Patriarchy as Institutional

1 What Counts as Institutional?

On what is arguably the most common understanding of institutions, they govern which moves that are open to whom and when. One prominent example is North (1991, p. 97): “Institutions are the humanly devised constraints that structure political, economic and social interaction. They consist of both informal constraints (sanctions, taboos, customs, traditions, and codes of conduct), and formal rules (constitutions, laws, property rights).” Another proponent is Ostrom (2005, p. 3): “[b]roadly defined, institutions are the prescriptions that humans use to organize all forms of repetitive and structured interactions.” And more recently, Ásta (2018, p. 106) has understood something having institutional status in terms of “rights, privileges, obligations, and other deontic constraints and enablements.” This is not to say that all forms of structuring are necessarily institutional, but that what will characterize institutional structures is that they sort actions into deontic categories like permitted, obligatory, and forbidden, or some variation on that division, such as acceptable, appropriate, and inappropriate. We can call this the deontic conception of institutions.

In seeking to account for patriarchy as institutional in this sense, i.e., as a deontic structuring of our actions, one potential obstacle immediately presents itself in applying the above model, namely that it is typically developed in a highly rules-centric way.^[2] For instance, North suggests that “[i]nstitutions are the rules of the game in a society” (1990, p. 3), and Gilbert (2018, p. 30) characterizes an institution as “a system of rules that is a blueprint for human behavior.” In political theory, Rawls (1999, pp. 47–48) takes the stance that an institution is “a public system of rules which defines offices and positions with their right and duties, powers and immunities, and the like. These rules specify certain forms of actions as permissible, others as forbidden; and they provide for certain penalties and defenses, and so on, when violations occur.” This kind of account tends to be put forward primarily with formal institutions in mind, where there might even be laws regulating which kinds of moves that are open to whom and when. But especially if we seek to analyze ways in which contemporary Western societies are patriarchal, we will be interested in more informal structures. Let us take a couple of examples. They will be somewhat stereotypical, but can hopefully still work as illustrations.

We can imagine a person, Joe, who lives in a run-down apartment and works hard at a tedious and poorly paid job. Joe is not a socio-economic winner in his community. But that community can still be institutionally structured so that Joe is at liberty in a variety of ways that his wife Jane is not. When Joe is at work he might not worry about such things as who cleans up after him. When he comes home he has the right to just sit down and relax after a hard day’s work. And when he and Jane converses at night, he always gets the last word and is free to go on at length explaining for Jane how the world works. Jane, on the other hand, not only works hard at her even worse-paying job, she keeps picking up trash and doing dishes after her male coworkers, after which she picks up the kids, shops for groceries, cooks, and does the dishes at home. Finally, when she and Joe sits down and converses a bit, the expectation that they both share is that she mostly listens and affirms him in his understanding of how the world works. Now, presumably, neither of them need subscribe to anything like Millett’s first principle of patriarchy, male shall dominate female. They might even strongly object to it. They just repeat certain types of actions, similar to actions they have done again and again, and have an inarticulate sense of appropriateness in doing so. But the fact that they are not guided by any sexist principles or rules does not mean that there is no pattern here, nor that they are not leading their daily lives within the bounds of a sexist division of labor.

Now imagine two persons, Jack and Jill, at a bar. Jill is there with her friends and mostly just interested in hanging out with them. Jack takes an interest in Jill and does a number of things, such as smiling at her when she can see him, paying for a drink and letting the bartender give it to her, passing a compliment when he just happens to walk past her in very close proximity. Jill feels uncomfortable, but when Jack after a while addresses her, she still feels vaguely obliged to answer him in a polite way. What are we to make of this situation? There is a reading of the situation in which Jack is harassing Jill. But there is also a reading where Jack is just doing a series of nice things: he gives her attention (who does not like attention?), he is generous (who does not like getting a drink for free?), and he is amiable (who does not like a nice compliment now and then?). Let us assume that in the setting in which Jack and Jill find themselves there is no common shared reading of this type of scenario, and that many might not even have a definite reading at all, but are uncertain about how to read it. Consider then that eventually Jill calls Jack out and angrily tells him to stop harassing her. He responds that how could he have been harassing her – he has done nothing but nice things. Jill looks to her friends for support and one of them, uncomfortable with the situation (as well as with Jill being so loudly assertive), says “I don’t know, Jill, he hasn’t really done that much, has he?”

