Substantive Social Metaphysics



Social metaphysics is a source of important philosophical and moral insight. Furthermore, much social metaphysics appears to be substantive. However, some have recently argued that standard views of metaphysics cannot accommodate substantive social metaphysics. In this paper I offer a new diagnosis of this problem and defend a new solution, showing that this problem is an illuminating lens through which to examine the nature and boundaries of metaphysics. This case instantiates a broad, common pattern generated by attempts to align distinctions between realism and anti-realism, mind-independence and mind-dependence, and legitimate and non-legitimate inquiry. I show that the best response is to abandon the association between substantive metaphysics and mind-independence, and I sketch a new definition of substantivity, given in terms of explanatory power, that makes room for substantive social metaphysics while also offering an attractive basis for general metaphysics.

Keywords: Social metaphysics, realism, substantivity

How to Cite: Taylor, E. (2023) “Substantive Social Metaphysics”, Philosophers' Imprint. 23(0). doi:

1. Introduction

Social metaphysics is a source of important philosophical and moral insight. For instance, social metaphysicians investigate questions about the nature of race and gender, which are of urgent concern.1 It is a commonplace claim that some entities are socially constructed; social metaphysics provides tools to articulate this idea and its implications.2 Social metaphysics helps us to understand social structure and the harms in which social structure is implicated, such as patriarchy, white supremacy, and oppression.3 Social metaphysics also provides insight into the nature of social science and the relationships between scientific inquiry and social theory.4

Given this, social metaphysics appears to be (for the most part) substantive. That is, debates in social metaphysics concern deep, genuine questions about the nature of social reality that cannot be settled by a choice of framing device, such as a chosen language or a choice of quantifier.5 However, some philosophers have recently argued that on standard conceptions of metaphysics, most social metaphysics is not substantive. This seeming exclusion has been the subject of much recent discussion, as when Mari Mikkola describes the “apparent antagonism” between mainstream and feminist metaphysics, and Elizabeth Barnes argues that standard approaches to mainstream metaphysics preclude realist social metaphysics.6

In this paper, I offer a new diagnosis of this apparent exclusion of much social metaphysics from substantive general metaphysics, and I offer a new solution. My diagnosis is that this case instantiates a broad, common pattern generated by attempts to align distinctions between realism and anti-realism, mind-independence and mind-dependence, and legitimate and non-legitimate inquiry. Many philosophers take some version of these to align such that realism, defined in terms of mind-independence, marks the boundaries of legitimate inquiry. For example, realists about natural kinds, who think that the kinds that feature in scientific inquiry have boundaries determined by mind-independent causal mechanisms, endorse a version of this view.7 So do explanatory realists, who think that all legitimate explanations must give information about mind-independent metaphysical determination.8 However, attempts to align all three distinctions face counterexamples from inquiry responsive to mind-dependent phenomena.

As I show in Section 3, the case of social metaphysics instantiates this pattern. Realist metaphysicians define substantivity, and hence genuine metaphysical inquiry, in terms of responsiveness to mind-independent features of reality such as structure or fundamentality. Insofar as the answer to a question is settled by mind-dependent factors, debate about that question is not substantive. In much social metaphysics, however, debates often are responsive to mind-dependent factors, so extant definitions of substantivity cannot accommodate substantive social metaphysics.

Placing the problem for social metaphysics under this pattern illuminates the range of possible responses. In Section 4, I argue that the best response in the case of social metaphysics is to abandon the association between substantivity and mind-independence. The solution, therefore, is to develop an account of substantivity without these rigid connections to mind-independence. In Section 5, I sketch a new definition of substantivity, given in terms of explanatory power, that avoids this problem and also offers a plausible, ecumenical basis for general metaphysics.

2. Social Metaphysics

2.1 What is Social Metaphysics?

Recent discussion of the apparent non-substantivity of social metaphysics focuses on one central strand of social metaphysics. This tradition is constructivist in that it portrays many, if not all, social entities as socially constructed. It is emancipatory in that theory choice is motivated at least in part by moral and political considerations. And it is realist in that theory choice is taken by its practitioners to be responsive to the objective structure of reality (though, as we will see, part of what is at issue in this discussion is how to define realism). A leading figure in this tradition is Sally Haslanger, and much of this discussion focuses on Haslanger’s work.9 But many other authors in social metaphysics work in a similarly realist, constructivist, emancipatory spirit.

This kind of social metaphysics takes on a variety of tasks. One is to give an account of particular social kinds, such as race, gender, or class.10 Another is to give an account of social groups, or of social kinds, more generally.11 Another is to address questions about social structure, including accounts of social structure in general, and of particular instances of social structure.12 This latter cluster includes work on the nature of oppression and institutions, and even more detailed and local work on particular double binds.13

There are other forms of social metaphysics beyond this realist, constructivist, emancipatory tradition. For example, deflationists reject the realist idea that their theory choice is responsive to the objective structure of reality, holding that views in social metaphysics are relative to, and settled by, a particular chosen framing.14 Alternatively, many authors approach social metaphysics as a branch of philosophy of science.15 A growing cluster of essentialists use neo-Aristotelian tools to address questions about the nature of the social world.16 And there is the social metaphysics from outside the analytic tradition, which typically does not label itself as ‘metaphysics’ at all.17 In this discussion, however, I will focus on the realist, emancipatory, constructivist strand because it is the central locus of these concerns about substantivity, but I will also address these other approaches as they become relevant.18

2.2 Sources of Non-Substantivity

Three issues are particularly relevant to the apparent exclusion of this realist, constructivist, emancipatory strand of social metaphysics from general metaphysics: amelioration, structure, and fundamentality.19

To say that an area of inquiry is ameliorative is to say that theory choice in that area is responsive to political and more broadly normative considerations, such as whether adopting the theory will improve society.20 This feature has been widely discussed in the metaphysics of gender, where amelioration is taken by many as a requirement for an account of gender.21 Not all social metaphysics is ameliorative, but this is a central feature of many emancipatory, politically oriented projects. However, amelioration appears to be straightforwardly at odds with mainstream approaches to metaphysics. A standard methodology for metaphysics takes it to be a kind of modelling, proceeding through inference to the best explanation.22 In this process, theoretical virtues play a role in theory choice, but normative considerations do not come into play. Accordingly, taking normativity seriously generates a clash between the standard commitments of mainstream metaphysics and of social metaphysics.

A second source of tension is the idea that metaphysics is concerned with structure. The connection between metaphysics and structure has a long heritage, and in contemporary philosophy it is perhaps most famously associated with David Lewis, and more recently Theodore Sider.23 Lewis held that certain properties are perfectly natural, and that these perfectly natural properties are an “elite minority” with features such as appearing in the laws of nature, bearing responsibility for causal powers and resemblance, and acting as reference magnets. Sider extends Lewisian naturalness into the broader notion of structure, which goes beyond properties into domains such as quantification. Sider defends a view of substantivity that appeals to structure, in that what makes a metaphysical debate genuine and substantive is that the candidate answers differ with respect to how accurately they pick out structure.24 Substantive metaphysical debates concern, and are settled by, how well each rival answer limns the structure of reality.

Not every metaphysician shares this picture of structure or Sider’s definition of substantivity, but the rough idea that reality has structure and that it is the concern of metaphysics to describe it is widely shared. However, social groups, properties, and entities are highly non-structural on standard views of structure, which generates problems for social metaphysics. On a structure-based view of substantivity, debates in social metaphysics appear to be automatically non-substantive, because rival views are likely to be as non-joint-carving as each other, given that any view will be about the social world rather than a more structural domain.25 Features in virtue of which a view about social metaphysics is correct will typically not track facts about metaphysical structure, given that the social facts such views aim to capture are often context-dependent, contingent, and subject to change through collective action and shifts in shared beliefs, norms, and language.

