Jara-Ettinger and Dunham argue that the institutional stance constitutes a distinct cognitive system for interpreting social behavior through role-based structures. Their three-part formalization of role content – tasks, socially granted powers, and limitations – is a genuine advance, specifying behavioral boundaries around role performance. All three occupy the same conceptual space: they delineate the behavioral requirements and permissions that attach to a role. Tasks specify what must be done, whereas powers and limitations define what one can and cannot do. And while a power described with its boundary conditions is itself a limitation (and vice versa), they may be pragmatically useful as separate categories.
This is what makes one of their own central examples so conspicuous. Jara-Ettinger and Dunham classify as a limitation the requirement that the barista “must be friendly” (Section 4.1). The requirement is real, yet it is not a behavioral boundary described from any direction. It cannot be cashed out as a discrete action or sequence of actions. It specifies, instead, an orientation: a quality of engagement with which the role’s tasks are to be carried out. One might respond, however, that “must be friendly” just is a constraint on how tasks are performed and therefore a proper limitation whose violation constitutes role failure: a barista can observe every behavioral boundary while being surly or be warm and accommodating while making procedural errors. These are indeed familiar patterns of role failure, evaluable from within the institutional stance without mentalizing (that is, without positing beliefs or desires to explain the failure), yet the institutional stance’s formalization offers no way to distinguish them. I call the norms of orientation dispositional entailments, extending the sense in which normative entailments of roles are already discussed in this literature – namely, as expectations. Recognizing them as a distinct type of role content enriches the institutional stance without collapsing it into the mentalistic stance.
This matters because many roles central to social life – parent, teacher, leader, judge, host – have dispositional entailments as core content. A parent who feeds, clothes, and shelters a child but does so in a manner marked by indifference or contempt is failing in the role, not failing on some separate dimension that only mentalizing could detect. The failure is, to borrow their language, institutional: the role of the parent specifies that care should be enacted with a certain orientation. The current formalization of role content does not capture this difference – but an expanded formalization could. Indeed, the ease of identifying roles without dispositional entailments (compared to the difficulty of finding roles without limitations) suggests it may be a more natural third category.
One might pause: aren’t dispositional entailments just requirements to hold certain mental states, as Jara-Ettinger and Dunham discuss in Section 6.3.3? The examples they provide, though, involve propositional attitudes: a stock person must know where items are located, and a role-occupant may be expected to want to fulfill their role’s goals. Dispositional entailments are different. “Be caring” does not specify that the role-occupant must believe P or desire Q. Any attempt to decompose it into a propositional attitude either misses the target (“must believe caring is important” is compatible with cold, perfunctory conduct) or covertly reintroduces the dispositional content (“must want to engage warmly” simply restates an orientation requirement in mentalistic vocabulary). Dispositional entailments specify a quality of relational engagement, not an inference about an individual’s psychology.
A community that expects its parents to exhibit care is not thereby predicting what individual parents think or feel; it is specifying a condition of adequate role performance. And like other role content, dispositional entailments are socially transmitted and publicly assessable. They are learned as part of what the role demands, and failures to meet them are recognized and evaluated by others within the institutional framework itself.
The dispositional entailments of roles become especially visible in normative traditions – such as classical Confucianism – that have theorized them explicitly. Confucian ethics are organizes around roles specifying not only what the role-occupant must do but also the orientation with which they must do it (King Reference King2025; Sarkissian Reference Sarkissian, Szabó Gendler, Hawthorne and Chung2023, Reference Sarkissian2025). A passage from the canonical Mengzi (4B28) illustrates the point. When a role-occupant is treated with unexpected hostility by an institutional counterpart, the prescribed response is not to infer the offender’s mental states, much less to examine their own. Instead, they should ask whether their scrutable role conduct may have prompted it somehow (perhaps inadvertently). Only when this kind of self-examination confirms full compliance with dispositional norms (tenderheartedness, decorum, and commitment) is it appropriate to reach a judgment about the other. What’s more, the judgment is institutional, not mentalistic: this person is no longer within the space of role-governed interaction – they are simply “lost.” The entire sequence operates within the institutional stance: the role-occupant evaluates their own performance against institutional standards and from the perspective of the counterpart. The line of interrogation is “What did I do?” (role assessment), not “What was in my mind?” (introspection).
Compare this to Jara-Ettinger and Dunham’s account of competition between the stances (Section 6.3.2). They propose that explanatory or predictive failure in one stance triggers recruitment of the other. Their examples include both discrete task failures (the coffee doesn’t arrive) and failures of manner (a barista-friend acting unusually formal). Yet, their formalization accommodates only the first kind. Once dispositional entailments are recognized as distinct role content, the institutional stance gains evaluative resources that extend beyond compliance with discrete role requirements.
A frivolous judge, an inattentive host, a callous parent – these are recognizable instances of role failures assessed against dispositional standards without shifting to inferences about individual beliefs and desires. The failures can be read from the overt, scrupulous behavior of the role occupant (their tone, manner, and demeanor). The standard itself, however, cannot be reduced to any enumerable set of behavioral requirements. The institutional stance thus does more evaluative work than Jara-Ettinger and Dunham’s formalization, with its emphasis on prediction, currently allows. But that only becomes visible once we recognize dispositional entailments as part of what roles specify – something their barista who “must be friendly” already suggests.