Let us assume that neither example is atypical, that we are looking at a society where similar interactions happen relatively frequently and are relatively wide-spread. For something to be institutional, a minimal requirement would be to involve patterns of reoccurring behavior, but would patterns like these really rise to the level of institutionally structured behavior? One version of this worry would be that even if there are patterns, they might not be deontic in character. And certainly, the participants themselves would not characterize their behavior in terms of rights and duties. But at the same time, both John and Jack are acting as if they are entitled to behave in certain ways and can legitimately expect certain things of Jane and Jill. At the very least, it seems fair to say that there are normative expectations involved. And the normative aspect of these expectations is arguably deontic in character, since being entitled to something is a kind of right, and being normatively expected to behave in a certain way can be understood as being under a duty to behave in that way. This is not to say that these are bona fide rights and duties, that John and Jack are in a deep sense entitled to behaving in these ways, but simply that the expectations involved can be understood as deontic in character.

Another version of the worry is that while there are normative expectations involved here, they should be understood in terms of norms instead. Trying to draw a sharp distinction between norms and institutions will however be difficult, since on a deontic conception of institutions, norms (in some sense) would presumably be central to institutions being in place.^[3] Some norms might certainly still be mere norms, but at least for present purposes, the sense of institutions that is of interest is of something that is systemic in character. A single norm might be particular to some quite specific set of actions, but if we ask if a society is patriarchal, we are asking about how patterns for many different types of behavior are similar enough in character so that they add up to something systemic. This aspect of institutions is stressed by Hodgson (2006, p. 13), who suggests that we understand institutions as “durable systems of established and embedded social rules that structure social interactions”. Setting Hodgson’s emphasis on rules to the side, it does not seem unreasonable to think that if there is a significantly systemic character to certain durable normative patterns, then those patterns qualify as institutional.

If we stress the systemic character of institutions, the answer to the question about whether institutions always involve rules, or what roles rules play in institutions, can be understood as a question about whether durable patterns of behavior and synchronized normative expectations presuppose certain rules being in place. This question about the role of rules is not just a question for the present project, but a more general one about how to best understand institutions. As pointed out by North in the passage cited above, institutions can involve both formal and informal elements. Yet many social ontologists, as well as liberal political theorists, tend to focus on institutions that are relatively formalized, such as money. It then becomes important from a feminist perspective to emphasize that institutions are about more than what is defined through laws and other explicit rules, lest we risk ending up with an overly narrow conception of the domain of justice and of what can be proper objects of political interventions.^[4]

2 Towards a Distributivist Account

With respect to informal institutions, their more inarticulate character seems to be an integral part of being informal – indeed, participants themselves might be hard pressed to articulate rules summing up their behavior. Take an institution like the family: while there are codified aspects to it (family law), much of the normative expectations that we have on each other and which govern our interactions in the realm of family life are much vaguer. And yet, it would certainly seem that depending on which positions we occupy in a family, there would be different normative expectations on us and that very similar actions can have different deontic statuses depending on whether, say, a father or a mother performs them. For instance, a father who prioritizes work and does not spend much time with family might safely be in the region of permissible action, at least in the eyes of the normative peer community, while the same might not hold for a mother.

It should however be recognized here that there are at least two senses in which we can talk about institutions in terms of rules. One is that rules can function as coordination devices or decision-making tools, i.e., where individuals relate to and are guided by these rules in navigating the social world. The other is that the relevant behaviors can be rule-like in character. Some writers primarily talk about rules in the first sense, such as Greif and Kingston (2011, pp. 14–15) who suggest that “while a ‘rule’ may serve as a coordination device, it is fundamentally the expected behavior of others, rather than the rule itself, which motivates people’s behavior.” In contrast, someone like Hodgson (2006, p. 18) is clearly interested in both and suggests a disjunctive conception of rules in order to cover both senses, namely that rules “are understood as socially transmitted and customary normative injunctions or immanently normative dispositions” – where in the latter case, talk about rules then refers to the rule-like character of sets of dispositions that agents have. To some extent this is mainly a terminological matter, but not all terminological questions are merely terminological. Even when two terminologies might analytically be used to explicate the same sets of cases, depending on which framework one uses different cases might come out as typical or atypical and different features of those cases might come out as core features or not. Opting for a different conceptual apparatus can accordingly sometimes be about which way of analyzing phenomena that will best facilitate making certain aspects clearly visible.^[5]

