not be represented to exist (e.g., refugee), (2) the kind needs to be represented but not each member of the kind (e.g., senior citizen), and (3) the kind and each member both need to be represented (e.g., permanent resident) (see Khalidi 2015). It is also worth keeping in mind that exemplars of the first kind of social kind (opaque) need not be represented, but they can be. Thus, refugees existed before the concept REFUGEE, and although they need not be explicitly represented and recognized as refugees, they often are. And once they are, they might occupy a different social role, particularly if the category refugee is overtly enshrined in laws or international treaties. With this trichotomy in place, we can ask whether ontic injustice applies to all three kinds of social kinds, and if so, whether it applies differently. To pursue these questions, first, recall Jenkins’s (2023, 71) definition: Ontic injustice: An individual suffers ontic injustice if and only if they are socially constructed as a member of a certain social kind where that construction consists, at least in part, of their falling under a set of social constraints and enablements that is wrongful to them. At first glance, given the use of the terms “constructed” and “construction,” which imply a certain type of deliberateness, it might seem that the concept of ontic injustice applies only to transparent social kinds, which must be conceived of and recognized as such. More correctly, it might be thought to pertain to those social kinds that are actually represented, since, as mentioned earlier, opaque social kinds can be represented; they just do not have to be in order to exist or be manifested. It is clear, however, that members of opaque and unrepresented social kinds can also be unjustly treated by virtue of being members of those kinds. To use an example (of mine), consider members of the proletariat in the Middle Ages, as Europe emerged from the feudal order, and a new class of propertyless laborers came into being alongside a property-owning bourgeoisie. Members of that class suffered injustice before anyone thought of or represented them as such. Injustice and exploitation were part of the social structure, even though that structure was largely opaque, and proletarians were, at first, not explicitly represented or conceptualized. If we grant, at least for the moment, that ontic injustice applies to opaque social kinds (including those that are not represented) as well as to transparent ones, there do seem to be some differences in the ways that injustice and oppression apply to these kinds. Once an opaque social kind comes to be explicitly represented and Feminist Philosophy Quarterly, 2025, Vol. 11, Iss. 3, Article 2 Published by Scholarship@Western, 2025 10 conceived in a certain way, and once this informs action and policy, that paves the way for new forms of injustice and oppression. Note that in such cases, the opaque kind has been represented but has not been rendered transparent, since opacity and transparency are about the conditions that must be in place for something to exist. To illustrate, social class is opaque, in that it does not have to be represented to exist, as with the distinction between the proletariat and the bourgeoisie; but it can be represented, and certain changes might take place once it is. Consider legislation in a certain society that denies voting rights to propertyless persons. This constitutes a more entrenched and systematic form of oppression than might be associated with propertyless people before they are explicitly represented as such, as proletarians. To be sure, members of kinds that are not explicitly represented can be subject to all manner of prejudice and bias. But the changes that ensue with explicit representation seem to be of a different character, so it is worth considering how representation alters the nature of oppression. Take the case of Arab immigrants to the United States in the nineteenth and twentieth centuries, who were often classified as “white” or “Caucasian” during certain historical periods (but also as “colored,” “nonwhite,” and “Semitic” at other points). Despite the fact that they were not always conceived as a distinct ethnic group, there is clear evidence that Arab Americans were the subjects of prejudice, discrimination, and racism on the basis of ethnicity even before they began to widely identify or be identified as “Arab” or “Arab American” (see, e.g., Naber 2000; Kayyali 2013). But researchers argue that the introduction of the label “Arab American,” beginning around the late 1960s, led to increased prejudice against this community and new forms of discrimination (Naber 