1. What is an oppressive social practice?

To introduce the basic notion of a social practice, I begin with an example. C. J. Pascoe introduces the concept of the “fag discourse” in her ethnographic study of the way high-school boys reproduce dominant ideas of masculinity (Pascoe Reference Pascoe2012). The concept refers to the many interactions between the boys in which the titular slur was used as an insult or in service of “humorous imitation” (Pascoe Reference Pascoe2012, 60).Footnote ⁶ While one might assume its pervasiveness was due to widespread homophobia, Pascoe argues, this cannot explain why the slur was used to target non-sexual behavior or why it was not applied to lesbians or openly gay boys who conformed to stereotypical masculinity. To grasp the usage and meaning of the slur, it must be understood primarily as “a symbol around which contests for masculinity took place” (Pascoe Reference Pascoe2012, 82). The boys used the slur to stigmatize what Pascoe describes as “failures of masculinity” or “fag behavior,” a social category that included shows of same-sex affection, walking in an “effeminate” way, showing concern about style or cleanliness, or displaying even minute forms of incompetence. Behavior that fell into this category was seen to violate a specific ideal of masculinity: one that prized self-control, the exertion of sexualized dominance over others, and the concealment of perceived weakness.

The slur had the power to symbolically transform a boy into “the fag,” a figure who appeared in the collective imagination as “simultaneously predatory and passive and … at all costs, to be avoided” (Pascoe Reference Pascoe2012, 53). The “fag” was a “fluid” identity all sought to avoid, and one “most boys could escape, usually by engaging in some sort of discursive contest to turn another boy into a fag … One off the best ways to move out of the fag position was to thrust another boy into that position” (Pascoe Reference Pascoe2012, 61). Hence the “fag discourse” was like a game of “hot potato” (Pascoe Reference Pascoe2012, 60): the boys became hyper-vigilant about ensuring that they exhibited no signs of “fag behavior” and, as deflection was an advantage, hyper-alert to others’ lapses. Moreover, because what counted as “fag behavior” was not made explicit and could be expanded with novel uses of the slur, the boys could never be sure they would not be targeted. Consequently, the practice created an anxiety-ridden environment that pressured boys to conform to a stereotypical masculine ideal and punished those who failed.

The example of the “fag discourse” illustrates basic features of a social practice. First, as is well-known, practices are made up of rules that are constitutive. Constitutive rules do not regulate an activity that could be carried out in their absence, but establish the conditions under which a behavior counts as a specific kind of action, thereby enabling a practice to be carried out in the first place. Classic examples include rules of baseball and chess, but those unspoken rules that governed the “fag discourse” may also be understood in this vein.Footnote ⁷ Now, the boys did not necessarily recognize this as an activity with rules: one boy reported to Pascoe that one could be called the slur for “‘anything … literally anything’” (Pascoe Reference Pascoe2012, 57), which suggests he perceived the activity as arbitrary. The myriad rules that were in effect, however, established the conditions under which certain behaviors would count as “fag behavior.” Insofar as the point of the practice was to identify “real masculinity” and distinguish it from “failed masculinity,” the rules that rendered these categories meaningful also made it possible for the practice to be carried out in the first place. The slur was, then, not just a tool with which the boys condemned “fag behavior,” but one with which they reproduced this as an intelligible and significant social category.

Practices are constituted by “resources.” William Sewell, drawing on Giddens, defines these as anything an agent can use to exercise power and specifies that they can fall into two camps (Sewell Reference Sewell1992, 9). “Nonhuman” resources include “animinate or inanimate” (9) objects, while “human” resources include capacities like physical strength, knowledge, or emotional skills that can be used to enhance one’s power. I shall return to the role of material resources shortly. At this point, however, it is sufficient to observe that resources shape how, and to whom, the rules of a practice apply. One can imagine, for instance, that a greater share of “human” resources, like physical strength or charisma, would have made boys in the critical case less vulnerable to the slur or more easily able to repudiate it. To fully grasp how a practice structures a given social context, the practice must be understood as composed of rules and resources, as rules shape the resources in a practice and vice versa.Footnote ⁸

What makes a practice oppressive? Joining this sketch of practices to the initial definition of oppression, one can say that social practices are oppressive when their rules and resources undermine the conditions agents need to engage in self-development and self-determination. Consider, for example, a recent civil rights investigation that sought to uncover whether a Mississippi Sheriff’s Office had engaged in a “pattern or practice of ‘systematically violat[ing] people’s constitutional rights through excessive use of force; unlawful stops, searches, and arrests; and discriminatory policing’” (Kim Reference Kim2024, my emphasis). Oppression, in this case, consisted in wrongs like violence, degradation, and the violation of constitutional rights to which law enforcement subjected black citizens. What made the practice of policing oppressive, however, was the fact that its rules and resources—its structure—created conditions that made these wrongs a likely outcome, systematically undermining the ability of black citizens to exercise self-development and self-determination. The relevant rules and resources may have included ideas about what counted as good policing or who counted as a criminal, or the increased militarization of American police departments (Balko Reference Balko2013).