A third source of tension between social and general metaphysics comes from the idea that metaphysics is concerned with the fundamental. The nature of the challenge this presents for social metaphysics depends on the precise role of fundamentality in metaphysics. A straightforward version is the idea that metaphysics is concerned with characterizing the most fundamental entities.26 Given that social entities are not fundamental on most definitions of fundamentality, this immediately generates a problem for social metaphysics. An alternative association between substantivity and fundamentality is that substantive metaphysical inquiry concerns whether entities are fundamental or ungrounded, and if not, what they are grounded in, as in Jonathan Schaffer’s approach.27 On this view, metaphysics is about fundamentality but need not be about the fundamental, and Schaffer argues that his framework offers useful tools for feminist metaphysics.28 However, this accommodation of social metaphysics portrays all substantive social metaphysics as concerned with fundamentality, and in Section 4 I will argue that this view is too restrictive.

Not every metaphysician shares these commitments. For example, Karen Bennett holds that there is substantive metaphysics of the non-fundamental; Kit Fine endorses a neo-Aristotelian framework with little emphasis on structure; and Timothy Williamson practices modal logic as metaphysics.29 However, as I will show, the apparent exclusion of this central strand of social metaphysics by some mainstream conceptions of substantivity is an instance of a much broader problem, which every metaphysics must address. As such, it offers a useful lens through which to examine accounts of substantivity and the foundations of metaphysics more generally.

2.3 Where to Go from Here

Each of these features is a source of tension between social metaphysics (of the realist, constructivist, emancipatory sort) and standard approaches to general metaphysics. At this point, one might wonder why anyone should care. After all, feminist metaphysics proceeded without much recognition from mainstream metaphysicians until fairly recently, and need not be recognized as substantive to be useful philosophical and political work. Alternatively, one could adopt a deflationist approach to social metaphysics, or find some other framework that avoids these clashes. One could also reject the idea that we need social metaphysics at all, and pursue projects in philosophy of gender, say, or philosophy of social science, without attempting to show that those projects meet a realist metaphysician’s criteria for substantivity.

However, this problem is worth pursuing further — for diagnosis and for resolution. In its realism, this strand of social metaphysics is close to the heart of mainstream metaphysics. To understand metaphysics in general, it is important to establish whether or not these clashes preclude the possibility of genuinely substantive social metaphysics of this kind. In doing so, we will not merely determine the viability of this kind of social metaphysics but can also use this case as a lens through which to examine the nature and boundaries of metaphysics more generally. Furthermore, diagnosing these issues and illuminating the range of possible responses will clarify the force of this challenge for alternative approaches to social metaphysics, and for other projects in general metaphysics. So while much social and general metaphysics can proceed without addressing this issue, it is necessary to do so if we want to fully understand what metaphysics is, and what social metaphysics is such that it might be a branch of metaphysics.

3. A Pattern: Realism, Mind-Independence, and the Legitimacy of Inquiry

Many philosophers attempt to align realism, mind-independence, and legitimate inquiry such that realism, defined in terms of mind-independence, marks the boundaries of legitimate inquiry. In this section, I will discuss this general pattern and a problem it generates, then return to the case of social metaphysics to show that it instantiates both the pattern and the problem.

There are many different definitions of realism, framed in terms of notions such as objectivity, fundamentality, and structure, and most invoke some aspect of mind-independence.30 For example, in metaphysics, Carrie Jenkins defines metaphysical and ontological realism in terms of mind-independence; Matti Eklund notes that mind-independence is often used to characterize realism in metaphysics; and Sider, Schaffer, and Cian Dorr all defend versions of metaphysical realism that invoke mind-independence.31 In metaethics, the picture is complicated, given the well-documented difficulties involved in distinguishing between sophisticated versions of metaethical realism, irrealism, and anti-realism, but even in this literature, mind-independence is a standard criterion for realism.32 For instance, while acknowledging how difficult it is to unify the many different forms of metaethical realism, Stephen Finlay notes: “we shall see that the variety of metaethical claims labeled ‘realist’ cannot be collectively characterized any less vaguely than as holding that ‘morality’, in some form, has some kind or other of independence from people’s attitudes or practices”.33 In mathematics, Mark Balaguer characterizes mathematical Platonism as the view that mathematical objects exist independently of us and our theorizing, while Justin Clarke-Doane characterizes the question of realism as the question of whether the subjects of our thought and talk exist “independent of us”.34

Given this robust conceptual connection between realism and mind-independence, a further connection to inquiry is motivated by the idea that good inquiry tells us about the world beyond our ideas. It is constrained by, and responsive to, mind-independent reality. Inquiry includes discovery, the formulation of explanations, the classification of phenomena into kinds, and the formulation of generalizations and laws. On this line of thought, part of what makes this activity legitimate is that its subject of study is mind-independent such that the facts discovered, the explanatory relationships, the kinds, and the laws are mind-independent in some way. If we understand realism in terms of mind-independence, and think of proper inquiry as having to tell us about the world outside of our own ideas, then we have some motivation for taking realism, defined in terms of mind-independence, as marking a boundary of legitimate inquiry.

Philosophy is full of attempts to align these distinctions which instantiate this pattern. But such attempts face stubborn counterexamples from legitimate inquiry responsive to mind-dependent phenomena. I call this the Realism-Mind-Independence-Legitimacy (RML) problem:

RML Problem: Any view on which realism, defined in terms of mind-independence, marks the boundaries of legitimate inquiry faces counterexamples from legitimate inquiry responsive to mind-dependent phenomena.

This problem is not new (indeed, as stated here, it is obvious), and has been recognized and discussed across different areas of philosophy. Furthermore, I have left each of these concepts — realism, mind-independence, and legitimacy — undefined, and the seriousness of each instance of the RML problem depends on the precise definitions in play. However, my goal in describing the RML problem at this coarse-grained level is to identify the structure of the problem, and to use the space of responses to the general problem to illuminate the specific case of social metaphysics.

Examples of the RML problem vary in scope, much like accounts of realism. I mentioned two narrower instances in the introduction: explanatory realism faces counterexamples from legitimate explanations that do not give information about mind-independent metaphysical determination, and realism about natural kinds faces counterexamples from kinds central to scientific inquiry that do not have boundaries determined by mind-independent causal mechanisms.35 The RML problem has also troubled attempts to define general forms of realism. For instance, in the face of the RML problem, Gideon Rosen abandons attempting to define realism at all, while Amie Thomasson rejects the idea that realism marks a general boundary of good inquiry, as she argues that it cannot accommodate inquiry into human institutions and artefacts.36

Let us now turn to the case of social metaphysics to see how it instantiates the RML problem. In this case, “realism” is the metaphysical realism characteristic of mainstream metaphysics. “Legitimate inquiry” is substantivity. The notion of “mind-independence” in question is that associated with extant definitions of substantivity, which tie the legitimacy of metaphysical inquiry to its responsiveness to mind-independent factors.

To see how this latter aspect works, consider two approaches to substantivity that have been targeted in these discussions: Sider’s structure and Schaffer’s fundamentality. On the structure approach, a metaphysical debate is substantive if and only if the rival answers vary in the extent to which they accurately limn structure, and the right answer is determined by which view is more structural.37 Structure is a mind-independent phenomenon, in that whether x is more structural than y is a mind-independent matter, and this is crucial to the theoretical role played by structure.38 The fundamentality approach displays a similar profile: a metaphysical debate is substantive if and only if it concerns absolute or relative fundamentality, and the right answer is determined by which view most accurately captures the facts about fundamentality.39 Fundamentality is also mind-independent, in that whether x is more fundamental than y is a mind-independent matter. From these definitions, we can see the alignment of realism, mind-independence, and legitimate inquiry characteristic of the RML pattern.

Social metaphysics is an area of inquiry responsive to mind-dependent factors. There are active debates about the correct definitions of mind-dependence for the purposes of social metaphysics, but the idea that much social metaphysics is concerned with a domain of mind-dependent entities — including kinds, institutions, norms, and laws — has been extensively documented and discussed.40 Some social categories, such as disability classifications, include physical aspects, and many debates in philosophy of race and gender are about whether these categories are mind-dependent in some way. Nonetheless, the idea that much social metaphysics is responsive to mind-dependent factors is uncontroversial. This is particularly evident in the realist, constructivist, emancipatory strand of social metaphysics, as mind-dependence is a central aspect of social construction.