2000). Significantly, in March 2024, the US Census Bureau formally announced that it would introduce the category “Middle Eastern and North African” (MENA) starting with the 2030 census.9 Though many in the MENA community had long advocated for the prospective change, others cautioned that it would intensify and expand “antiterror policing” by providing demographic data to government agencies that associate this community with terrorism (Beydoun 2015, 6).10 Legal scholars and others have argued that establishing 9 The terms “Arab” and “Middle Eastern and North African” (MENA) are not coextensive, but there is considerable overlap, as there is with the term “Southwest Asian and North African” (SWANA), which is increasingly preferred by some people with those geographic origins. 10 Khaled Beydoun (2015, 6) also writes: “In short, the MENA American classification may be a step forward with regard to extinguishing the per se erasure of Arab Americans, but simultaneously, it is a step backward with respect to community surveillance, profiling, and policing.” Moreover, he cites a New York Times report from 2004, three years following the attacks of September 11, 2001, which showed that the Census Bureau provided population statistics on Arab Americans to the Department Khalidi – Pluralism, Opacity, and Values in the Social World: Commentary on Ontology and Oppression Published by Scholarship@Western, 2025 11 an official category of “Arab” or “Middle Eastern and North African” in the US census “may erode Arab American civil liberties by augmenting the precision of government surveillance and monitoring programs” (Beydoun 2015, 3; see also Beydoun 2016). Thus, it is likely that explicit classification and codification in administrative and legal codes can lead to new forms of injustice and oppression. Changes in the nature of injustice and oppression that occur as a result of representing a kind do not always concern institutional or legal kinds, ones that are explicitly codified in the laws or administrative codes, but also social kinds that are communally based, where the constraints and enablements are not as overt, rigid, or precise.11 This can be seen, for example, in the transition from informal racism to “scientific racism,” an elaborate pseudoscientific theory of the nature of races and their characteristics in early modern Europe. As Robert Bernasconi (2020, 83) argues, the introduction of an explicit classificatory concept with a pseudoscientific definition altered and legitimized racist practices: It was possible for the Spanish or the English to exploit Jews, Native Americans, and Africans, as Jews, Native Americans, and Africans, without having the concept of race, let alone being able to appeal to a rigorous system of racial classification. We have no difficulty identifying these as cases of racism, but they were not sustained by a scientific concept of race. However, the introduction of that concept lent an air of apparent legitimacy to these practices. Similarly, in discussing the origins of white supremacy in early modern Western thought, Cornel West (2002, 99) stresses the importance of the introduction of classificatory categories that originated in natural history: “The initial basis for the idea of white supremacy is to be found in the classificatory categories and the descriptive, representational, order-imposing aims of natural history.” Even before racial laws were introduced in various societies, explicit representation and conceptualization had a significant impact, even though race may have been conceptualized differently by different thinkers. It seems clear, therefore, that the character of injustice and oppression may well change substantially if a social kind is explicitly represented (as opposed to being unrepresented), and this is something that merits further discussion. But a worry arises here that can be put starkly as follows. If ontic injustice and oppression apply to of Homeland Security, “including detailed information on how many people of Arab backgrounds live in certain ZIP codes,” breaking down the Arab American population in more than five hundred zip codes by seven nationalities (Beydoun 2015, 7–8). 