This approach highlights the way a practice’s structure incentivizes agents to act, such that oppression does not appear as a result of “a few bad apples” but of rules and resources that can recur across otherwise distinct practices. Indeed, if one can show similarities between the structure of policing and that of other practices, one can articulate the systemic character of racism. Moreover, by focusing on the structure of practices one can bring out the oppressive dimension of practices that may not appear to be oppressive in a clear-cut way. The “fag discourse,” for example, involves some ambiguity insofar as it blurs the line between perpetrators and victims: it’s true that some boys were targeted more intensively than others, but all were conscripted into using the slur and were its potential targets. Considering the connection between its rules and resources and conditions for self-development, however, reveals a sense in which it counts as an oppressive social practice. The practice as a whole defined “real masculinity” in terms of heterosexuality, proof of which was demonstrated through sexualized aggression. Hence, the practice created conditions that undermined the ability of gay male students and female students to gain the respect they needed to exercise self-development, unhindered. The practice meant that gay male students were accepted on the condition of social invisibility, and that female students were frequently subject to sexual harassment.Footnote ⁹

2. Schematic materiality and internalization

With this brief sketch of an oppressive social practice in view, I turn now to the way such an account can capture the dynamics of structural oppression. Sally Haslanger’s influential account of social practices is useful in this regard as it is explicitly oriented toward the question of why oppression persists or changes. This section shows the limits of understanding the critical case from §1 in terms of one of her proposals, on which oppressive meanings are instantiated in the material world, and the limits of a second proposal, in which individuals internalize oppressive meanings. The shortcomings of these proposals will motivate the turn, in §3, to intersubjective meanings as a necessary component of a practice-based social ontology that sheds light on oppression.

For Haslanger, social practices are composed of “schemas,” or collective meanings and norms, and “resources,” or material elements like objects, the built environment, and power. Social practices also set up the “nodes,” or social roles and positions, agents occupy, as well as relationships between them. By Haslanger’s lights, the concept of a practice furnishes a robustly social account that can explain how it is possible for oppression to “push back” (Haslanger Reference Haslanger2012, 415) on agents—that is, compel agents to act in ways that might conflict with their own beliefs, attitudes, and desires. This, in turn, promises to explain oppression’s “sticky” character, or the way it can persist despite the beliefs, attitudes, or desires of those who uphold it. Philosophically, the task is as follows: (1) to develop an account of oppression that counts as robustly social insofar as it invokes something other than mental states of individuals as part of oppression’s socio-ontological foundations; (2) to show, on the basis of this account, how oppression can compel agents to act in ways that come apart from their own beliefs, attitudes, or desires; (3) and so, to explain why it is that social transformation cannot be solely a matter of changing the minds of individuals.

Haslanger’s first proposal invokes “schematic materiality” (Haslanger Reference Haslanger2012, 417), or the interdependence of schemas and resources in practices. One can think, for example, of the way that wheelchair ramps in a city are shaped by schemas about who is thought to belong in a community, while also reinforcing those schemas. What makes this a robustly social account of oppressive meanings is that one has to point to the material world itself, not just the mental states of individuals, in order to capture the socio-ontological foundations of those meanings. Consequently it sheds light on the way schemas “push back” on agents in virtue of their instantiation in the material world. I may, for example, be adamantly opposed to car-centric infrastructure and the individualistic ethos to which it gives rise, and yet end up sustaining this ethos if the only way I can get to work is on the highway. The concept of schematic materiality clarifies, then, why social transformation requires a change to material dimensions of the social world. It is clear that, in order to undermine an individualistic ethos, one cannot change individual attitudes alone, but must also take aim at the built environment in which such an ethos is embodied.