The fact that social metaphysics concerns a domain of mind-dependent phenomena is not in itself enough to generate RML counterexamples, as many areas of metaphysics concern mind-dependent entities. For example, it would be absurd to say that metaphysics of mind is not genuine metaphysics. Given that, it might be tempting to think that social metaphysics is straightforwardly in the clear. However, the problem raised by the mind-dependence of social metaphysics is not the mere fact that social metaphysics concerns a domain of mind-dependent entities. Instead, the problem lies in the fact that what determines the answer to a question in social metaphysics is often something mind-dependent, such as a norm, a law, an expectation, or an institution. This is not the case in metaphysics of mind. In a debate between a dualist and a physicalist, say, the rival positions vary in their claims about the relative fundamentality of physical and mental states, or the relative joint-carvingness of each position. And those relations — relative fundamentality and structure — do not depend on or vary in accordance with anyone’s beliefs. This is true even when mind-dependent entities feature in the relevant relation. If subatomic particles are more fundamental than thoughts, then this is true regardless of any individual’s beliefs, even though a mind-dependent entity lies at one end of the fundamentality relation.

Social metaphysics operates in a different way. Unlike in metaphysics of mind, the correctness of a position in social metaphysics often is determined by its responsiveness to mind-dependent social features. For example, consider debates about social structure, such as “is there oppression?”41 This is a debate about how society works, and the correct answer will be determined in large part by mind-dependent social factors, which are subject to change through human thought and action. Indeed, projects of this kind often aim to identify harmful aspects of the social world in order to change them. This illustrates the dis-analogy with metaphysics of mind — imagine formulating the view that dualism is correct in order to overcome it and eventually establish the truth of physicalism! The problem for social metaphysics is generated not by the mind-dependence of the domain of entities in question, but by the fact that what determines the answer to a question in social metaphysics is often something mind-dependent.

The key to this issue is the connection between mind-independence and substantivity. Traditional definitions of substantivity tie the substantivity of metaphysical debates to their responsiveness to mind-independent metaphysical factors. That connection is reflected in the fact that appeals to structure and fundamentality have been used to protect metaphysics from the deflationist charge that metaphysical debates can be settled by human choices (such as semantic choices or choices of quantifiers) rather than mind-independent features of the world.42 One role for a notion of substantivity is to appropriately ground metaphysics in mind-independent reality and thereby show that metaphysics is not just about human choice. Given this emphasis on mind-independence in definitions of substantivity, the clashes between mainstream metaphysics and this strand of social metaphysics are generated by the worry that the social metaphysics is responsive to mind-dependent factors in a way that the general metaphysics is not. A similar point applies to the prohibition on amelioration in general metaphysics. This is based on the idea that the truths of metaphysics are in some sense independent of human affairs, where the norms that guide ameliorative theory choice reside. The correct view of part-hood, say, or of substance, is not determined by political concerns or considerations about social justice, but instead by mind-independent features of reality.43

Accordingly, this failure to accommodate realist, emancipatory, constructivist social metaphysics instantiates the RML problem. Other approaches to social metaphysics have different resources to address this instance of the RML problem. For example, deflationary approaches straightforwardly avoid it because they reject the realism that treats substantivity as a desirable feature.44 Whether or not essentialist social metaphysics can avoid it depends on the relevant view of essence. Some (if not all) of those who take social metaphysics as a branch of philosophy of science face different versions of the RML problem.45 Although I have focused on one strand of social metaphysics here, the RML problem is relevant to social metaphysics more generally, because it presents a challenge that any viable social metaphysics must overcome.

4. Ways Out

Instances of the RML problem are generated by attempts to align realism, mind-independence, and legitimate inquiry. Ways out involve dropping one of these commitments or attempting to resolve the apparent tension. In this section, I will consider responses to the general RML problem alongside corresponding responses to the specific version of this problem for social metaphysics.

One natural response to the RML problem is to abandon realism in general. Indeed, similar considerations have led many philosophers to embrace anti-realism in other areas. Rejecting general realism about social metaphysics amounts to adopting a deflationary approach; and, as we have noted, the deflationist avoids this problem. However, the starting puzzle was to respond to the exclusion of the realist, emancipatory, constructivist strand of social metaphysics from metaphysics understood in realism terms, and, if possible, to find space for substantive social metaphysics of this type. Below, I will argue that there are promising alternatives that do just that, and so will continue to explore non-deflationary alternatives.

A second response is to abandon the association between realism and mind-independence, and stop defining realism in terms of mind-independence. This strategy has a significant history in metaethics, and some philosophers have more recently defended this combination of views about social metaphysics.46 My primary reason for not recommending this option is that there is a robust, long-standing conceptual connection between realism and mind-independence. Because of this connection, almost any concept that we can use to define more than the most insubstantial notion of realism is likely to implicate mind-independence. This is the case even of fairly minimal definitions of realism given in terms of truth-aptness, because the connection to truth brings in a sense of objectivity and independence (though this depends on the account of truth with which the realism is combined).47 Given this conceptual association between realism and mind-independence, we should expect that attempts to cleanse a definition of realism from any association with mind-independence will leave us with a definition of realism unrecognizable as such. This expectation is borne out by the “problem of creeping minimalism” — when philosophers defend forms of realism that do not invoke mind-independence, their views are often almost impossible to distinguish from irrealist and anti-realist alternatives.48 The further we get from defining realism in terms of the mind-independence of subject matter, as when, for example, we define realism in terms of mere truth-aptness and combine this with minimalist theories of truth, propositions, and so on, the thinner the account of realism becomes, and the harder it is to distinguish from anti-realism and irrealism. This indicates that mind-independence is central to realism, as moving away from robustly mind-independent definitions of realism has historically undermined attempts to distinguish realism from its alternatives.

There is more to say here, as the rich literature in metaethics indicates.49 Furthermore, divorcing realism from mind-independence might be effective in some areas of philosophy but not others. Some of the most interesting attempts to embrace this response to the RML problem come from metaethics, which indicates that there may be something about metaethics that makes this strategy particularly promising.50 In the meantime, the robustness of the conceptual connection between mind-independence and realism provides good grounds to explore other solutions, at least in the case of social metaphysics.

A third option is to maintain the alignment between realism, mind-independence, and legitimate inquiry, but to argue that the apparent problem cases of inquiry into mind-dependent domains are not counterexamples because they bear the right relationship to mind-independent domains to count as legitimate. An intuitive way to do this is through an appeal to grounding. The rough idea is that the mere prospect of mind-dependence is not enough to threaten the legitimacy of inquiry, so long as the mind-dependent facts are grounded in mind-independent facts. This seems like a promising avenue because, given plausible naturalistic assumptions about mentality, we can expect mind-dependent facts to be grounded in mind-independent physical facts. This strategy mirrors the structure of physicalist defenses of the autonomy of the mind and brain sciences against the threat of non-naturalism.51 Furthermore, contemporary theories of grounding were developed in response to concerns about defining realism, and so there is precedent for using grounding to address similar worries.52

I will take grounding to be a non-causal explanatory relationship between facts, such that some fact(s) can ground, and therefore explain, some other fact.53 Grounding theorists often use the ‘in virtue of’ idiom to capture this explanatory aspect. I will presume that if some fact is grounded in another, it does not follow that the grounded fact is eliminated or reduced, but it does follow that the grounding fact is more fundamental than the grounded fact. I will also presume that grounded facts are necessitated by their grounds.

The worry in the case of social metaphysics is that substantive metaphysical inquiry responsive to mind-dependent social features is not permitted on most views of substantivity. However, if the facts about mind-dependent phenomena are grounded in facts about mind-independent phenomena, then we can rescue the idea that inquiry into the social world is substantive after all, because the mind-dependence of its subject matter is related to mind-independent facts in a way that preserves its legitimacy. A similar approach can be adopted towards the worry about amelioration. If facts about amelioration are appropriately grounded in facts about mind-independent metaphysical structure warranted by naturalistic presumptions about ethics, then the ameliorative aspect of social metaphysics is less worrying.