11 See Ásta (2018) for the distinction between institutional and communal social kinds. Feminist Philosophy Quarterly, 2025, Vol. 11, Iss. 3, Article 2 Published by Scholarship@Western, 2025 12 cases in which a social kind is not explicitly represented as such, how is this to be distinguished from injustice and oppression more generally? In other words, what makes ontic injustice, specifically in cases without explicit representation, different from plain old injustice, in cases in which one is not represented as a member of a certain social kind? So far, I have assumed that the answer is that these cases of injustice are ones that pertain to people not as individuals but insofar as they are members of social kinds, whether those kinds are explicitly represented or not. But it is not clear that this is sufficient for characterizing the distinctive wrong that Jenkins associates with ontic injustice, and whether the kinds of injustices described here, involving both opaque and transparent social kinds, all count as ontic injustice according to Jenkins.12 4. Values and Social Kinds There is a long-standing debate in the philosophy of social science about the role of value judgments in theorizing about the social world. For some theorists, the adoption of specific social values, no matter how admirable, should and can be avoided in conducting dispassionate social inquiry (DuBois 1898; Weber [1946] 2004), while for others, the very notion of value neutrality when it comes to scientific theorizing, especially when it comes to the social world, is both unworkable and undesirable (Longino 2008; Ludwig 2016). These latter theorists hold that nonepistemic values are inevitably involved in theorizing about the social world and can indeed play a salutary role in the social sciences. Although this debate is not explicitly referenced in Jenkins’s work, I think it is broached obliquely in her discussion of trans-exclusionary arguments and policies in ways that help illuminate this important question. In the context of a defense of trans-inclusion, Jenkins engages in a concerted discussion of the relationship between our explanatory projects and our emancipatory interests when it comes to social kinds, which provides an important case study regarding the place of nonepistemic values in social science. The social phenomenon that she engages with concerns contemporary transphobia in the United Kingdom, specifically as it manifested around plans to reform the Gender Recognition Act (2004, c. 7) in 2017. There is no need to go into the details of the debate here; suffice it to say that Jenkins identifies an assumption made by supporters of the trans-exclusionary position, “the ontology-first approach,” that has played a key 12 An anonymous referee suggests that the specific wrong of ontic injustice is moral injury, and that may provide the key to understanding the difference between ontic injustice and other forms of injustice associated with membership in a social kind. The discussion in this section has focused on the harms pertaining to membership in a social kind rather than the wrongs associated with such membership. Khalidi – Pluralism, Opacity, and Values in the Social World: Commentary on Ontology and Oppression Published by Scholarship@Western, 2025 13 role in public discussions. That approach posits that “settling questions about the current ontology of gender kinds will automatically determine what shape our gendered social practices ought to take” (Jenkins 2023, 201). When it comes to trans women, some parties to the debate argue that we should first establish the correct account of the ontology of the kind woman (and other gender kinds) and then let this determine social practice. She associates this approach with the slogan “Choose Reality,” adopted by some proponents of trans-exclusion, which is interpreted to mean that trans women who have not undertaken a medical transition should not be treated as women socially and politically. Since the claim among many proponents of trans- exclusion is that trans women are not, ontologically speaking, members of the social kind woman, they argue that this fact should shape policy and practice. According to this approach, the alleged ontology of the social world ought to be worked out before deciding what legislation to pass or which policies to pursue, and when it is, it vindicates trans-exclusion. How does Jenkins resist this attempt to let supposed social facts determine policy and practice? Before arguing against trans-exclusion and its purported commitment to the “ontology-first approach,” Jenkins engages in some metaphysical ground-clearing concerning social kinds. She points out that “one common way of conceiving of social kinds is as groupings of things in the world that can do explanatory work” and that this approach identifies gender with “the kind that does the explanatory work that we associate with the idea of ‘gender’” (2023, 213–14). As should be clear by now, I am in general agreement with this approach to social kinds, but I have a quibble with the wording in this quotation as well as a friendly suggestion. As Jenkins notes, kinds are groupings and