But not all oppressive meanings are mediated through the material world in this way, as Haslanger observes. What description can we give of these other cases that meets the philosophical challenge outlined above? Grappling with this question, Haslanger initially proposes that oppressive schemas can be “internalized” and “guide behavior” (Haslanger Reference Haslanger2016, 126). And yet, she recognizes, this does not meet the desideratum that the account be robustly social. As she writes, if “schemas are internalized and govern action through individual psychology … again we seem to be left with familiar individualistic explanation” (Haslanger Reference Haslanger2016, 127). In response, she suggests, psychological facts can “trigger” action but structural constraints “set limits, organize thought and communication, create a choice architecture; in short, they structure the possibility space for agency” (127). More broadly, while agents do internalize oppressive meanings, invoking such internalized states alone conceals other, structural factors that shape what agents do. It may be true, for example, that women take on an unequal share of domestic labor in virtue of internalized gender norms, but this alone is insufficient as an explanation as it occludes the role of factors like workplace and governmental policies.

Against this conceptual backdrop, how are we to make sense of the critical case from §1? On the one hand, schematic materiality is not entirely apt. For although the oppressive meanings in the practice were surely embodied in some material objects—the kind of clothing marketed to boys for instance—this was not the primary form these meanings took. The two even came into conflict: changing rooms at the school went unused because changing clothes after shop class was deemed “fag behavior.” On the other hand, an appeal to internalized meanings and norms is not entirely satisfactory either. This may seem to be a more apt account, as the “fag discourse” did not involve paradigmatically structural constraints of, say, an economic or political character. But even in this case, I shall argue, appealing to internalization obscures more than it illuminates in ways that highlight the need for an intersubjective approach.

In order to bring this into view, let us work backwards from Haslanger’s claim that internalization accounts conceal the social factors that “structure the possibility space for agency.” I would argue that the relevant factors in the critical case are interpersonal interactions. Consider the frequency with which the boys lobbed the slur at one another, the punishing attention they paid to even trivial “transgressions,” and the occasional confusion the boys betrayed about what, exactly, this practice demanded of them.Footnote ¹⁰ These features of the practice suggest that there was a need for the “rules of gender” to be perpetually reasserted and policed. Although the boys surely had internalized ideas about what it meant to be a man, taking this to be a sufficient explanation of their participation conceals the central role such interpersonal interactions played. Moreover, it shapes how one perceives the boys’ activity. If one understands the boys’ activity to have been primarily driven by an internal sense that certain behaviors were shameful or unmanly, it is hard to bring this activity’s “other-directed” character into view. By “other-directed,” I mean the way this activity was driven by desires to be seen or not seen in a certain light by others and, more broadly, a heightened consciousness of the way one appeared in the eyes of others.

Of course, focusing on internalized meanings and norms, on the one hand, and interpersonal interactions, on the other hand, are not mutually opposed. But, as a practice approach aims to highlight the social processes through which oppression is reproduced, it is especially important to bring out the inner dynamics of the processes by which the boys internalize dominant ideas about masculinity, which here crucially involve interpersonal interactions. Such internalization may be an outcome of these social processes; our theoretical frameworks, however, should give us the tools to talk about these processes themselves and not just their end result.Footnote ¹¹

If this is the case, one might wonder why internalization-talk is common in explanations of oppression’s persistence? Revealing here is Haslanger’s remark that if a set of oppressive meanings is to “push back” on agents, there must exist “an actualization of it in the world … that involves something material” (Haslanger Reference Haslanger2012, 415). One might imagine, for instance, that ideas about what it is to be a man can exist in the social world only by being embodied in something material. As a corollary, it may seem that the only other place such ideas can be “located” is “embodied in individuals” (415): not external, but internal to individuals. One point of a practice-based social ontology, however, is to challenge this conceptual topography and the dualism on which it rests, and to make space for a third possibility, on which meanings emerge out of relationships between agents. One lesson to draw from Haslanger’s critique of internalization is that we need to think about the social ontology of oppressive meanings in a way that is neither just material nor just in the minds of individual agents, but in a way that captures their robustly intersubjective character. The subsequent section develops such an account and show how it responds to the challenge laid out at the start of §2.