Let us return to the two examples above. They do not feature rules in the sense of something that the agents relate to. If we take the John and Jane example, however, it might perhaps be interpreted in terms of something like an unwritten rule being in place, or that the dispositions exhibited by John and Jane are rule-like in character. But at the same time, it does not seem obvious that their respective dispositions should be understood as rule-like, so at the very least it seems wise to consider our options. If we turn to the Jack and Jill example, it is more complicated, not just because it is a one-off situation between the two, but also because it potentially takes place in a gray area. It is arguably also a type of case that highlights how there can be vague patterns with certain tendencies even within gray areas, as well as how these might even serve to protect those who are already privileged and put others in a position of vulnerability. Patterned gray-area behavior seems at least prima facie difficult to make sense of in terms of rules, because this would not just be about certain behaviors falling between rules. This is not to say that it would be impossible to provide a rules-based account, but again it seems reasonable to at least consider our options.

If we assume that institutions are grounded in our attitudes and behaviors, with shared ideas and sentiments establishing different entities and features of the social world that we share, then the shape of those entities and features will presumably be determined by how we categorize the situations and persons we encounter and the actions that we contemplate or react to.^[6] And there can be systemic patterns in how we navigate the world without these being rule-like in character. While applying rules is obviously one way in which human beings sort things, it is not the only one. Indeed, the last 40 or so years has seen considerable empirical work on how human beings categorize, complicating the picture of how categorization works. While rules provide sharply bounded conditions for some X being a Y or not, many theorists of categorization have pointed to how human beings instead often sort by using exemplars or prototypes, where an X is being sorted as a Y by being similar enough to either a set of previous examples or to a statistical type.^[7] If we are categorizing actions, then a rule-based assessment would be based on the feature(s) of the action made salient by the rule. For instance, if some people are guided by the rule that it is appropriate for a woman to defer to a man’s judgment, then this rule points out three salient features, one about a speaker (being a man), one about a listener (being a woman), and one about an activity (making judgments), as well as providing the appropriate response (deferring to someone). What characterizes a rule is that it zooms in on certain features as the essential features.

In similarity-based categorizations, however, we operate with richer descriptions of the relevant objects of assessment, where there is a number of features that all have some relevance in that they are present in many of the cases where we see things in a certain way. For example, on the question of whether a person is to be deferred to on certain matters there might be a range of features that serve as partial cues for which response is felt to be appropriate: gender, skin color, age, height, body shape, hairstyle, clothing style, vocal timbre, dialect, sociolect, and so on. Some of these might have more weight than others, but whether a certain response is seen as appropriate or not will be a function of how many of the relevant cues are present (and to which degree, given that some of them might themselves come in degrees).

Similarity-based categorizations turn on matters of typicality and will always be characterized by the existence of gray areas, where certain instances have some of the features that characterize the prototypes or exemplars in question, but where the classification of those instances will be less than clear-cut. To the extent that a sorting of actions into deontic categories works by exemplars or prototypes rather than rules, we would accordingly expect it to be characterized by many actions falling into gray areas (some similarity to the relevant exemplars or prototypes, but not enough to clearly fall into the relevant categories). If we consider the Jack and Jill example, it can be understood precisely in terms of how Jack’s behavior is not seen as positively correct: there is no rule that he is living up to, but since it has some similarity with clearly acceptable behavior, the response to it is wavering and uncertain.

Of course, one can always stretch the notion of rule-like so that is covers typicality as well, but doing so is not unproblematic. Many institutions involve both formal and informal aspects, and since the former are characterized by there being clear-cut rules that people relate to, referring to both kinds of aspects as rules could involve setting up a misleading image of what institutions typically are like.^[8] It should also be said that, at least if we start from the deontic conception of institutions, the suggestion being made here is really quite modest. Human beings make use of both rule-based and similarity-based categorizations,^[9] so the main point here is just that we should not prematurely restrict the kinds of patterns that we might explicate with our fleshed-out conception of institutions by just assuming that patterns must always be rule-like in order to be patterns.

The proposal is that instead of fleshing out the deontic conception of institutions in terms of systems of rules, we instead understand institutions as being systemic distributions of deontic statuses that structure social interactions. While such distributions can be the product of a system of rules being in place, with people relating to and being guided by that system of rules, they can also be established through more organic and gradual processes where our normative expectations with respect to certain action prototypes or exemplars have become coordinated and synchronized to the point where it makes sense to speak of such distributions being in place. The notion of deontic statuses can serve as the common denominator between the two.