Despite appearing promising, however, this version of the grounding strategy fails to resolve the RML problem in the case of social metaphysics. Recently, a number of authors have defended grounding-based approaches to social metaphysics, so my replies here cannot be exhaustive, but my primary concerns are: first, that these views fail to defend social metaphysics as legitimate inquiry; and second, that grounding-based conceptions of social metaphysics are too restrictive.

On the first concern, if the defense of the legitimacy of inquiry into certain facts comes through a defense of the legitimacy of inquiry into their grounds, then this suggests that inquiry into the more fundamental facts is better, and that inquiry into the grounded facts is a ladder to eventually kick away in favor of the more fundamental inquiry. The real action is happening in the grounds, and in the social case, these grounds are very far away from the social entities themselves, as they are the fundamental physical facts that ground the relevant mental states. This is an inadequate defense of social metaphysics, because it fails to defend social metaphysics as pitched at the social level. The parallel case from philosophy of science would be to defend the legitimacy of economics through an appeal to the legitimacy of physics. It is no real defense of the practice of economics as a science to say that economic facts are ultimately grounded in physical facts, and that inquiry into physical facts is real science. Such considerations are familiar from well-trodden debates about reductionism.54

On the second concern, Schaffer has defended an alternative grounding strategy to accommodate social metaphysics in response to Barnes’ argument that his definition of substantivity leaves no room for social metaphysics.55 On Schaffer’s view, substantive metaphysical debates are about fundamentality — about whether entities are grounded, and if so, what it is in which they are grounded.56 This approach avoids the first concern because it takes the primary concern of metaphysics to be with fundamentality rather than the fundamental. Schaffer argues that his framework offers useful resources for articulating claims about social construction, and as such is an appropriate basis for feminist, and more broadly social, metaphysics. I agree that this framework offers useful framing for claims about social construction, but to take all substantive social metaphysics to be focused on questions about fundamentality is simply too restrictive. Some debates in social metaphysics can be framed in terms of grounding, but others cannot. Some social metaphysics involves identifying elements of social structure, such as oppressive double binds. For example, Talia Mae Bettcher argues that people who are transgender face a specifically pernicious double bind between disclosing their status as transgender, and thereby risking violence and censure, or not disclosing it, and facing an equal risk of violence and censure.57 The work here is not about establishing what the double bind is grounded in. Instead, the work focuses on identifying this aspect of social structure and the role that it plays in generating transphobic violence. It would be a struggle to reframe this work in terms of grounding, and this example indicates that social metaphysics includes, but also goes beyond, inquiry into the grounds of social categories.

There is further room here for the defender of grounding-based approaches to social metaphysics to reply. For instance, Aaron Griffith has argued that, contra Barnes, debates about the metaphysics of gender can be reconstructed as debates about grounding, and perhaps a similar strategy could be applied to the Bettcher case.58 However, my preferred strategy is to take grounding as a useful resource for certain tasks while recognizing its limitations as a basis for social metaphysics in general.

The final option is to abandon the association between realism (understood in terms of mind-independence) and legitimate inquiry. Abandoning this association will look like this: feel free to endorse realism about a given domain, such that there are facts or entities distinctive of that domain which do not depend on human thought and interest. But do not make the further connection between realism and inquiry. That is, do not endorse general theories of explanation, generalization, discovery, laws, theories, or kinds given in purely realist, and therefore mind-independent, terms. Feel free to draw distinctions between different kinds of inquiry. We could even call some areas of inquiry mind-dependent and others mind-independent. But do not expect that good explanations will only give information about mind-independent relationships, or that the kinds and predicates that appear in good theories and laws will only limn mind-independent structure, or that the kinds that appear in good scientific theories will have their boundaries determined by purely mind-independent causal mechanisms. On this approach, some explanations, theories, laws, and kinds may have these features. But we should not expect them all to have these features. Nor should we generally prize inquiry with the right connections to mind-independence. It might be that inquiry with certain connections to mind-independence is particularly useful for certain purposes. But overall, embracing this view amounts to abandoning the idea that inquiry is better in general when it displays this connection to realism.59

In the case of social metaphysics, abandoning the connection between realism and legitimate inquiry amounts to abandoning a definition of substantivity given in terms of mind-independence. On this view, insofar as social metaphysics is responsive to mind-dependent phenomena, it is not realist and not mind-independent, but it can be substantive. We can, for example, use talk of social joints, and have social kinds feature in laws and generalizations, safe in the knowledge that this work is as substantive as the metaphysics of the mind-independent. We might sometimes want to draw distinctions between mind-independent and mind-dependent phenomena and can use metaphysical tools to do so. However, on this new picture, we need not take responsiveness to mind-independent features such as structure and fundamentality to mark a general boundary of substantive metaphysics.

5. Substantivity

5.1 Desiderata

A primary role for an account of substantivity is to protect metaphysics from the threat that metaphysical debates can be settled by framing choices that are arbitrary with respect to describing reality, though they may not be arbitrary with respect to other goals, such as aligning with everyday language use. Consider, for example, an ontological debate about whether or not tables exist.60 On a roughly deflationist line of thought, this debate is about how to use language. Once we settle our use of the word ‘table’, the debate is settled, and the choice between different uses of the word is arbitrary in that none is better than any other for the purposes of describing reality.61 A similar threat is generated by the worry that debates about existence involve a choice of quantifier, or a domain for a quantifier.62 Established definitions of substantivity ground the genuineness of metaphysical debates in mind-independent reality and so avoid the worry that those debates can be so trivially settled through human choice. However, as we have seen, in tying substantivity to mind-independence so robustly, these definitions face the RML problem. We need a way to avoid the worry that metaphysical debates can be settled by arbitrary human choice, without tying substantivity to mind-independence in a way that raises the RML problem.

For insight, let us turn to the social domain. This is a useful place to begin not only because it is our motivating subject-matter, but also because it illustrates a central point between mind-independence and arbitrariness. Although many social entities are mind-dependent, debates about social structure cannot be settled by the arbitrary choices of individual inquirers. To illustrate, consider once again the question: “Is there oppression?” This is a question about how society works and how it is structured. In engaging with this question, we must ask: are there patterns in the ways that people from certain groups are advantaged or disadvantaged? If so, what are the mechanisms through which such patterns are created and maintained? The answers are not up to the choice of individual inquirers, but oppression is mind-dependent in that its existence relies on people’s beliefs and on the institutions, norms, laws, and so on constituted by those beliefs. Furthermore, whether or not there is oppression is up to us, in that we can take social action to undermine true generalizations about how advantage and disadvantage work, and in doing so change the answers to the questions. But the mind-dependence of oppression, and its being subject to change through human thought and action, does not make debate about oppression arbitrary. This example illustrates what it is for debates about mind-dependent phenomena to be non-arbitrary.

The case of inquiry into oppression illuminates two broad desiderata for a new approach to substantivity. The first is the need to avoid deflationary arbitrariness. Substantive metaphysical debates cannot be settled by arbitrary human choice, whether that involves selection of a conceptual framework, a definition, a quantifier, or some other framing device. This is a primary, original role for substantivity. The second is that a definition of substantivity must not in itself settle whether or not inquiry responsive to mind-dependent phenomena is substantive. An account of substantivity must leave open the possibility of substantive inquiry responsive to mind-dependent phenomena, though all such inquiry may turn out to be non-substantive. To put this differently, the substantivity or otherwise of inquiry responsive to mind-dependent phenomena should not be a matter of definition. This requirement is generated by the need to avoid the RML problem.

5.2 Substantivity and Explanatory Power

I propose to define substantivity in terms of explanatory power, as follows:

Explanatory Substantivity: A debate is substantive if and only if the rival views differ in their explanatory power.

Although a full development and defence of this proposal cannot be given here, I will sketch enough detail to show that this proposal can meet the desiderata given above, can accommodate social metaphysics, and has some independent plausibility as an approach to general metaphysics.

