A. Conditions of resentment

With that basic framework in place, I will now describe what I take to be two important conditions for fitting resentment, and thus conditions of this sort of accountability. Though much of the literature is concerned with the freedom or control conditions for accountability, I will be concerned with the way that an agent must be able to reasonably know and endorse the demands to which we hold her accountable. Call these the known demand condition and the justification condition respectively. As I will indicate, it is often difficult for these conditions to be satisfied and reasonably known to be satisfied, but stable social norms meliorate these difficulties. This will bring out some important ways that social norms can facilitate relations of mutual accountability.

The conditions of resentment come from the way that the emotion presents its object. When we resent someone, we see her as having had ill will toward us. Our resentment seems misplaced if directed at things we know to not have any will at all, such as a wasp or the weather. Similarly, our resentment is not fitting if directed toward those who cause us harm by pure accident or without relevant control, as when they had no reason to think that the action would cause us harm. Though the accidental harmer has a will, and her action was an expression of her will, she did not show any ill will toward us. It is not that she failed to give our interests due regard, but rather that she failed to see how our interests were at risk. Furthermore, even someone who knowingly and intentionally causes you some harm is not necessarily displaying ill will, for she may have been justified. It would not be fitting to resent someone, say, for pushing you if you know that she only did that because it was necessary in her attempt to quickly get to and save a drowning child. In that case, action does not show insufficient regard for your interests insofar as there were sufficient reasons to act despite the minor harm to you; pushing you was the morally right thing to do. I raise these fairly simple examples merely to bring out the most basic conditions of resentment in which we can see the nature of the quality of will at stake.

B. The known demand condition

Among other things, resentment seems to include what I call the known demand condition. When we see agents as having shown ill will, we see them as having acted against certain demands on their action, or normative expectations and requirements directed to them. It will only count as acting against the demand if the agent knew the demand, or at least reasonably should have known it.^{Footnote 12} The charge of ill will seems undermined if this condition is not met. Imagine, for instance, that the resented party was to sincerely ask: “How was I supposed to know not to do that?” Were we to not only lack an answer, but also to flatly admit that there is no reason for her to have known, it would seem odd to continue to hold her accountable through resentment and maintain that she had shown ill will. If, say, God told me alone of some new law, not discernable to natural reason or otherwise than through revelation, it would not be fitting for me to begin resenting violators before at least telling them of the new law. Other responses do not require such knowledge, but resentment seems to presuppose that demands are known rather than simply made.

The known demand condition will strike many as benign and unimportant, for moral requirements seem obvious to most people; we do not think we generally come to know morality only if we are lucky enough to receive God’s secret dictates. Instead, moral theories tend to appeal to basic considerations that are (supposedly) widely recognized and perhaps universally recognizable.^{Footnote 13} Be that as it may, it seems uncontroversial that moral values and requirements are obscure at significant margins. This obscurity comes both because there are different foundational possibilities that often produce different answers and because applying any one of them can be extremely difficult, particularly where our moral concepts are vague, the particular interests of individuals are hard to discern, values conflict, and the considerations of a particular case are complex.^{Footnote 14} Insofar as morality is obscure, even well-meaning people acting with due diligence may simply not know what they are supposed to do or what is demanded of them. These cases will tend to undermine the appropriateness of resentment and holding people accountable, for they may be unfortunate, ignorant, or misguided, rather than expressing any ill-will.^{Footnote 15} Moreover, knowing that a person is likely ignorant of the relevant demands, even if in fact he is not ignorant, will tend to undermine the rationality of resentment, and thus to undermine relations of mutual accountability.

C. Social norms and the known demand condition

One of the reasons we seem to be able to hold each other mutually accountable as widely as we do is because our lives are infused with social norms. To a great degree, we hold each other accountable based on rules governing particular sorts of social circumstances. We guide our actions in accordance with these norms, expect the compliance of others, and hold others to those norms. With such norms, we do not have to appeal to fundamental values, the complexities of our competing interests, advanced exercises of practical reasoning, or similar considerations about which good willed people might reasonably disagree. Social norms, by their very nature, will not be so prone to obscurity, for the very existence of a social norm depends on shared beliefs about expectations. For instance, on Bicchieri’s account, a social norm, defined in terms of a rule, only exists when for a sufficiently large part of the population each person believes that a sufficiently large part of the population expects compliance with the rule.^{Footnote 16} That is, most people believe that most people hold the norm. It is generally safe to presume, until given reason to doubt, that someone will know the social norms for her society; she is not likely to be able to sincerely ask: “How was I supposed to know?”

It is worth emphasizing that social norms, as Bicchieri describes, “are embedded into scripts,” which are “generic knowledge structures” and “essentially prescriptive sequences of actions of varying levels of specificity that people automatically engage in (and are expected to engage in) while in particular situations.”^{Footnote 17} Scripts are themselves activated in response to common understandings of, or “schemata” for, situations, objects, and persons. Roughly, an individual will quickly and automatically conceive her present environment through prior generic categories, as by discerning that she is in a restaurant, a classroom, or a rock concert, based on environmental cues, such as table settings, desks, or a stage. This schema will trigger a corresponding script indicating the actions expected of the various agents involved, including the individual herself. People within a system of social norms will tend to share scripts and schemata that allow them to readily identify the same considerations as relevant and the same requirements as applicable to particular people.