3 Deontic Statuses as Relational

When institutions are formalized, there are rules in place. And while people might not always obey them straightforwardly, we can at least take those rules as a starting-point in explicating those institutions. Informal institutions present more of a challenge. The approach suggested above is an attempt to articulate a unified conception that covers both formal and informal institutions, and it will now be suggested that the idea of institutions as systemic distributions of deontic statuses can be further developed though a schema where rights and duties are understood as relational.^[10] The difference between formal and informal rights will be that while a formal right is clearly defined and has relatively sharp boundaries, an informal right is prototype- or exemplar-based – it will be fuzzy, although certainly still part of the social reality that we navigate.^[11]

In doing micro-level analysis of deontic relations, what is arguably the classic account was articulated by Hohfeld (1913, 1917. A defining feature of the Hohfeldian schema is that even broad and general rights, such as the right not to be killed, are really “many separate and distinct rights, actual and potential, each one of which has a correlative duty resting upon some one person” (1917, p. 742). There are four main types of incidents (as they are known) which we can possess in relation to each other:

1. A has a claim that B φ if and only if B has a duty to A to φ.

2. A has a liberty to φ in relation to B if and only if A has no duty to B not to φ.

3. A has a power with respect to B if and only if A has the ability to alter or determine B’s Hohfeldian incidents.

4. A has an immunity with respect to B if and only if B lacks the ability to alter or determine A’s Hohfeldian incidents.

Claims and liberties, the first-level incidents that we hold in relation to concrete others, are about which moves are open and which moves are not. Hohfeldian liberties are not about being positively able to accomplish what one wants, but simply about freedom from being deontically restrained: others have no legitimate standing to rebuke or punish you if you pursue that course of action. The second-order incidents, powers and immunities, introduce a dynamic dimension to which moves that are open or not (and where immunities set limits for exactly how dynamic a situation can be in terms of how incidents can change during the course of our interactions).

Because of their inherently relational character, the above four incidents all come with certain correlates:

A has a claim ↔ B has a duty

A has a liberty ↔ B has a no-claim

A has a power ↔ B has a liability

A has an immunity ↔ B has a disability

What this schema means is that for any concrete person, we can characterize (at least in principle) all her deontic relations to all other incident-holding persons to get a complete deontic profile of how she stands in the world. This kind of schema can clearly be used to analyze formal institutions involving legal rights and duties (indeed, that is what the schema was originally intended for). This is an advantage. Even for someone contemplating a shift to a distributivist conception of institutions, it should still be recognized that, at least with respect to formal institutions, rules can play a central role in creating and maintaining the relevant distributions of rights and duties. But there is nothing in the schema that prevents it from also being applied to prototypical actions as well when used as an analytical tool to explicate patterns of deontic statuses based in our normative expectations of and responses to each other.

Given the model of institutions and social positions suggested above, our main analytical categories for understanding domination or oppression will be in terms of comparative differences in distributions of right and duties, where some individuals will be less constrained, others more. The Hohfeldian table of correlates provides a straightforward way of understanding positions as advantaged or disadvantaged: the left-side incidents (claims, liberties, powers, immunities) are all about enabling and protecting certain moves as being available to us, while the right-side incidents (duties, no-claims, liabilities, disabilities) are all about constraining us in terms of the moves that are available or by leaving their availability open to decisions by others. For a social position to be dominated or oppressed is then for it to be systemically circumscribed in terms of the moves available compared to some other related social position. And for a society to be patriarchal would accordingly involve a distribution of rights and duties so that not only is being a woman a systemically disadvantaged social position compared to being a man, but being older and being younger are also factors that systemically shape these advantages and disadvantages, so that being an older man typically comes with certain additional advantages (and where being older need not be understood in terms of a definite age-point but can be a position that one occupies to different degrees).

The fundamental way in which there can be an asymmetry between two social positions concerns the extent to which they are characterized by liberties or by being duty-bound. If one occupies a position that is strongly characterized by being duty-bound, many actions will typically come with social costs, and if one is duty-bound in informal ways this means that when one performs certain kinds of actions, people will predictably (although not necessarily universally) have a variety of negative reactive responses to them (and further responses by additional others will tend to be supportive of those responses). To the extent that interactions in a certain community are characterized by a pattern of what Manne (2018, p. 68) has called down girl moves – actions and responses that serve a punitive, deterrent, or warning function, which constrain women in informal ways and where the extent to which men tend to be subjected to similar moves is significantly lower – then that community can be understood as being institutionally sexist in that being a man is a position of greater liberty and being a woman is a more duty-bound position. Since being at liberty is about an absence of claims against one, and accordingly an absence of negative reactive responses to one’s behavior, one should not expect being at liberty to be something that is necessarily experienced as a pronounced feature of one’s situation. The social advantages that come with occupying a position of liberty is not about positively having been granted certain absolute advantages, but a comparative advantage in relation to some other social position that is more duty-bound. To be advantaged in some such ways is also perfectly compatible with being disadvantaged in other ways.