You have
Access
Jara-Ettinger and Dunham argue that the institutional stance constitutes a distinct cognitive system for interpreting social behavior through role-based structures. Their three-part formalization of role content – tasks, socially granted powers, and limitations – is a genuine advance, specifying behavioral boundaries around role performance. All three occupy the same conceptual space: they delineate the behavioral requirements and permissions that attach to a role. Tasks specify what must be done, whereas powers and limitations define what one can and cannot do. And while a power described with its boundary conditions is itself a limitation (and vice versa), they may be pragmatically useful as separate categories.
This is what makes one of their own central examples so conspicuous. Jara-Ettinger and Dunham classify as a limitation the requirement that the barista “must be friendly” (Section 4.1). The requirement is real, yet it is not a behavioral boundary described from any direction. It cannot be cashed out as a discrete action or sequence of actions. It specifies, instead, an orientation: a quality of engagement with which the role’s tasks are to be carried out. One might respond, however, that “must be friendly” just is a constraint on how tasks are performed and therefore a proper limitation whose violation constitutes role failure: a barista can observe every behavioral boundary while being surly or be warm and accommodating while making procedural errors. These are indeed familiar patterns of role failure, evaluable from within the institutional stance without mentalizing (that is, without positing beliefs or desires to explain the failure), yet the institutional stance’s formalization offers no way to distinguish them. I call the norms of orientation dispositional entailments, extending the sense in which normative entailments of roles are already discussed in this literature – namely, as expectations. Recognizing them as a distinct type of role content enriches the institutional stance without collapsing it into the mentalistic stance.
This matters because many roles central to social life – parent, teacher, leader, judge, host – have dispositional entailments as core content. A parent who feeds, clothes, and shelters a child but does so in a manner marked by indifference or contempt is failing in the role, not failing on some separate dimension that only mentalizing could detect. The failure is, to borrow their language, institutional: the role of the parent specifies that care should be enacted with a certain orientation. The current formalization of role content does not capture this difference – but an expanded formalization could. Indeed, the ease of identifying roles without dispositional entailments (compared to the difficulty of finding roles without limitations) suggests it may be a more natural third category.
One might pause: aren’t dispositional entailments just requirements to hold certain mental states, as Jara-Ettinger and Dunham discuss in Section 6.3.3? The examples they provide, though, involve propositional attitudes: a stock person must know where items are located, and a role-occupant may be expected to want to fulfill their role’s goals. Dispositional entailments are different. “Be caring” does not specify that the role-occupant must believe P or desire Q. Any attempt to decompose it into a propositional attitude either misses the target (“must believe caring is important” is compatible with cold, perfunctory conduct) or covertly reintroduces the dispositional content (“must want to engage warmly” simply restates an orientation requirement in mentalistic vocabulary). Dispositional entailments specify a quality of relational engagement, not an inference about an individual’s psychology.
A community that expects its parents to exhibit care is not thereby predicting what individual parents think or feel; it is specifying a condition of adequate role performance. And like other role content, dispositional entailments are socially transmitted and publicly assessable. They are learned as part of what the role demands, and failures to meet them are recognized and evaluated by others within the institutional framework itself.
The dispositional entailments of roles become especially visible in normative traditions – such as classical Confucianism – that have theorized them explicitly. Confucian ethics are organizes around roles specifying not only what the role-occupant must do but also the orientation with which they must do it (King Reference King2025; Sarkissian Reference Sarkissian, Szabó Gendler, Hawthorne and Chung2023, Reference Sarkissian2025). A passage from the canonical Mengzi (4B28) illustrates the point. When a role-occupant is treated with unexpected hostility by an institutional counterpart, the prescribed response is not to infer the offender’s mental states, much less to examine their own. Instead, they should ask whether their scrutable role conduct may have prompted it somehow (perhaps inadvertently). Only when this kind of self-examination confirms full compliance with dispositional norms (tenderheartedness, decorum, and commitment) is it appropriate to reach a judgment about the other. What’s more, the judgment is institutional, not mentalistic: this person is no longer within the space of role-governed interaction – they are simply “lost.” The entire sequence operates within the institutional stance: the role-occupant evaluates their own performance against institutional standards and from the perspective of the counterpart. The line of interrogation is “What did I do?” (role assessment), not “What was in my mind?” (introspection).
Compare this to Jara-Ettinger and Dunham’s account of competition between the stances (Section 6.3.2). They propose that explanatory or predictive failure in one stance triggers recruitment of the other. Their examples include both discrete task failures (the coffee doesn’t arrive) and failures of manner (a barista-friend acting unusually formal). Yet, their formalization accommodates only the first kind. Once dispositional entailments are recognized as distinct role content, the institutional stance gains evaluative resources that extend beyond compliance with discrete role requirements.
A frivolous judge, an inattentive host, a callous parent – these are recognizable instances of role failures assessed against dispositional standards without shifting to inferences about individual beliefs and desires. The failures can be read from the overt, scrupulous behavior of the role occupant (their tone, manner, and demeanor). The standard itself, however, cannot be reduced to any enumerable set of behavioral requirements. The institutional stance thus does more evaluative work than Jara-Ettinger and Dunham’s formalization, with its emphasis on prediction, currently allows. But that only becomes visible once we recognize dispositional entailments as part of what roles specify – something their barista who “must be friendly” already suggests.
Financial support
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Competing interests
None.