To illustrate the proposal, consider the two cases discussed earlier in this section: the debate about tables (“are there tables?”), and the debate about oppression (“is there oppression?”). On this proposal, each debate is substantive if and only if the rival answers differ in their explanatory power, which includes their capacity to explain relevant phenomena, and the number of facts the proposals leave unexplained. The explanandum phenomena in the case of the table may include scientific findings, human linguistic practice, common-sense intuition, and other established views about the nature of reality, while the target phenomena in the case of oppression may include facts about patterns of distribution and access to material goods, employment, education, violence, and so on. In order for this proposal to satisfy the two desiderata for an account of substantivity, explanatory power must be understood in neither wholly realist nor wholly anti-realist terms. To avoid the RML problem, explanatory power cannot reduce to one of the extant, mainstream bases of substantivity, such that a theory displays higher explanatory power the closer it gets to limning structure or capturing fundamentality relations. It must be possible, on this view, for a theory to do worse with respect to these features while doing better with respect to explanatory power. I will not argue for this view here, but will note that it captures uncontroversial intuitions about explanation. If you think that sometimes an explanation pitched at the social level is better than one pitched at a more fundamental, or joint-carving, level, then you are on board with the view of explanatory power I have in mind. In order to avoid arbitrariness, however, explanatory power cannot be completely subjective either. It cannot be the case that the extent to which some proposal displays explanatory power is entirely up to the choice or interests of some individual. As before, I will not argue for this here, but it is also a fairly minimal requirement, amounting to the denial of the claim that whether and how well x explains y is entirely up to the individual.

Let us now see how this sketched proposal handles cases. A central test case in recent conversations about the substantivity of social metaphysics is the metaphysics of gender. Because there is so much explanatorily at stake in these debates, my approach straightforwardly classifies them as substantive. Consider an extract from Simone de Beauvoir’s The Second Sex where she considers and rejects a series of views about what it is to be a woman on the basis of their not adequately explaining certain phenomena.63 For example, the position de Beauvoir describes as “nominalism”, on which women are, “among human beings, merely those who are arbitrarily designated by the word ‘woman’”, she rejects as simply denying, rather than explaining, the evident fact that “humanity is split into two categories of individuals”.64 On my approach, this is substantive inquiry into the metaphysics of gender. Alternatively, consider an explanatory framing of more recent conversations about gender. Some have argued that transgender women experience oppression of a kind similar to that faced by cis women.65 Arguably, a view of gender on which transgender women are, straightforwardly, women, does a better job of explaining this fact than alternatives. Given that, this debate is substantive because the alternatives differ in their capacity to explain the relevant facts.

In both of these cases there are open debates about which facts require explanation by the theory. For instance, de Beauvoir’s nominalist may reject the claim that there is an obvious difference between different categories of humans, and some may reject the claim that the oppression faced by transgender women is similar in nature to the oppression faced by cis women. However, this kind of openness about the data that must be accommodated by a theory is in keeping with standard metaphysical practice. When selecting between metaphysical theories on the basis of their explanatory power, questions about what is at stake in theory choice are traditionally subject to as much philosophical attention as theory choice itself.66

Now, let us consider some non-substantive debates. Debates about whether to classify a gathering of around eight people as ‘people hanging out’ or as ‘a party’ are unlikely to be substantive on this approach, because there is so little explanatorily at stake. We may have preferences about how to use words, but there are no significant explanatory losses on either choice of classification. A less straightforward case is Bennett’s evocative example of whether it is correct to call a fruit-flavoured alcoholic drink served in a V-shaped glass a ‘martini’.67 Bennett presents this as a merely semantic debate, and as such non-substantive, but on my proposal the result is not so straightforward. It may turn out that there is no explanatory difference between the proposal that the fruit-flavoured drink is a martini and the proposal that it is not (and I suspect that this is the case). However, the proposals may differ with respect to how well they capture certain social, historical, normative, or legal facts, and as such this debate may turn out to be substantive. For instance, if the name ‘martini’ indicates a connection to the name of a brand of vermouth, then views that accommodate this history and so require the presence of at least traces of vermouth in a genuine martini, may differ in explanatory power from views that do not.

There is much more work to be done in developing this proposal in detail. For instance, some may worry that it is too permissive, as it may end up treating any genuine debate about classification, or debates between deflationists, as substantive metaphysics. In order to address this kind of concern, more detail about what explanation is and what it is for proposals to differ in explanatory power is needed. Furthermore, if explanatory power comes in degrees, then substantivity may also be a matter of degree, which would offer enriched resources for handling cases. Overall, however, this sketch is enough to show that a definition of substantivity given in terms of explanatory power can make room for social metaphysics without falling foul of the worry that metaphysical debates can be settled through arbitrary framing choices. Furthermore, there is space to reject this particular approach to substantivity while accepting that the solution to this problem for social metaphysics is to loosen the ties between substantivity and mind-independence. Those who adopt this route are free to develop an alternative account of substantivity that avoids the threat of arbitrariness while making room for social metaphysics.

This explanatory approach to substantivity is also generalizable. Wherever we find ties between realism, mind-independence, and the legitimacy of inquiry, we will face the RML problem. Accordingly, an account of substantivity that does the original, motivating work of avoiding arbitrariness, without tying substantivity to mind-independence, will prove valuable for all areas of metaphysics. I have mentioned some examples in the literature on the metaphysics of explanation and natural kinds of views in non-social metaphysics that generate RML-style counterexamples. However, any project in metaphysics that attempts to apply metaphysical tools to mind-dependent phenomena, or to articulate connections between metaphysics and inquiry, will benefit from this approach.

6. Conclusion

I have argued that the apparent exclusion of a central strand of social metaphysics from substantive general metaphysics is an instance of a broader pattern in which attempts to align realism, mind-independence, and legitimate inquiry face counterexamples generated by legitimate inquiry responsive to mind-dependent phenomena. My solution is to abandon the connection between realism (defined in terms of mind-independence) and substantive metaphysics. Doing so allows for the possibility of substantive social metaphysics without abandoning realism or attempting to cleanse realism of any association with mind-independence. This proposal requires a new definition of substantivity, and I have sketched a proposal to define substantivity in terms of explanatory power.

Some might argue that allowing metaphysics to have a closer association with the mind-dependent makes it unrecognizable as metaphysics. However, the idea that metaphysics is so purely concerned with the mind-independent is a fairly recent idea. Many historical figures — think of Kant or perhaps Spinoza — conceived of metaphysical inquiry as having a much closer relationship with the mind and as being driven by human interest and moral considerations as much as the mind-independent structure of reality.68 So although abandoning the association between mind-independence and substantive metaphysical inquiry might seem radical at first, it has historical precedent. And, as we have seen, this view has the pleasing upshot of making room for substantive social metaphysics.69