4 Hierarchies of Authority in Patriarchy

Saying that a society is patriarchal involves saying more than that it is sexist. While it is not explicit in something like Millett’s second principle, one key feature of a patriarchal system is presumably a governing idea of the father figure as an authority figure. In making sense of patriarchy as an informal institution, we accordingly need an account of such authority.

Using the Hohfeldian schema we can distinguish between at least two aspects of authority. One has to do with having claims on others for their attention and respect, i.e., to occupy a position of authority is about others being duty-bound to at least show a certain level of respect or deference. The other has to do with possessing powers. Now, the Hohfeldian notion of power is not intended to capture everything we mean by power, but many things we mean by someone having power in an institutional context can be understood in Hohfeldian terms. Presidents and prime ministers, CEOs and high priests, deans and heads of department, and so on – the institutional power that lies in such social positions is to change the incidents held by those in subordinate positions. This can be done by giving an order, thereby creating a duty that is now held by the subordinate in question, but also by delegating certain powers, subordinating certain people to that person. Holding such powers of appointment is characteristic of being in a position of authority within a more organized institution and depending on how different powers and immunities are distributed, a hierarchy is created.

Groups of people are however also ordered hierarchically in more informal ways. We are all participating in economies of esteem and attention, where being well-placed involves having claims to respect and attention, and where having power will involve holding what might be called powers of anointment, to be able to make moves that authoritatively single out certain others as also being owed respect or deference. For instance, if the professor of great prestige publicly heaps praise on his young male protégé, people’s perceptions might be realigned so that this person is owed respect as well. Such powers of anointment need not be about anointing a person as a whole, but can also be about being able to act as a form of authoritative judge on the quality of different parts of a conversation or discussion – being able to say or indicate what remarks that were relevant or irrelevant, important or unimportant. Another aspect is what might be called powers of arbitration, i.e., when there is a dispute about who owes what to whom, the judgments of a person possessing such powers can settle that dispute and situationally solidify patterns of expectations and responses in specific ways. While there are, as already pointed out, many ways in which a society can be patriarchal, it seems reasonable to think that to the extent that being an older man is an advantaged position, it is largely about asymmetries in how these kinds of claims and powers connected to authority are distributed.

Authority comes in different forms and while it is perfectly possible for a quarter or so of the population to hold a disproportional share of some such claims and powers, it is not possible for that many people to hold positions of major authority. Of course, it is simply a fact that in basically all societies such positions of major authority are disproportionately held by older men (albeit in some societies less so than what has historically been the case). This disproportionality seems unlikely to be a matter of chance, and given the present account, one potential explanation of the underlying mechanisms would be in terms of how our shared sense of appropriateness is aligned so that being an older man involves disproportionately being perceived as fit for shouldering the role of an organizational or societal father figure, i.e., being in line with the prototypical authority figure in a patriarchal society. This explanation is perfectly compatible with most older men, to the extent that they pursue positions of major authority, losing out in the struggle for them. The patriarchal character of this struggle would rather be about how they at least had better odds at ascending to, as well as staying in, such positions. Indeed, the struggle for positions of major authority can be rigged in ways that need not be clearly visible to those who are advantaged. For instance, someone occupying a more duty-bound position might struggle to put in the kind of concerted effort needed to reach those positions, because a lot of time and energy is being taken up by various duties. But it can also be about how specific types of effort, such as those involving more aggressive or assertive behavior, might run counter to normative expectations dictating that occupants of certain social positions largely abstain from such behavior.