divisions in the world, so rather than think of the kinds themselves as doing the explanatory work, I would rephrase this by saying that the categories that correspond to the kinds do the explanatory work. Meanwhile, categories belong to our discourse and practices, and they can be used by us to generalize, project, explain, predict, and perform other epistemic functions. My more substantive adjustment would be to suggest that the reason that some social categories can play these epistemic roles is precisely that they denote kinds, which can be thought of as part of the causal structure of the social world or, as I have put it elsewhere, “nodes in causal networks” (Khalidi 2018). With these epistemological and metaphysical preliminaries in place, Jenkins observes that, in line with the pluralism defended in the rest of the work (and previously explicated in section 2), no single gender category is going to do all the explanatory work. Rather, we will need to posit identity, interpersonal, and hegemonic gender kinds to explain various social phenomena, and these may classify individuals differently. Interestingly, Jenkins points out that the slogan “Choose Reality” has also been coopted by advocates of trans rights who have argued that trans people were indeed “choosing reality” by living in accordance with their gender identity. Feminist Philosophy Quarterly, 2025, Vol. 11, Iss. 3, Article 2 Published by Scholarship@Western, 2025 14 Additionally, Jenkins (2023, 217) notes that if we recognize the phenomena of ontological injustice and oppression, we should acknowledge that some social kinds are oppressive, and hence we should not automatically defer to such kinds in arranging our social practices. This means that just because trans women may be classified as men in existing society, that does not mean we must always accede to existing classifications. Social practices should sometimes track existing kinds and sometimes depart from them, and indeed resist them, pace the ontology-first approach. While our social practices cannot ignore the reality of social oppression, they should be in the service of our emancipatory efforts, which aim at changing social reality rather than always accepting it as it is. I would venture that another way to approach this debate involves distinguishing between the context of explanation and the context of emancipation in social theorizing. One way to diagnose the problem with the ontology-first approach is to say that it confuses an investigative question with a normative one. I think Jenkins would agree since she points out the following: “I want to suggest that discussions of gender recognition could helpfully be framed as primarily practical and normative rather than primarily metaphysical” (2023, 226). If that reading of Jenkins is right, it seems to carry with it an implicit account of the role of values in social science, according to which our moral and political values ought not to determine how to classify existing social phenomena but can and ought to be brought to bear on decisions on how to alter social reality. But this type of approach is sometimes challenged on the grounds that we cannot neatly separate the explanatory and emancipatory projects. So how should we respond to these challenges? Elizabeth Anderson (2012, 402) helpfully distinguishes two roles for moral and political values in social inquiry: Value judgments guide inquiry toward the concepts, tools, and procedures it needs to answer our value-laden questions. But facts— evidence—tell us which answers are more likely to be true. These two roles must be kept distinct, so that inquiry does not end up being rigged simply to reinforce our evaluative preconceptions. So long as they are distinct, the active direction of scientific inquiry by value judgments is not only legitimate, but indispensable. As I understand it, Anderson is drawing attention to two different roles that value judgments can play in social science. They can guide the choice of inquiry and can help define our research priorities, or they can direct us in deciding which theories to adopt; but even though they ought to do the former, they ought not to be implicated Khalidi – Pluralism, Opacity, and Values in the Social World: Commentary on Ontology and Oppression Published by Scholarship@Western, 2025 15 in the latter.13 For example, because of concern for the well-being of trans people, our moral values can lead us to conduct an inquiry into the effects of misgendering on trans people in certain social contexts, but they ought not to lead us to adopt a particular theory of how trans people are perceived. If we think that it is immoral for trans people to be