3. An intersubjective account of oppressive meanings

In this section, I introduce the concept of intersubjective meanings and show how it applies to oppression in a way that meets the three desiderata laid out at the start of §2. I conclude with some remarks about its implications for social change.Footnote ¹²

The concept of intersubjective meanings is most closely associated with Taylor’s discussion of social practices, where he defines intersubjective meanings as a “vision of the agent and his relation to others and to society” (Taylor Reference Taylor1985, 35) that is “not just in the minds of the actors but … out there in the practices themselves, practices which cannot be conceived as a set of individual actions, but which are essentially modes of social relation, of mutual action” (Taylor Reference Taylor1985, 36). In what sense are intersubjective meanings “out there in the practices themselves”? As a first step, consider the way intersubjective meanings are instantiated in the rules of a practice as a whole: a vision of democratic citizenship that prizes individual autonomy is not just expressed through one rule but through a whole host of rules that, taken together, establish how agents must relate to others to successfully perform the practice. Voters must, for instance, cast private ballots, and campaigners must remain at a distance from the voting site. The relationships these rules establish express, and indeed actualize, the idea that autonomy is a key component of democratic citizenship.

Intersubjective meanings are also grounded in modes of mutual action: an action, like teaching, whose successful performance requires multiple agents to play distinct roles. Such meanings necessarily require, therefore, more than one agent to sustain them. A society that highly prizes civic responsibility in its conception of democratic citizenship may implement a public defense in which voters must articulate reasons for their vote before a group of fellow citizens to successfully cast a ballot.Footnote ¹³ Successfully casting a vote, in this case, requires two parties to play their respective roles: whether articulating reasons or attending to the reasons given by others. The public defense also establishes a specific kind of social relationship between its participants that differs from the one that exists in the case of the private ballot: in order to engage in the give-and-take of reasons it demands, I have to orient myself toward others not as if to a stranger or a friend but to a fellow citizen to whom I have a distinctive kind of responsibility.

Intersubjective meanings are robustly social insofar as they “do not bring together two free subjectivities, but two partners who must do different things and whose roles or statuses are determined by an established rule” (Descombes Reference Descombes2014, 301), as Descombes observes. While a group of agents might share the same set of beliefs about what, say, democratic citizenship means, this alone does not give this vision social reality.Footnote ¹⁴ Intersubjective meanings are robustly social not in virtue of “a certain intellectual content … simultaneously present in the minds of individuals” (Descombes Reference Descombes1994, 104), but in virtue of the way they structure relationships between agents in a practice. Looking solely at the beliefs and attitudes of participants in a practice may not, therefore, permit the intersubjective meanings in that practice to come fully into view as these can come apart. Perhaps I participate in the practice of voting with all kinds of beliefs about the merits of different candidates or the political process itself and yet, to paraphrase Taylor, what I do not bring with me is the vision of democratic citizenship that underpins the social practice. This vision is, rather, embedded in how I must relate to others in order to cast my ballot.

It is in virtue of their robustly social character that intersubjective meanings can constrain agents. They do so in two interrelated ways. First, as is true of practices generally, the rules through which intersubjective meanings get expressed have an independent, “stage-setting” quality: they set out how one must relate to others in order to participate, they apply to anyone who wants to do so, and they cannot be changed by individuals at will. If I want to vote, I must participate in the public defense; to refuse to do so because “that’s not how I vote” would be both nonsensical and result in a failure to vote at all. Second, because intersubjective meanings are grounded in forms of mutual action, we need others’ cooperation in order to uphold them; conversely, others may conscript us into upholding intersubjective meanings. If I want to actualize the idea that education consists in the unhindered pursuit of curiosity but students treat me as a “sage on the stage” and themselves as passive consumers, my vision of education will be frustrated. I may find myself upholding a set of meanings that I do not personally endorse. Agents actively uphold intersubjective meanings, but they do not do so as “free subjectivities” insofar as their capacity to do so depends on established relationships to others.

This concept of intersubjective meanings better illuminates the robustly social character of oppressive meanings.Footnote ¹⁵ Consider what it means to be a “real man” in the “fag discourse.” On this account, we can see this idea as grounded in the myriad rules of the practice as a whole, rules that govern how one should move, dress, show affection, enjoy, and so forth. These rules establish how the boys must relate to others—the forms of mutual action in which they must engage—to achieve the esteemed status of a “real man.” To achieve this esteemed status or to avoid the “fag” status, they must construct and come to stand in a hierarchical relationship to another boy who is devalued. Multiple agents have to “play their part” in a way that does not involve others’ cooperation as much as it does their forceful conscription into the “fag” status. Just as the public defense aims as the successful performance of voting, this mutual action aims, in some sense, at the successful performance of masculinity as well as at the achievement of status, esteem, and recognition, and the avoidance of social devaluation.