An example here might be about how when running for higher political office, like the presidency of some country, older men have been and arguably still are advantaged, especially compared to women candidates. The relevant kinds of regularities in expectations and responses can then be in place without voters necessarily understanding themselves as focusing on a candidate being a woman as a disqualifying factor when they respond to a specific woman candidate in a certain way. They might just have a vague feeling of there being something off-putting about that woman (as well as that other woman, and so on). The proof here is ultimately in the pattern – it is not about how people interpret themselves, but about how their expectations and responses line up to establish certain regularities. Of course, one can always debate the interpretation of specific examples, but for present purposes the important thing is just to illustrate how there can be regularities in our normative expectations and reactive responses in place without there being some rule, such as women should not become president, in place. In fact, to continue with this example, it is even possible that a woman candidate will eventually win – a game being rigged need not mean that it is impossible to win, just that it is consistently more difficult for occupants of some social position(s) to do so.

5 Matters of Degree and Gray Areas

A key test for a model supposed to be able to account for informal institutions is whether it can be helpful in explicating gray areas between straightforwardly appropriate and straightforwardly inappropriate behavior, including how the existence of gray areas can function to create opportunities for behaving problematically, yet with relative impunity, for those who are already privileged. Let us return to some of the examples considered above.

First, take the example of a woman deferring to a man’s judgment, guided by the normative expectations in place in that setting. Even given such a background of patterned normative expectations, there will typically still be many differences of degree involved in determining the exact responses that are felt to be suitable. Not all men might be seen as equally worthy of being deferred to, and even some women might be seen as somewhat worthy. There could also be some people who occupy gray-area statuses where responses to them are highly uneven and wavering. Given the model proposed here, such things are only to be expected. The deontic relations which are in play in a setting can easily vary in how strong they are depending on what the relevant prototypes or exemplars are and how similar the duty-bearers, the right-holders, and the relevant actions are to these. It might be that the prototypical person to defer to would be a white, older middle-class man and that certain other cues like being tall, having a deep voice, being dressed in a suit, having a strong jawline, and so on, complete the prototypical idea of the kind of person to whom deference is owed. The distance of different individuals to that prototype (or set of exemplars) would then determine the extent to which there is a shared sense of appropriateness with respect to deferring to their judgment. For instance, even if the perfect image of being an authority figure involves being a man, there might for some women still be certain overlaps with some of the details in the fuller image guiding people, so that depending on skin color, vocal timbre, clothing style, etc. there might be less distance to the prototypical person that is owed deference or to be taken seriously.

Apart from certain kinds of persons being more or less clear-cut as the kinds of people to whom different things are owed or not, matters of degree can also enter into the kinds of behaviors that are seen as being owed or not – and the two can of course come together. As already pointed out, on the account proposed here, gray areas are not anomalous. They are only to be expected. The prototypes or sets of exemplars in place will establish and maintain what might be called deontic reference points with respect to which we can navigate the social world, but where many actions will fall into the category of being deontically indeterminate. What characterizes gray-area behavior is that others might give one a pass for behaving in that way, but also that they might not. These are forms of behavior that have not been fully negotiated on a more general level and where there is more room for negotiation in context. This also means that when concrete gray-area actions become contested, it will typically come down to interpretation in context whether someone’s behavior is close enough to an established type of liberty or whether it is close enough to violating an established duty. One consequence of this is that, for any given agent, whether engaging in gray-area behavior primarily constitutes a risk or an opportunity will depend on the interpretative authority of that agent. To the extent that one’s social position is fraught with interpretative deference, gray areas will predominately be areas of risk, and to the extent that one’s social position is characterized by interpretative authority, gray areas will predominately present opportunities where one can make pushes, performing actions that start to approximate behavior which is clearly out of bounds, but where one can still feel relatively confident that one will be able to counter challenges to one’s behavior.

When analyzing the normative terrain that we have to navigate as socially situated individuals, we accordingly need to distinguish between two ways in which a person can be considered to be, on the whole, free to do something, either (i) in terms of an action being institutionally supported, where I have a (clear-enough) liberty to perform actions of that type, or (ii) in terms of an action being institutionally facilitated, where because of my overall standing in the relevant institutional framework, I can relatively reliably get away with performing actions of that type, even if they are not positively established as liberties. If we return to the Jack and Jill example and apply these notions, then to the extent that Jack can relatively reliably get away with this kind of behavior, this would be an example of assertive gray-area behavior that is institutionally facilitated. Jack pushes into a territory of behavior that is generally seen as inappropriate, but because of how the social positions that he and Jill occupy are structured, she is at an interpretative disadvantage when calling him out.^[12] Nearby deontic reference points can then provide a form of interpretative refuge points where someone can seek cover when being called out in some way, at least given that they occupy a contextually empowered social position. With this gray area in place, for Jack going to the bar there is predominately an opportunity for engaging in a certain kind of behavior, whereas for Jill going to a bar there is predominately a risk of being subjected to a certain kind of behavior.

