The Metaphysics of Social Powers: Introduction to the Special Collection

1. Dispositions and Social Powers: Where Are We Now?

Dispositions, or causal powers,¹ have been the subject of a flourishing literature since the 1970s. Moving away from attempts to analyze and reduce sentences containing dispositional predicates to those devoid of such predicates, the focus has shifted to dispositions themselves: dispositional properties understood as entities in the world and as proper objects of philosophical inquiry. Dispositional properties are standardly understood as causal properties, namely properties that generate certain effects when certain circumstances arise and that are grounded in an object’s physical structure.² Many metaphysicians have put dispositions to work in accounting for a wide range of phenomena, including causation (Mumford & Anjum 2011, Williams 2019), the identity conditions of properties (Shoemaker 1980), laws of nature (Bird 2005), persistence over time (Williams 2019). Dispositions have also been mobilized as explanans for modal statements, accounting for claims about possibility and necessity (Borghini & Williams 2008; Jacobs 2010; Vetter 2015). More broadly, Groff and Greco (2013) present realism about causal powers as part of a ‘Neo-Aristotelian’ framework characterized by a rejection of neo-Humeanism and by a picture of properties as genuinely powerful and interconnected, capable of bringing about change through their interactions with one another, rather than as passive qualities forming a mosaic on which laws of nature supervene (Lewis 1986).

In parallel, a power turn is emerging within social ontology: this is an attempt to sharpen existing theoretical frameworks by moving beyond collective intentionality and the assignment of status functions (Searle 1995), in order to capture the power dynamics that structure the social world. Social powers are often understood as a type of normative power, and they include deontic powers, such as the power of a president to determine that a country ought (or ought not) to go to war (Searle 2010), communal powers, such as the power of someone cool to decide who is going to sit at the table (Ásta 2018), and telic powers, such as the ability to effect certain outcomes in virtue of one’s correspondence to a social ideal (Burman 2023). Because their possession depends on collective practices, normative powers do not satisfy the intrinsicness condition often imposed on dispositions. Moreover, they are standardly understood to produce effects by giving agents reasons to act in certain ways—as when a person complies with an arrest because of the police officer’s authority—and they cannot easily be equated with causal powers’ ability to produce a change. Since social powers restrict our possibilities for action, and often do so in a counterfactually stable way, they also bear on questions of modality. While some (Searle 1995; Hansson Wahlberg 2020; Burman 2023) have insisted on the dissimilarities between normative powers and causal powers, others have attempted to unify the two (Lawson 2013). The issue of causal powers also raises a deeper ontological question: can certain social entities, understood as substances or structures, be said to have causal powers? For example, if one’s opportunities are constrained by a gendered structure, can that structure be understood as an emergent entity with causal powers of its own?

This is precisely the kind of question for which contemporary metaphysical tools associated with Neo-Aristotelian metaphysics should prove useful. Yet these resources have not been extensively brought to bear on the question whether social entities can have causal powers, with the exception of discussions around critical realism.³ This relative neglect may be explained by the tendency of social ontologists to focus on constitution, rather than causation, in the social realm. To clarify the notion of social constitution, concerning how an entity’s nature can depend on social factors, social ontologists have applied contemporary metaphysical tools such as grounding (e.g., Epstein 2015; Griffith 2018; Schaffer 2019) and essence to social entities (e.g., Mason 2020; Passinsky 2025), while leaving questions of social causation relatively underexplored.⁴ Yet this latter inquiry could fruitfully engage with causal powers theories. Just as dispositions have been usefully applied to the realm of the natural sciences and have been argued to accurately capture scientific discourse (e.g., Cartwright, 1989; Bird, 2007), one may wonder whether understanding social properties as dispositional could also lead to a better understanding of the social world. Moreover, causal powers theories may illuminate not only the familiar question of whether social entities have emergent causal powers, as well as the relation between those powers and scientific inquiry, but also questions concerning the existence of social entities themselves. Indeed, some social ontologists have moved away from treating mind-independence as the touchstone of reality (Thomasson 2003; Baker 2007; Khalidi 2015, 2016), and have instead appealed to the Eleatic Principle, the causal criterion for reality, according to which something is real if (and only if) it has causal powers (Baker 2007; Khalidi 2015, 2016; Morrison 2025). But this appeal to causal efficacy has not typically been developed through engagement with theories of causal powers and is rarely, if ever, accompanied by a clear anti-Humean and realist view of causation, without which the principle cannot be adequately put to use. Such theories must therefore help clarify what it would mean for a social entity to satisfy a causal criterion of reality: for social entities to count as real under the Eleatic Principle, must they be understood as objects that bear causal powers, as powerful properties, or merely as difference-makers whose causal efficacy can be captured counterfactually? Finally, as mentioned above, causal power theories have also been used to ground modal statements. This is especially relevant given that current discussions in social ontology now focus more on the modal features of social entities (e.g., Epstein 2015; Griffith 2018; Raven, 2024). One fruitful subject of discussion would then be to examine whether social powers also ground certain modal statements related to social entities just the way causal powers do.⁵