  1. Such as Appiah, K. A. (1996), Ásta (2018), Dembroff, R. (2018), Glasgow, J. et al. (2019), Griffith, A. (2020), Jenkins, K. (2016), Haslanger, S. (2000), Mills, C. (2000), and Ritchie, K. (2013). [^]
  2. Such as Ásta (2018), Díaz-León, E. (2015), Griffith, A. (2020), Hacking, I. (2000), and Haslanger, S. (2012). [^]
  3. Such as Barnes, E. (2014), Barnes, E. (2017), Cudd, A. (2006), Bettcher, T. M. (2007), Haslanger, S. (2012), Ritchie, K. (2021), and Taylor, E. (2016). [^]
  4. Such as Epstein, B. (2015), Godman, M. (2015), Guala, F. (2016), Khalidi, M. A. (2013), Khalidi, M. A. (2015), and Mallon, R. (2016). [^]
  5. For further discussion of substantivity and deflationism, see Bennet, K. (2009), Chalmers, D. (2009), Hirsch, E. (2005), Sider, T. (2011, Chapter 5), and Thomasson, A. (2015, Part 1). [^]
  6. Mikkola, M. (2017), Barnes, E. (2014), and Barnes, E. (2017). The term ‘mainstream metaphysics’ is used for a popular realist conception of metaphysical inquiry. For discussion, see Chalmers, D. et al. (2009, pg. 3–4). Rebecca Mason and Katherine Ritchie also document the exclusion of social metaphysics from prestigious metaphysics venues in Mason, R. & Ritchie, K. (2020). [^]
  7. Realists about natural kinds include Boyd, R. (1991), and Mallon, R. (2016). [^]
  8. Explanatory realists include Audi, P. (2012), Audi, P. (2015), Kim, J. (1988), Kim, J. (1993), Kim, J. (1994), and Ruben, D-H. (1990). See discussion in Kovacs, D. (2017) and Taylor, E. (2018). [^]
  9. I will focus on Mikkola, M. (2017), Barnes, E. (2014), and Barnes, E. (2017) in particular, but see also Díaz-León, E. (2018). [^]
  10. Such as Dembroff, R. (2018), Haslanger, S. (2000), Griffith, A. (2020), and Jenkins, K. (2016). [^]
  11. Such as Ritchie, K. (2013) and Ritchie, K. (2020). [^]
  12. Such as Griffith, A. (2018) and Haslanger, S. (2012). [^]
  13. Such as Bettcher, T. M. (2007), Cudd, A. (2006), Taylor, E. (2016), and Young, I. M. (1988). [^]
  14. Such as Ásta (2018). Esa Díaz-León has also argued that, contra Barnes, a deflationary approach accommodates even apparently realist projects in social metaphysics. See Díaz-León, E. (2018). [^]
  15. Such as Epstein, B. (2015), Godman, M. (2015), Guala, F. (2016), Khalidi, M. A. (2015), and Mallon, R. (2016). [^]
  16. Such as Mason, R. (2020), Passinsky, A. (2020), and Passinsky, A. (2021). [^]
  17. Such as de Beauvoir, S. (2011), Butler, J. (2006), and Foucault, M. (1977). [^]
  18. For ease of expression, I will sometimes refer to “the” problem for social metaphysics, or “the” case of social metaphysics, meaning the apparent exclusion of this strand of social metaphysics from substantive general metaphysics. [^]
  19. Primary texts focusing on these issues are Barnes, E. (2014), Barnes, E. (2017), Díaz-León, E. (2018), Mason, R. & Ritchie, K. (2020), and Mikkola, M. (2017). [^]
  20. In literature on conceptual engineering, the term ‘amelioration’ is used more broadly to mean any change to a concept that aims to improve the concept, which can include but is not exhausted by politically motivated improvements. For example, Carnapian explication of a scientific concept is ameliorative in this broad sense. I am using the term in a narrower sense to focus on specifically political and ethical motivations. See Cappelen, H. (2018), Burgess, A. & Plunkett, D. (2013), Flocke, V. (2020), and Haslanger, S. (2012). [^]
  21. For instance, Katherine Jenkins has argued that Haslanger’s account of gender fails to meet Haslanger’s own ameliorative standards because it fails to always classify people who are transgender as belonging to the gender with which they identify. See discussion in Jenkins, K. (2016), Haslanger, S. (2012), and Haslanger S. (2000). [^]
  22. See discussion in Paul, L. (2012), Sider, T. (2009), and Swoyer, C. (1999). [^]
  23. Lewis. D. (1983), Sider, T. (2009), Sider, T. (2011). [^]
  24. Sider, T. (2011, Chapter 4). [^]
  25. Though Sider resists this objection in Sider, T. (2017). [^]
  26. See Dorr, C. (2005) and discussion in Barnes, E. (2014). [^]
  27. Schaffer, J. (2009). [^]
  28. Schaffer, J. (2009), Schaffer, J. (2017a). [^]
  29. Bennett, K. (2017), Fine, K. (2001), Fine, K. (1994), Williamson, T. (2013). [^]
  30. I am presupposing a notion of realism which is richer than mere truth-aptness, typically contrasted with irrealism rather than anti-realism. However, even the thinner notions of realism understood in terms of mere truth-aptness can preserve an aspect of mind-independence through the connection to truth, though the extent to which this is the case depends on the account of truth (see discussion in Section 4). Furthermore, accounts of realism not explicitly framed in terms of mind-independence typically maintain some connection to mind-independence. For instance, Michael Dummett takes commitment to bivalence as a marker of realism, and invokes mind-independence through the realist’s understanding of meaning and of truth. See Dummett, M. (1978). See also discussion in Devitt, M. (1991). [^]
  31. Jenkins, C. (2005), Jenkins, C. (2010), Eklund, M. (2006), Sider, T. (2009), Sider, T. (2011), Dorr, C. (2005). [^]
  32. See discussion in Dreier, J. (2004). Some metaethicists do explicitly reject the mind-independence criterion for realism, such as Blackburn, S. (1993). I will return to this point in Section 4. [^]
  33. Finlay, S. (2007, pg. 820). [^]
  34. Balaguer, M. (1998, pg. 5), Clarke-Doane, J. (2020, pg. 4). [^]
  35. Explanatory realists include Audi, P. (2012), Audi, P. (2015), Kim, J. (1988), Kim, J. (1993), Kim, J. (1994), and Ruben, D-H. (1990). See also discussion in Taylor, E. (2018) and Kovacs, D. (2017). Realists about natural kinds include Boyd, R. (1991) and Mallon, R. (2016). [^]
  36. Rosen, G. (1994), Thomasson, A. (2003). [^]
  37. Sider, T. (2011, Chapter 4). [^]
  38. As Sider, T. (2011, pg. 65) puts it while discussing the epistemic role of structure:

    … joint-carving languages and beliefs are better. If structure is subjective, so is this betterness. This would be a disaster …. If there is no sense in which the physical truths are objectively better than the scrambled truths, beyond the fact that they are propositions that we have happened to have expressed, then the postmodernist forces of darkness have won.