misgendered, we ought to acknowledge that misgendering takes place. Otherwise, we would be misrepresenting the social reality that we are attempting to understand. That may seem obvious, but some might object that if, say, trans men are treated as women by some others (genderP) and within the hegemonic social structure (genderH), we ought not to acquiesce in this in our social theorizing and ought rather to redraw the boundaries of social kinds so as to ameliorate injustice and oppression. On this alternative view, the very choice of the categories of interpersonal and hegemonic gender merely serves to entrench the social structure that we regard as morally unjust. If our aim is to rectify the injustice, however, we need to understand it, and the only way to do so is by explaining how it functions in the social world. Notice that our moral values can also lead us to look for other categories in the vicinity and may direct us to posit a gender-identity category (genderI) to denote a distinct kind. But unless that kind is also part of the causal structure of the social world, it would be misleading to adopt it in our theorizing. Once again, moral values ought not to influence our choice of categories, though they should provoke us to use our findings to change the social world in an emancipatory vein, so as to oppose misgendering. This seems to be in keeping with Anderson’s intention in the passage quoted above, though there is some ambiguity in saying that value judgments can “guide inquiry toward the concepts.” As I read it, this does not mean that value judgments ought to play a role in directing us how to define classificatory concepts or where to draw category boundaries. Rather, they can guide us to look for certain configurations in the social world that we might miss, such as genderI. This is not a matter of helping us to decide which kinds exist in the social world but rather a matter of helping us to decide which kinds to focus on, of the many 13 In discussions of the role of social values in science, normative and descriptive claims are not always clearly distinguished. Thus, even though the first two sentences in the quotation from Anderson are phrased descriptively, I think they are intended to have normative force, with appropriate “oughts” understood. Zina Ward (2021, 58) has a very useful breakdown of the strengths of various claims made in this connection, framed in terms of scientific theory choice: (a) Social values ought to be considered in theory choice; (b) Social values can be considered in theory choice; (c) Social values must (in a nonmoral sense) play a role in theory choice; (d) Social values do play a role in theory choice. Feminist Philosophy Quarterly, 2025, Vol. 11, Iss. 3, Article 2 Published by Scholarship@Western, 2025 16 kinds that exist.14 To sum up, the role of moral values ought to consist, first, in guiding us to inquire into causal structures in the social world that we might otherwise overlook, and second, in finding ways to alter that social structure. Value judgments can direct inquiry by setting research priorities and deciding what to do with research results, not by helping to decide which categories are needed to understand social phenomena.15 But it is not evident whether Jenkins sees things in this way or whether this reading faithfully captures her view of the role of moral and political values in theorizing about the social world. 5. Conclusion Jenkins’s Ontology and Oppression is an indispensable contribution to “analytic emancipatory theory,” which gives injustice and oppression a central role when it comes to understanding the metaphysics of the social world. In this paper, I have tried to engage with the work on three different issues: pluralism, opacity, and values. On the first issue, I have defended Jenkins’s variety of ontological pluralism against some objections and have proposed expanding it further using her notions of scope, breadth, and granularity, thus yielding a range of different race and gender kinds in the social world, which manifest at different levels and in different contexts. At the same time, I have raised a question as to whether a hybrid approach is also sometimes warranted: might there be room for amalgamated concepts of gender or race for some theoretical purposes? When it comes to the second issue, I have tried to draw out the implications of Jenkins’s concept of ontic injustice for the distinction between transparent and opaque social phenomena (those that are and are not concept-dependent, respectively). I argued that injustice and oppression interact differently with transparent and opaque social phenomena, and that once opaque phenomena