Against this background, this special collection brings metaphysical work on powers into closer dialogue with social ontology, seeking to advance discussion of several issues that, despite their importance, have received relatively limited attention. More specifically, the present collection is organized around three questions. Can social entities be understood as substances capable of bringing about change through their causal powers, thereby perhaps vindicating their existence? Despite the differences in their existence conditions and causal mechanisms, are normative powers and causal powers both properties of things directed toward the realization of a change, or is talk of normative powers merely predicative and ultimately reducible? What is the connection between social powers and modal claims?

2. The Contributions

The three articles of the collection address these questions in turn: Cox (2025) considers whether social entities can be substances bearing active powers within Aristotle’s metaphysics; Hansson Wahlberg (2025) addresses a puzzle concerning the apparent backward generation of institutional properties, thereby illuminating the relationship between institutional statuses, on the one hand, and social enablements and constraints, on the other; and Locke (2026) investigates how social powers structure a distinctively social space of possibility and necessity.

In his contribution ‘Aristotle’s Plural Power Problem’, Joel Cox identifies a tension internal to Aristotle’s metaphysical system. If only substances can bear what Cox calls ‘active powers’, and if social entities such as cities, families, or mobs are not substances, then such entities cannot exercise active powers. Cox argues, however, that Aristotle himself treats cities and families in strikingly substantive terms in the Politics. He places them on a sliding scale of unity that brings them closer to individuals, attributes to them a proper telos, compares them analogically to the cosmos and to living bodies, and ascribes to them forms of joint deliberation that appear to function as active powers irreducible to those of individual members. This tension gives rise to a dilemma. Either Aristotle is wrong to exclude social entities from the category of substance, or he is wrong to claim that only substances can bear active powers. In order to sharpen the terms of this dilemma, Cox draws on Gilbert (2014, 2015) and Jansen’s (2014, 2017) plural subject theory. On this view, plural subjects are unified through joint commitment and thereby constitute wholes capable of acting and causing as units that cannot be reduced to the aggregate behavior of their members. Cox argues that such entities cannot be adequately captured by reductionist models of social action. He concludes that the more plausible option for an Aristotelian is to accept that certain social entities are substances and therefore legitimate bearers of active powers. This would require extending Aristotle’s account beyond the cases considered in Categories 5 and Metaphysics VII, where social entities do not figure among his examples of substance, to include social entities characterized by joint action, a joint identity, and awareness of that identity among most of their members.

In a paper called ‘Is There Backward Generation in the Institutional Realm?’, Tobias Hansson Wahlberg addresses a puzzle internal to Searle’s (2010) framework of social ontology. Status Function Declarations play an important role in that framework: these are declarations whose successful performance brings into existence the institutional state of affairs they represent, and they are standardly understood as synchronic, in the sense that the relevant institutional status is created at the time of the declaration. Yet many actual declarations are directed toward the absolute past, as in retroactive grade assignments and job appointments or the disqualification of winners in sports years after the event, such as the 2012 decision to strip Lance Armstrong of all his results and prizes from August 1, 1998 onwards, because of doping violations. Negative retroactive declarations, such as disqualifications, are handled through what Hansson Wahlberg calls ‘modal mere Cambridge change.’ Assuming the B-theory of time, the later revoking declaration is already part of reality at the time of the initial declaration, which is therefore void rather than first valid and subsequently invalidated. Thus, the athlete was never the official winner, although everyone at the earlier time believed and behaved as if they were. No institutional property is therefore removed from the past.