  39. Schaffer, J. (2009, pg. 350–354). [^]
  40. See Guala, F. (2013), Mason, R. (2021), and Mason, R. & Ritchie, K. (2020). [^]
  41. Cudd, A. (2006) offers an overview of this debate. [^]
  42. For example, see discussion in Eklund, M. (2006) and Hirsch, E. (2005). [^]
  43. If moral norms are mind-independent, then ameliorative theory choice might not involve mind-dependence, but social metaphysicians who endorse amelioration do so regardless of metaethical debates about the mind-independence of morality, which indicates that amelioration is acceptable to social metaphysicians even if ameliorative factors are mind-dependent. [^]
  44. As in Díaz-León, E. (2018). [^]
  45. For example, in Taylor, E. (2020), I argue that Mallon’s realist view of social categories as natural kinds faces a version of the RML problem (though not under that description). [^]
  46. Such as Blackburn, S. (1993) in metaethics, and Mason, R. (2020) in social metaphysics. Muhammad Ali Khalidi argues that mind-dependence is not a defining characteristic of social kinds, and thus that realism or otherwise about social kinds should not be understood in terms of mind-independence. See Khalidi, M. A. (2015). [^]
  47. Something similar is true of Dummett’s use of bivalence as a marker of realism. Although Dummett does not explicitly define realism in terms of mind-independence, his realist about a given domain takes the meaning and truth of statements about that domain to be “transcendent” of inquirers. See Dummett, M. (1978). [^]
  48. See discussion in Dreier, J. (2004). [^]
  49. For example, some forms of moral realism invoke facts about humans, such as Peter Railton’s moral realism. On Railton’s view, moral value is human value, and as such is dependent on facts about humans. Perhaps this, then, is a robust, genuine form of realism that divorces realism from mind-independence. However, although Railton’s conception of moral value is dependent on facts about humans, it is importantly not dependent on human ideas or beliefs. Accordingly, although there is an importantly human aspect at the heart of Railton’s realism, the relevant facts about humans are not facts about human minds, and so he maintains the traditional connection between realism and mind-independence. See Railton, P. (1986). [^]
  50. For discussion of parallels between realism/anti-realism debates in ethics and in philosophy of science, and including some reflection on differences in the nature of the role of inquiry, see Callender, C. (preprint). [^]
  51. Traditionally, non-reductive physicalists appealed to supervenience to articulate such views, but see Ney, A. (2016) for a defense of grounding in a similar role. [^]
  52. Fine, K. (2001), Rosen, G. (2010), Berker, S. (2018), Schaffer, J. (2009), Schaffer, J. (2017b). [^]
  53. This cluster of commitments would not be endorsed by every grounding theorist, but it is a fairly uncontroversial summary of non-sentential forms of grounding. For discussion, see Correia, F. & B. Schnieder (2012). [^]
  54. Such as Fodor, J. (1974). [^]
  55. Schaffer, J. (2017a). [^]
  56. Schaffer, J. (2009). [^]
  57. In Bettcher, T. M. (2007). Barnes makes a similar point about debates about gender in Barnes, E. (2014, pg. 344), arguing that certain debates about gender remain unresolved even after the fundamentality facts are established. Griffith, A. (2018) responds, arguing that these debates are appropriately understood as debates about grounding. [^]
  58. Griffith, A. (2018). [^]
  59. There is precedent for such views. For example, Dasgupta, S. (2018) defends a form of realism without what he calls the valuing of joint-carving properties that is a feature of Lewisian approaches to metaphysics. Cohen, J. & Callender, C. (2009) have developed an adapted version of Lewis’ Best System Account of laws, on which the predicates that feature in the laws are not necessarily metaphysically joint-carving. None of these authors is a full-blown anti-realist, but they reject general connections between realism and the legitimacy of inquiry. [^]
  60. An account of substantivity need not deliver the result that this particular debate is substantive. I offer this case simply to illustrate the threat of arbitrariness. A permissivist about existence questions may wish to ask an alternative question, such as “Are tables fundamental?” [^]
  61. Such as one of the classic Carnapian critiques of metaphysics in Carnap, R. (1950). [^]
  62. See discussion in Chalmers, D. (2009). [^]
  63. de Beauvoir, S. (2011). [^]
  64. The full quote reads: [^]
  65. …the truth is that anyone can clearly see that humanity is split into two categories of individuals with manifestly different clothes, faces, bodies, smiles, movements, interests, and occupations; these differences are perhaps superficial; perhaps they are destined to disappear. What is certain is that for the moment they exist in a strikingly obvious way” (de Beauvoir, S. 2011, pg. 4). [^]
  66. For example, Dembroff, R. (2019). [^]
  67. Consider debates about the role that science, common-sense judgment, or ordinary language use should play in first-order debates, as in recent conversations about the role of scientific data in metaphysics. See Bryant, A. (2021) and McKenzie, K. (2020). [^]
  68. Bennett, K. (2009, pg. 50). [^]
  69. For instance, Moore, A. W. (2012, pg. 44–66) describes Spinoza’s metaphysics as “metaphysics in service of ethics”. [^]
  70. With thanks to Simona Aimar, Fatema Amijee, Justin Bledin, Craig Callender, Patrick Connolly, Michael Della Rocca, Aaron Griffith, Michelle Kosch, Marc Lange, Jacob Lettie, and Hanna Pickard for helpful comments and discussion. Particular thanks to Elizabeth Miller for working through an early version of the project. Thanks to audiences at the Southern California Metaphysics Network, the Pacific Division Meeting of the American Philosophical Association, the Social Metaphysics Workshop, the University of Wisconsin-Madison, the University of Oklahoma, and seminars at the University of Gothenburg, Frei Universität Berlin, and the Arché Center at the University of St Andrews for stimulating and helpful discussion. Much of this paper was written during various pandemic lockdowns and could not have been completed without feedback and support from two online groups: the Pre-Tenure Women’s online working group and the WiM Research Network. Deepest thanks to the members of both. Finally, many thanks to two anonymous referees from this journal for their detailed, engaged, and supportive critique and comments.


Appiah, Kwame Anthony (1996) “Race, Culture, Identity: Misunderstood Connections.” In Appiah, Kwame Anthony & Gutmann, Amy (1996) Color Conscious: The Political Morality of Race. Princeton University Press.

Ásta (2018) Categories We Live By: The Construction of Sex, Gender, Race, and Other Social Categories. Oxford University Press.

Audi, Paul (2012) “Grounding: Towards a Theory of the In-Virtue-of Relation.” Journal of Philosophy 109(12): 685–711.

Audi, Paul (2015) “Explanation and Explication.” In Daly, Chris (Ed.) (2015) The Palgrave Handbook of Philosophical Methods. London: Palgrave Macmillan.

Balaguer, Mark (1998) Platonism and Anti-Platonism in Mathematics. Oxford University Press.

Barnes, Elizabeth (2014) “Going Beyond the Fundamental: Feminism in Contemporary Metaphysics.” Proceedings of the Aristotelian Society 114 (3): 335–351.

Barnes, Elizabeth (2017) “Realism and Social Structure.” Philosophical Studies 174 (10): 2417–2433.

De Beauvoir, Simone (2011) (Translated by Borde, Constance & Malovany-Chevallier, Sheila) The Second Sex. Vintage.

Bennett, Karen (2009) “Composition, Colocation, and Metaontology.” In Chalmers, David, Manley, David & Wasserman, Ryan (2009) Metametaphysics: New Essays on the Foundations of Ontology. Oxford University Press.

Bennett, Karen (2017) Making Things Up. Oxford University Press.

Berker, Selim (2018) “The Unity of Grounding.” Mind 127 (507): 729–777.

Bettcher, Talia Mae (2007) “Evil Deceivers and Make-Believers: On Transphobic Violence and the Politics of Illusion.” Hypatia 22 (3): 43–65.

Blackburn, Simon (1993) Essays in Quasi-Realism. Oxford University Press.

Bryant, Amanda (2021) “Epistemic Infrastructure for a Scientific Metaphysics.” Grazer Philosophische Studien 98: 27–49.

Burgess, Alexis & Plunkett, David (2013) “Conceptual Ethics I.” Philosophy Compass 8 (12): 1091–1101.

Butler, Judith (2006) Gender Trouble. Routledge.

Callender, Craig (preprint) “Humean Laws of Nature: The End of the Good Old Days.” Archived:

Cappelen, Herman (2018) Fixing Language: An Essay on Conceptual Engineering. Oxford University Press.

Carnap, Rudolf (1950) “Empiricism, Semantics, and Ontology.” Revue Internationale de Philosophie 4 (11): 20–40.

Chalmers, David (2009) “Ontological Anti-Realism.” In Chalmers, David, Manley, David, & Wasserman, Ryan (2009) Metametaphysics: New Essays on the Foundations of Ontology. Oxford University Press.

Chalmers, David, Manley, David & Wasserman, Ryan (2009) Metametaphysics: New Essays on the Foundations of Ontology. Oxford University Press.

Clarke-Doane, Justin (2020) Morality and Mathematics. Oxford University Press.

Cohen, Jonathan & Callender, Craig (2009) “A Better Best System Account of Lawhood.” Philosophical Studies 145: 1–34.

Correia, Fabrice & Schnieder, Benjamin (2012) Metaphysical Grounding: Understanding the Structure of Reality. Cambridge University Press.

Cudd, Ann (2006) Analyzing Oppression. Oxford University Press.

Dasgupta, Shamik (2018) “Realism and the Absence of Value.” Philosophical Review 127(3): 279–322.

Dembroff, Robin (2018) “Real Talk on the Metaphysics of Gender.” Philosophical Topics 46 (2): 21–50.

Dembroff, Robin (2019) “Trans Women are Victims of Misogyny, Too — and All Feminists Must Recognize This.” The Guardian:

Devitt, Michael (1991) Realism and Truth: Second Edition. Princeton University Press.

Díaz-León, Esa (2015) “What is Social Construction?” European Journal of Philosophy 23 (4): 1137–1152.

Díaz-León, Esa (2018) “On Haslanger’s Meta-Metaphysics: Social Structures and Metaphysical Deflationism.” Disputatio 10 (50): 201–216.

Dorr, Cian (2005) “What We Disagree About When We Disagree About Ontology.” In Kalderon, Mark Eli (2005) Fictionalism in Metaphysics. Oxford University Press.

Dreier, James (2004) “Meta-Ethics and The Problem of Creeping Minimalism.” Philosophical Perspectives 18: 23–44.

Dummett, Michael (1978) “Realism.” In Dummett, Michael (1978) Truth and other Enigmas. Harvard University Press.

Eklund, Matti (2006) “Metaontology.” Philosophy Compass 1 (3): 317–334.