Of course, apart from there potentially being problematic inequalities between different persons in terms of the extent to which they are institutionally facilitated in their behavior, the psychological effects of such asymmetries can also be restrictive in that if you are (at least implicitly) aware of occupying a social position for which uncertain normative terrain often equals risky normative terrain, you will presumably be prone to avoid certain social situations or even entering certain physical locations, in order to reliably steer clear of unpredictable interactions where you are at disadvantage if things come to hinge on disputes about how the situation should be interpreted. In contrast, the already privileged person might go into such social situations much more confidently, and that confidence in itself might then also serve a further suppressive function in relation to those for whom the situation is one of risk. Having tools for analyzing asymmetries in gray-area interactions is accordingly very important for addressing structural injustices.

6 Addressing Possible Worries

With the basic framework in place, let us now look at our initial worries. To begin with, how does intersectional analysis enter into the picture? One key feature of the model proposed here is that which moves that are available to us, either by being institutionally supported or merely facilitated can be based in complex processes of categorization involving a large number of features. Relevant prototypes and exemplars can be intersectional in character, as shown by our earlier example of the white, older middle-class man as typically possessing a certain level of authority in many contexts. Many stereotypes also involve bundles of traits, and these stereotypes might then serve to establish patterns of normative expectations between members of different social groups. Other patterns might be more connected to practices that have been negotiated in-group, and which play less of a role in interactions between members of different social groups. Complications like these are all in line with the account of institutions proposed here, and it should accordingly be highly compatible with an intersectional analysis of social positions.

Consider the Joe and Jane example. As described they are a working-class couple. In an abstract sense, they can very well be similar to middle-class or upper-class couples who are also characterized by there being an asymmetry in terms of the incidents they hold, but on a more fine-grained analysis, the normative terrain occupied by Joe and Jane will in many ways be distinctly shaped by their class positions. The concrete forms in which Joe as a man and Jane as a woman can excel socially, given their context, might in many ways be different from, say, James and Jenny, a couple of comparatively well-paid academics who lead very different lives, partly because of how economic circumstances can enable normative expectations to shift. Race is, of course, then also a further factor which can complicate the normative terrain that any concrete individual has to navigate. Say that James and Jenny are both Asian Americans. They are certainly privileged in some ways, having much better-paid jobs than Joe and Jane, and being able to afford help so that a lot of household work does not become an issue of a gendered division of labor between them. While both intelligent and hard-working, they are however rarely seen as owed credit for what they do in their places of work, since the expectations on them tend to be that they are simply natural perfectionists. And especially Jenny finds that she is under an expectation to always be very agreeable and that people tend to just assume that she will acquiesce in discussions, something which holds in relation to her white women co-workers as well.^[13]

If we compare Jane and Jenny, it might be difficult to articulate specific rules that constrain both of them. On an abstract level, exemplified by Millett’s two principles, they are both expected to be submissive in important ways, but the concrete ways in which their social positions are dominated by right-side Hohfeldian incidents might be quite different. On the present analysis, however, there need not be any specific rules that regulate both of their lives in terms of more specific constraints. Indeed, at a fine level of grain there might even be no specific duties shared by all persons occupying the social position of being a woman, but simply similarities between different complex intersectional positions of which being a woman is a part. For a society to be sexist on a general institutional level will rather be about how, when the only difference between two people is that one is categorized as a man and the other as a women, there will typically be an asymmetry in how the normative terrain that they navigate is deontically structured. A society is institutionally sexist not because there is a specific form that this asymmetry takes for all women, but because occupying the position of being a woman relatively consistently brings with it a comparative deontic burden. As a marginal contributing factor being a woman regularly means a greater tilt towards holding right-side Hohfeldian incidents.