An important consequence of this account concerns the relationship between institutional properties and social powers, constraints, or enablements. Hansson Wahlberg argues that these can systematically come apart. An entity may possess an institutional status without any corresponding enablement or constraint being realized, as in cases where a retroactively assigned grade has no practical consequences at the time to which it is deemed to apply. Conversely, enablements and constraints may be realized in the absence of any genuine institutional status, as when a disqualified athlete has already spent the prize money. If Hansson Wahlberg’s account is correct, it would seriously undermine theories that identify, or assume a perfect correlation between institutional statuses and enablements and constraints, such as Ásta’s (2018) account.

Social constraints and enablements are also central in Theodore Locke’s ‘Varieties of Social Necessity’. In this contribution, Locke argues that there exists a genuine kind of modality, social modality, which he presents as a fourth kind alongside the metaphysical, natural, and normative modalities distinguished by Fine (2002). In particular, Locke defends social necessities as real modalities. Following Rosen (2006), he characterizes such modalities as ‘alethic’, ‘non-epistemic’, and ‘objective’. This means, respectively, that claims expressing social necessities concern what is true rather than what is prescribed, that they do not concern what we are in a position to know, and that they concern the non-representational world rather than our representations of it. Locke’s argument draws on Fine’s (2002) parallel treatment of natural necessity: just as laws of nature may be necessary in a natural sense without being metaphysically necessary, Locke suggests that some claims about social reality may be necessary in a social sense without being metaphysically necessary. Actual social systems may make certain arrangements socially impossible even though they remain perfectly coherent from the standpoint of metaphysical possibility. Locke argues for this by appealing to conceivability. For instance, we can coherently imagine a possible world such as Barbieland, where women occupy dominant social and political positions. Such a world involves no metaphysical incoherence. Yet it may violate what is socially necessary relative to actual heteronormative structures, insofar as these structures generate patterns of constraint and enablement that make such an arrangement socially impossible, or at least not a nearby social possibility.

Locke further argues that social modality cannot be captured simply by relativizing metaphysical necessity to a set of basic social facts, since this would leave unexplained why those facts themselves constrain what is socially possible. The upshot is that the space of social possibility is best understood as underwritten by social systems and by the patterns of constraints and enablements they generate, rather than by metaphysical essences or laws alone. To put this in the terms of our discussion in Section 1, one could say that the space of social possibility depends on the distribution of social powers. For Locke, moreover, his account has normative stakes: insofar as social powers shape what is socially possible for individuals and groups in counterfactually stable and unjust ways, there is a form of modal injustice, a harm suffered not only in virtue of one’s social position, but in virtue of social possibilities that are closed off altogether.

3. Conclusion and Directions for Future Work

Taken together, the contributions to this collection show that a metaphysics of social powers raises related questions about the kinds of entities that can bear powers, the relationship between normative and causal powers, and the role of powers in structuring social possibility. At the same time, much remains to be explored, and we would like to indicate some directions for future work here. One important direction concerns the relationship between social powers and the causal criterion of reality: if causal efficacy is to vindicate the reality of mind-dependent social entities, it remains to be determined whether this requires genuine productive powers or whether some weaker form of causal relevance or difference-making is sufficient. A related direction concerns the nature of causal social construction. While non-causal social construction has frequently been accounted for in terms of grounding, it remains less clear what causal social construction consists in and how it relates to non-causal social construction: must it proceed by causally producing the grounds of a socially constructed fact (e.g., Griffith 2018), or can it take some other form? Another direction concerns whether and how social powers can play for social properties the individuating and modal roles that dispositions have been claimed to play elsewhere in metaphysics, for instance, by bearing on the identity, essence, or modal profiles of social properties. A further question concerns the range of possible bearers of social powers, including not only individuals, groups, and structures but also artifacts and other socially constituted entities. There is also more to be said about the mechanisms through which social powers constrain agents: for example, are the constraints associated with social powers always causal, or can social structures also constrain agents through non-causal, for instance constitutive, relations? These are among the directions worth pursuing in future work. More generally, these questions point toward the need for a more systematic metaphysics of social powers and their place within the broader metaphysics of the social world.

Author Contributions

Both authors contributed equally to the conception, writing, and revision of this manuscript.