are explicitly represented, new forms of injustice and oppression may come into place that depend on representations of social phenomena, whether formally or informally, institutionally or communally. But this leads to a question about the ways of distinguishing ontic injustice from other forms of injustice that pertain to people insofar as they belong to specific social kinds: how is ontic injustice to be 14 See Khalidi (2024) for further justification of the claim that nonepistemic or social values should not determine our choice of taxonomic categories in science, though they can determine which taxonomic categories to deploy in a given context. Since a pluralist view of scientific taxonomy holds that there may be a plurality of valid taxonomic schemes in a given domain, nonepistemic values can help decide which scheme to use for a given purpose, rather than which schemes are valid. 15 In addition, of course, our social values along with the values of others, as manifested in our attitudes and actions, may be efficacious in fashioning the nature of social reality in the first place, irrespective of any inquiries we might carry out. Khalidi – Pluralism, Opacity, and Values in the Social World: Commentary on Ontology and Oppression Published by Scholarship@Western, 2025 17 distinguished from other forms of injustice, particularly when members of social kinds are not explicitly represented as such? Finally, with regard to the third issue, I have argued that Jenkins’s treatment of trans rights and critique of the ontology-first approach would seem to implicitly align her with a particular conception of the role of nonepistemic values in social theorizing, one that distinguishes the context of explanation from the context of emancipation and accords a role to such values in the former rather than the latter context. Does this mean that Jenkins adopts a view according to which taxonomic categories in the social choices ought not to be demarcated based on moral and political values? On all these points, I hope to have made a little further progress on the trail that Jenkins has blazed and have tried to raise questions that will promote further discussion of some foundational issues about the social domain. Acknowledgments For valuable feedback, I am grateful to the participants and audience at an author-meets-critics session at the American Philosophical Association Pacific Division meeting in Portland, March 2024, especially Katharine Jenkins. I would also like to thank Itsue Nakaya-Pérez and Maximiliana Rifkin for insightful discussion of Jenkins’s work and two anonymous referees for this journal for very helpful comments on an earlier draft. References Anderson, Elizabeth. 2012. “Uses of Value Judgments in Science: A General Argument, with Lessons from a Case Study of Feminist Research on Divorce.” In Out from the Shadows: Analytical Feminist Contributions to Traditional Philosophy, edited by Sharon L. Crasnow and Anita M. Superson, 377–404. Oxford: Oxford University Press. Ásta. 2018. Categories We Live By: The Construction of Sex, Gender, Race, and Other Social Categories. Oxford: Oxford University Press. Bach, Theodore. 2012. “Gender Is a Natural Kind with a Historical Essence.” Ethics 122 (2): 231–72. https://doi.org/10.1086/663232. Barnes, Elizabeth. 2020. “Gender and Gender Terms.” Nous 54 (3): 704–30. https://doi .org/10.1111/nous.12279. Bernasconi, Robert. 2020. “Who Invented the Concept of Race?” In Theories of Race and Racism: A Reader, 2nd ed., edited by Les Black and John Solomos, 83–103. New York: Routledge. Feminist Philosophy Quarterly, 2025, Vol. 11, Iss. 3, Article 2 Published by Scholarship@Western, 2025 18 Beydoun, Khaled A. 2015. “A Demographic Threat? Proposed Reclassification of Arab Americans on the 2020 Census.” Michigan Law Review First Impressions 114: 1–8. https://repository.law.umich.edu/mlr_fi/vol114/iss1/12/. ————. 2016. “Boxed In: Reclassification of Arab Americans on the US Census as Progress or Peril.” Loyola University Chicago Law Journal 47 (3): 693–759. https://lawecommons.luc.edu/luclj/vol47/iss3/3/. Brigandt, Ingo. 2003. “Species Pluralism Does Not Imply Species Eliminativism.” Philosophy of Science 70 (5): 1305–16. https://doi.org/10.1086/377409. Du Bois, W. E. B. 1898. “The Study of Negro Problems.” The Annals of the American Academy of Political and Social Science 11 (1): 1–23. Dupré, John. 1999. “On the Impossibility of a Monistic Account of Species.” In Species: New Interdisciplinary Essays, edited by Robert A. Wilson, 3–22. Cambridge, MA: MIT Press. Ereshefsky, Marc. 1998. “Species Pluralism and Anti-realism.” Philosophy of Science 65 (1): 103–20. https://doi.org/10.1086/392628. Hey, Jody. 2001. “The Mind of the Species Problem.” Trends in Ecology & Evolution 16 (7): 326–29. https://doi.org/10.1016/S0169-5347(01)02145-0. Jenkins, Katharine. 