It should also be kept in mind here that social status is, on this model, ultimately held by concrete individuals and is not anything we hold simply qua being an individual of type X. It is always shaped by the particularities of our person and our history with those concrete other individuals with whom we mostly interact. To the extent that we interact frequently with some people, they will also be the people whose expectations and reactive responses are the most important in structuring the specific normative terrain that we navigate. Repeated interactions involve perpetually ongoing negotiations and renegotiations, where each of us might make moves that push against the boundaries of what we are at liberty to do. Gradually our spaces of meaningfully available actions might expand or contract in different ways. This means that even if we can still speak of generic social positions in terms of certain typical packages of Hohfeldian incidents, these are more like opening bids in the ongoing negotiations about social status that we participate in through our daily interactions with others. To the extent that we deviate from the generic patterns, such deviations will however primarily obtain in relation to those with whom we interact frequently and who know who we are. We will accordingly often find ourselves being back to the opening bids when meeting people we do not already know, and who then relate to us simply as generic occupants of the relevant social positions. This dynamic character of the present model should be seen as an advantage.^[14]

If we turn to the other worry, it should hopefully also be clear how the account proposed here should enable us to steer clear of the trap of Western people seeing patriarchy as something just characterizing “other” cultures or societies. It invites us to take a closer look at Western societies, not stopping at the formal level, which is typically relatively egalitarian. On the model proposed here, it is perfectly possible that certain reoccurring patterns can be identified even if many of us are not ourselves cognizant of these as reoccurring patterns, at least not with respect to our own societies. We might not see the forest for the trees, i.e., we might be synchronized in our normative expectations and responses to a variety of actions, thereby establishing a series of deontic reference points that will structure the normative terrain that we occupy, but still fail to properly see the kinds of regularities that all these particular deontic relations add up to (cf. Brännmark 2019b). In stressing how patriarchy can be an informal institution, with many of its workings potentially being quite subtle, it is only to be expected that it can be quite difficult for specific individuals to see how their lives are shaped by patriarchal structures.

But is there, then, not a risk that patriarchy on this account becomes too nebulous to be graspable? Certainly, in concrete cases it will often be difficult to definitively trace different aspects of how an individual fares to the fact that this person occupies that social position. Say that a relatively competent older white man and a relatively competent younger woman of color both apply for a job, and the man gets the job. It could then be very difficult to separate out the difference made for that outcome by which social positions they occupy or by their more precise levels of competence. The bare fact that it is difficult to tell exactly what in an outcome depended on which difference in their social positions does not, however, mean that is safer to simply assume instead that the difference was due to the more distinctly personal qualities of the individuals involved. That difference-making cannot be neatly separated out either.

On the account suggested here, occupying a disadvantaged social position will be about risk, not destiny. And this is just what should be expected for such highly complex matters.^[15] Take an entirely different example: we know that smoking greatly increases the risk of lung cancer. But we also know that it is possible to get lung cancer without smoking. So when a specific smoker develops lung cancer we will typically not be able to say conclusively that it was because of the smoking. Thinking in terms of risk factors is however also compatible with in some cases it being relatively clear that the most likely explanation involves a specific risk factor. The point is just that qua explanations of specific events in the lives of concrete individuals, appeals to relevant patterns (or institutions more specifically) will often have a tentative character. Yet giving up on having simple explanations of particular events is not the same as giving up on the idea that certain patterns are perfectly real, nor does it mean giving up on the idea that such patterns can be proper subjects of policy interventions on a population or group level. These would simply be interventions that seek to address problematic risk factors, similar to how policies that target smoking do not presuppose that we can always tell exactly which persons who got lung cancer that did so because of smoking. For the prospect of political action, there being certain patterns is what matters.

7 Concluding Remarks

In articulating a piece of social ontology, one presumably tries to sharpen the contours of social reality, making various social phenomena more discernable, easier to talk about, and potentially even possible to address politically. We can opt for an institutions-as-rules account or an institutions-as-distributions account. Both will provide conceptual tools. But it has been argued here that the latter is superior to the former in that it allows us to better explicate informal institutions or informal aspects of institutions, e.g., in making sense of gray areas, and how such gray areas can involve not just vagueness, but also power asymmetries. At the end of the day, this is a completely general point. If the institutions-as-distributions account can handle both formal and informal institutions well, we have good reason for adopting it, at least as a way of fleshing out a deontic conception of institutions. But this is a bigger question, going beyond the scope of the present paper.

More specifically, the argument here has however hopefully shown that if we work with an account of institutions that can explicate informal, but still reoccurring, types of deontic relations between us, we can make sense of how patriarchy can be institutional in a way that both enables us to understand how the relevant structures might not be clearly visible even to those that live under patriarchy, but also how it can intersect with other institutional structures. Given this kind of account, the basic understanding of patriarchy can actually be quite simple, more or less in line with Millett’s classic formulation, but as an actual institution in any concrete society, it should be expected to be anything but simple, and far from monolithic. Yet it can still be there.