Agents uphold oppressive meanings through their relationships to others, the forms of which are not entirely up to them as individuals. Understanding oppressive meanings as intersubjective allows us to see one sense in which such meanings acquire an independent social reality, which sheds light in turn on how even agents who disagree with oppressive meanings may be complicit in their perpetuation. Consider, as an example of this, that despite the boys’ enthusiastic participation in the “fag discourse,” many disavowed homophobia when Pascoe asked them about their own beliefs and attitudes. One might wish to attribute this to a desire to avoid the appearance of bigotry, but the possible conflict between oppressive social meanings and the beliefs, attitudes, and intentions of individuals is a broader phenomenon, as Young’s description of structural oppression as grounded in the “everyday practices of a well-intentioned liberal society” (Young Reference Young1990, 41, emphasis my own) indicates.

If the oppressive meanings in the practice are grounded in hierarchical relationships, however, it is more evident how agents can be motivated to uphold them, regardless of their individual beliefs, attitudes, and intentions. Oppressive meanings in the “fag discourse” establish how the boys must relate to others in order to gain status, esteem, and recognition, and so they have a powerful motivation to participate: to achieve an esteemed status or avoid the painful experience of being degraded. Regardless of whether they object or remain indifferent to the ideas the practice expresses, it is hard to resist being drawn into it or to opt out. For one thing, because one had to stand in the right relationship to a devalued other in order to achieve the esteemed status, any boy was likely to find himself conscripted as an unwilling partner into someone else’s bid for status at some point. For another, because the rules governing these relationships had a public and independent character, he had to redirect the slur or risk his standing in the eyes of others. He might have objected to the practice and what it expresses, but there were no socially legible ways to challenge it that would not simply appear to be an acceptance of the devalued status—and so failure by the lights of the practice. It is not that individuals can never resist oppressive meanings; indeed, Pascoe observes some of the ways male students did transgress the ideas about masculinity embedded in the “fag discourse.”Footnote ¹⁶ Without recognizing how oppressive meanings can be embedded in intersubjective relationships, however, we will struggle to see the social obstacles that make it difficult for even those opposed to oppressive meanings to resist them.

The philosophical task outlined at the start of §2 had three desiderata. We sought, first, a robustly social account of oppression that would chart a path between prioritizing material aspects of the world and internalized states and, second, to explain how oppression “pushes back” on agents. On an intersubjective approach, oppressive meanings are instantiated in the rules that establish how agents must relate to others to perform a practice properly or, more specifically, to acquire the status, recognition, and esteem the practice makes possible. Because of the independent character of these rules, and because oppressive meanings depend on forms of mutual action, agents can be conscripted into perpetuating oppressive meanings even if they disagree with them. But what does this approach suggest about the third and final task—that of bringing out the possibilities for social transformation? While it is beyond the scope of this paper to address this task in full, I shall draw out two implications by way of conclusion. The first is quite general. The above account suggests that social theorists should devote greater attention to the way practices structure relationships between agents, especially relationships through which agents seek social goods of recognition, esteem, and status. The second implication concerns the difficulty of social change. When oppressive meanings structure recognitive relationships in this way, agents may be shaped by these meanings in ways that come apart from the recognition they are consciously seeking. Even if the boys were not aware of the wider meanings in the practice, their pursuit of esteem acculturates them into these meanings. For they do not need to comprehend the practice’s underlying principles to participate, but just have to “do the right thing in the right context” where this means observing how others use the slur, attempting to use it correctly themselves, and assessing their relative success or failure by gauging the reactions of others.Footnote ¹⁷ Similarly, a boy who is labelled a “fag” does not need to grasp what his “failure” consists in or why it counts as a failure to get that something about who he is or what he has done counts as “wrong.” It is through this process that the boys acquire a tacit and primarily non-discursive grasp of gender and sexuality that manifests in a sense of the shameful or admirable, or what is disgusting or arousing. By pursuing relational goods in oppressive contexts, agents can become acculturated into intersubjective meanings in ways that are hard to eradicate.

Social ontology shapes the diagnosis we give of what drives oppression and how it can be remedied. Many forms of oppression concern unequal relationships of power, esteem, or status. Yet even social ontologies that aim to capture the dynamics of oppression do not always foreground the relational sphere. As a result, it becomes difficult to theoretically articulate central drivers of oppression, or how it can be addressed. My aim in this paper has been to show that practice-based social ontologies can better explain some forms of structural oppression by appealing to the concept of intersubjective meanings. Understanding how oppressive meanings structure relationships permits a clearer vision of the social barriers that exist to changing an oppressive social world.