2019. “Two Routes to Radical Racial Pluralism.” Aristotelian Society Supplementary Volume 93: 49–68. https://doi.org/10.1093/arisup/akz008. ————. 2023. Ontology and Oppression: Race, Gender, and Social Reality. New York: Oxford University Press. Kayyali, Randa. 2013. “US Census Classifications and Arab Americans: Contestations and Definitions of Identity Markers.” Journal of Ethnic and Migration Studies 39 (8): 1299–318. https://doi.org/10.1080/1369183X.2013.778150. Khalidi, Muhammad Ali. 2015. “Three Kinds of Social Kinds.” Philosophy and Phenomenological Research 90 (1): 96–112. doi.org/10.1111/phpr.12020. ————. 2018. “Natural Kinds as Nodes in Causal Networks.” Synthese 195 (4): 1379– 96. https://doi.org/10.1007/s11229-015-0841-y. ————. 2023. Natural Kinds. Cambridge Elements in Philosophy of Science. Cambridge: Cambridge University Press. ————. 2024. “Ontological Pluralism and Social Values.” Studies in History and Philosophy of Science 104: 61–67. doi.org/10.1016/j.shpsa.2024.01.004. Kitcher, Philip. 1984. “Species.” Philosophy of Science 51 (2): 308–33. Longino, Helen E. 2008. “Values, Heuristics, and the Politics of Knowledge.” In The Challenge of the Social and the Pressure of Practice: Science and Values Revisited, edited by Martin Carrier, Don Howard, and Janet Kourany, 68–86. Pittsburgh, PA: University of Pittsburgh Press. Ludwig, David. 2016. “Ontological Choices and the Value-Free Ideal.” Erkenntnis 81 (6): 1253–72. https://doi.org/10.1007/s10670-015-9793-3. Khalidi – Pluralism, Opacity, and Values in the Social World: Commentary on Ontology and Oppression Published by Scholarship@Western, 2025 19 Ludwig, David. 2019. “How Race Travels: Relating Local and Global Ontologies of Race.” Philosophical Studies 176 (10): 2729–50. https://doi.org/10.1007/s110 98-018-1148-x. Mallon, Ron. 2004. “Passing, Traveling and Reality: Social Constructionism and the Metaphysics of Race.” Noûs 38 (4): 644–73. https://doi.org/10.1111/j.0029- 4624.2004.00487.x. Mason, Rebecca. 2020. “Rejecting the ‘Implicit Consensus’: A Reply to Jenkins.” Thought: A Journal of Philosophy 9 (2): 140–47. doi.org/10.1002/tht3.447. Mills, Charles W. 2014. “Notes from the Resistance: Some Comments on Sally Haslanger's Resisting Reality.” Philosophical Studies 171 (1): 85–97. https://doi .org/10.1007/s11098-013-0249-9. Naber, Nadine. 2000. “Ambiguous Insiders: An Investigation of Arab American Invisibility.” Ethnic and Racial Studies 23 (1): 37–61. https://doi.org/10.1080 /014198700329123. Root, Michael. 2000. “How We Divide the World.” Philosophy of Science 67 (S3): S628– S639. https://doi.org/10.1086/392851. Ruben, David H. 1989. “Realism in the Social Sciences.” In Dismantling Truth: Reality in the Postmodern World, edited by Hilary Lawson and Lisa Appignanesi, 58– 75. London: Weidenfeld & Nicolson. Thomasson, Amie L. 2003. “Realism and Human Kinds.” Philosophy and Phenomeno- logical Research 67 (3): 580–609. https://doi.org/10.1111/j.1933-1592.2003 .tb00309.x. Ward, Zina B. 2021. “On Value-Laden Science.” Studies in History and Philosophy of Science 85: 54–62. https://doi.org/10.1016/j.shpsa.2020.09.006. Weber, Max. (1946) 2004. “Science as a vocation.” In The Vocation Lectures, edited by David Owen and Tracy B. Strong, translated by Rodney Livingstone, 1–31. Cambridge: Hackett. West, Cornel. 2002. “A Genealogy of Modern Racism.” In Race Critical Theories: Text and Context, edited by Philomena Essed and David Theo Goldberg, 90–112. Oxford: Blackwell. MUHAMMAD ALI KHALIDI is Presidential Professor of philosophy at the Graduate Center, City University of New York. He specializes in general issues in the philosophy of science, as well as questions in the foundations of the special sciences, especially the cognitive and social sciences. His books include Cognitive Ontology: Taxonomic Practices in the Mind-Brain Sciences (Cambridge, 2023) and Natural Categories and Human Kinds (Cambridge, 2013). He also has research interests in the history of Arabic-Islamic philosophy and has published a translated anthology of Arabic philosophical texts, Medieval Islamic Philosophical Writings (Cambridge, 2005). He has Feminist Philosophy Quarterly, 2025, Vol. 11, Iss. 3, Article 2 Published by Scholarship@Western, 2025 20 published papers in a number of philosophical journals, including Journal of Philosophy, Philosophy of Science, British Journal for Philosophy of Science, Philosophy and Phenomenological Research, Synthese, and Mind & Language. 113002 Khalidi title page 113002 Khalidi Final Format