Epstein, Brian (2015) The Ant Trap: Rebuilding the Foundations of the Social Sciences. Oxford University Press.

Fine, Kit (1994) “Essence and Modality.” Philosophical Perspectives 8: 1–16.

Fine, Kit (2001) “The Question of Realism.” Philosophers’ Imprint 1: 1–30.

Finlay, Stephen (2007) “Four Faces of Moral Realism.” Philosophy Compass 2 (6): 820–849.

Flocke, Vera (2020) “How to Engineer a Concept.” Philosophical Studies 178 (10): 3069–3083.

Fodor, Jerry (1974) “Special Sciences.” Synthese 28: 97–115.

Foucault, Michel (1977) “Truth and Power.” In Calhoun, Craig J. (2007) Contemporary Sociological Theory. Blackwell.

Glasgow, Joshua, Haslanger, Sally, Jeffers, Chike & Spencer, Quay-shawn (2019) What Is Race? Four Philosophical Views. Oxford University Press.

Godman, Marion (2015) “The Special Science Dilemma and How Culture Solves It.” Australasian Journal of Philosophy 93 (3): 491–508.

Griffith, Aaron (2018) “Social Construction and Grounding.” Philosophy and Phenomenological Research 97 (2): 393–409.

Griffith, Aaron (2020) “Realizing Race.” Philosophical Studies 177 (7): 1919–1934.

Guala, Francesco (2016) Understanding Institutions: The Science and Philosophy of Living Together. Princeton University Press.

Hacking, Ian (2000) The Social Construction of What? Harvard University Press.

Haslanger, Sally (2000) “Gender and Race: (What) Are They? (What) Do We Want Them To Be?” Noûs 34 (1): 31–55.

Haslanger, Sally (2012) Resisting Reality: Social Construction and Social Critique. Oxford University Press.

Hirsch, Eli (2005) “Physical-Object Ontology, Verbal Disputes, and Commonsense.” Philosophy and Phenomenological Research 7 0(1): 67–97.

Jenkins, Carrie (2005) “Realism and Independence.” American Philosophical Quarterly 42 (3): 199–209.

Jenkins, Carrie (2010) “What is Ontological Realism?” Philosophy Compass 5 (10): 880–890.

Jenkins, Katherine (2016) “Amelioration and Inclusion: Gender Identity and the Concept of Woman.” Ethics 126 (2): 394–421.

Khalidi, Muhammad Ali (2013) Natural Categories and Human Kinds: Classification in the Natural and Social Sciences. Cambridge University Press.

Khalidi, Muhammad Ali (2015) “Three Kinds of Social Kinds.” Philosophy and Phenomenological Research 90 (1): 96–112.

Kim, Jaegwon (1988) “Explanatory Realism, Causal Realism, and Explanatory Exclusion.” Midwest Studies in Philosophy 12 (1): 225–239.

Kim, Jaegwon (1993) Supervenience and Mind. Cambridge: Cambridge University Press.

Kim, Jaegwon (1994) “Explanatory knowledge and metaphysical dependence.” Philosophical Issues 5: 51–69.

Kovacs, David (2017) “Grounding and the Argument from Explanatoriness.” Philosophical Studies 174 (12): 2927–2952. 2017.

Lewis, David (1983) “New Work for a Theory of Universals.” Australasian Journal of Philosophy 61 (4): 343–377.

Mallon, Ron (2016) The Construction of Human Kinds. Oxford University Press.

Mason, Rebecca & Ritchie, Katherine (2020) “Social Ontology.” In Bliss, Ricki & Miller, J. T. M. (2020) The Routledge Handbook of Metametaphysics. Routledge.

Mason, Rebecca (2020) “Against Social Kind Anti-Realism.” Metaphysics 3 (1): 55–67.

Mason, Rebecca (2021) “Social Kinds are Essentially Mind-Dependent.” Philosophical Studies 178: 3975–3994.

McKenzie, Kerry (2020) “Science-Guided Metaphysics.” In Bliss, Ricki & Miller, J. T. M. (2020) The Routledge Handbook of Metametaphysics. Routledge.

Mikkola, Mari (2017) “On the Apparent Antagonism between Feminist and Mainstream Metaphysics.” Philosophical Studies 174 (10): 2435–2448.

Mills, Charles (2000) “‘But What Are You Really?’: The Metaphysics of Race.” Radical Philosophy Today 1: 23–51.

Moore, Adrian W. (2012) The Evolution of Modern Metaphysics: Making Sense of Things. Cambridge University Press.

Ney, Alyssa (2016) “Grounding in the Philosophy of Mind: A Defense.” In Kenneth, Aizawa & Carl, Gillett (2016) Scientific Composition and Metaphysical Ground. Palgrave-Macmillan.

Passinsky, Asya (2020) “Social Objects, Response-Dependence, and Realism.” Journal of the American Philosophical Association 6 (4): 431–443.

Passinsky, Asya (2021) “Finean Feminist Metaphysics.” Inquiry: An Interdisciplinary Journal of Philosophy 64 (9): 937–954.

Paul, Laurie (2012) “Metaphysics as Modeling: The Handmaiden’s Tale.” Philosophical Studies 160: 1–29.

Railton, Peter (1986) “Moral Realism.” Philosophical Review 95 (2): 163–207.

Ritchie, Katherine (2013) “What Are Groups?” Philosophical Studies 166 (2): 257–272.

Ritchie, Katherine (2020) “Social Structures and the Ontology of Social Groups.” Philosophy and Phenomenological Research 100 (2): 402–424.

Ritchie, Katherine (2021) “Does Identity Politics Reinforce Oppression?” Philosophers’ Imprint 21 (4): 1–15.

Rosen, Gideon (1994) “Objectivity and Modern Idealism: What is the Question?” in O’Leary Hawthorne, John & Michael, Michaelis (1994) Philosophy in Mind (Philosophical Studies Series 60, vol. 60). Springer.

Rosen, Gideon (2010) “Metaphysical Dependence: Grounding and Reduction.” In Hale, Bob & Hoffmann, Aviv (2010) Modality: Metaphysics, Logic, and Epistemology. Oxford University Press.

Ruben, David-Hillel (1990) Explaining Explanation. Abingdon-on-Thames: Routledge.

Schaffer, Jonathan (2009) “On What Grounds What.” In Chalmers, David, Manley, David, & Wasserman, Ryan (2009) Metametaphysics: New Essays on the Foundations of Ontology. Oxford University Press.

Schaffer, Jonathan (2017a) “Social Construction as Grounding; Or: Fundamentality for Feminists, a Reply to Barnes and Mikkola.” Philosophical Studies 174 (10): 2449–2465.

Schaffer, Jonathan (2017b) “The Ground Between the Gaps.” Philosophers’ Imprint 17 (11):1–26.

Sider, Theodore (2009) “Ontological Realism.” In Chalmers, David, Manley, David, & Wasserman, Ryan (2009) Metametaphysics: New Essays on the Foundations of Ontology. Oxford University Press.

Sider, Theodore (2011) Writing the Book of the World. Oxford University Press.

Sider, Theodore (2017) “Substantivity in Feminist Metaphysics.” Philosophical Studies 174 (10): 2467–2478.

Swoyer, Chris (1999) “How Ontology Might be Possible: Explanation and Inference in Metaphysics.” Midwest Studies in Philosophy 23 (1): 100–131.

Taylor, Elanor (2016) “Groups and Oppression.” Hypatia 31 (3): 520–536.

Taylor, Elanor (2018) “Against Explanatory Realism.” Philosophical Studies 175 (1): 197–219.

Taylor, Elanor (2020) “Social Categories in Context.” Journal of the American Philosophical Association 6 (2): 171–187.

Thomasson, Amie (2003) “Realism and Human Kinds.” Philosophy and Phenomenological Research 67 (3): 580–609.

Thomasson, Amie (2015) Ontology Made Easy. Oxford University Press.

Williamson, Timothy (2013) Modal Logic as Metaphysics. Oxford University Press.

Young, Iris Marion (1988) “Five Faces of Oppression.” Philosophical Forum 19 (4): 270–290.