It is easy to overlook the importance of these shared scripts for behavior and schemata for understanding our circumstances. Without these general ways of conceiving things, someone would confront a dizzying array of potentially relevant considerations when she is deciding how to act. Schemata enable her to quickly understand her current situation as fitting into a specific category (e.g., “classroom,” or “rock concert”) with associated scripts for how she and others should act. The relevant demands, then, are made salient for the agent, while a plethora of other potential considerations are pushed into the background or filtered out. Such shared understandings thus facilitate people having warranted expectations that others will be adequately aware of the demands associated with social norms, including knowing that a demand is relevant and applicable in particular circumstances, as needed for resentment to be fitting.

Above, I indicated that resentment presents the target as having expressed ill will, or insufficient regard, for the interests of people or other morally relevant considerations.^{Footnote 18} The details of the sort of regard we expect are complex, particularly varying with context and the sorts of relationships involved.^{Footnote 19} Insofar as social norms are grounding the particular demands within relations of mutual accountability, those social norms come to define the relevant interests that we must regard, as well as dictate how those interests must be balanced against each other. When we resent someone for failing to show us due regard, the amount of regard we take ourselves to be due is given by the social norms for the situation we are in. We do not, for instance, reflect on the appropriate balance of interests in light of our personal virtues and value as autonomous agents, but instead look to the social norms of fairness governing the division of goods in our particular circumstances. We resent the person who fails to give us an equal share of the resource, or reward us proportionate to our productivity, or to flip a coin, or whatever our norms may require in the case, and we resent them specifically for failing to show due regard for the fact that we should get an equal share, be rewarded proportionate to productivity, or the like.^{Footnote 20} At no point need we demand that they have any regard for our interests in terms of what we might do with the resource to be weighed against their parallel interests, for within the norm-structured relations the relevant interest is an interest in the resources themselves, and these interests are to be weighted equally, or proportionately to productivity, or the like. That we pick out some interests or considerations as the relevant ones for some context, but not for other contexts, is an important part of our ordinary moral life that we can better understand when we recognize the role of social norms in giving content to the morality to which we hold each other.

Social norms, of course, do not guarantee mutual recognition of demands in all cases because social norms face a number of limitations. They may be vague or conflicting in some cases, and some circumstances may not have sufficiently clear cues to make the norms salient.^{Footnote 21} Furthermore, though the existence of a social norm requires that most people know of the relevant rule, it does not guarantee universal knowledge. Some people in the population may remain ignorant of the social norm, perhaps despite reasonable efforts on their part. Where these limitations manifest, we may need to withhold resentment in light of the significant possibility that people are acting from failure to understand the demands rather than simply disregarding them. Despite these limitations, however, where we have social norms they can significantly support satisfaction of the known demand condition, particularly because they tend to develop in ways that will be by and large clear rather than obscure to members of the population in the situations they typically find themselves in. Though social norms do not provide guarantees, they do significantly meliorate the epistemic problem and support defeasible presumptions for our mutual accountability.

D. The justification condition

We must now turn to the justification condition. This condition comes in because resentment cannot be adequately grounded in a demand simply as such. We will not believe someone has shown ill will for disregarding a demand if the demand is purely arbitrary, oppressive, or the like.^{Footnote 22} Someone is only showing ill will if she disregards a demand that is justified. Though it will undoubtedly be controversial, I believe that the sort of justification relevant for resentment is what I call “interpersonal justification,” in contrast to impersonal justification.^{Footnote 23} Focusing on demands grounded in social norms, the justification condition requires that the norm be justified to each person, optimally meaning each person in the population having the norm, but at least meaning the particular people holding and being held accountable. So, if Abe is holding Beth to some norm, it must be that the norm is at least justified to Abe and to Beth. For the norm to be justified to someone, she must have sufficient reason to endorse the norm as a source of moral demands given her beliefs and values.^{Footnote 24} It will not be enough to claim that the norm is justified by true values or the facts of the world if these facts and values are not accessible to the individuals being held accountable.

It may help to consider a few brief cases. At least part of morality helps solve collective action problems by requiring participation in mutually beneficial practices. This provides one key way of justifying a norm. Many practices, however, are not beneficial for all members. Abe and Beth may be in a society with norms requiring everyone to contribute to a firework show (a classic example of a public good in the economic sense), but fireworks create for Beth more harm than benefit (say, because they cause her anxiety or frighten her pets).^{Footnote 25} Assuming the norm is not justified to Beth in some other way, the fact that the norm imposes costs on her without providing benefits undermines the thought that she is showing ill will when she violates the norms. She is merely avoiding exploitation, and from her perspective there is not sufficient reason to see the demand as worthy of regard. Abe may see people who are free riders enjoying the show as ill willed, but if he knows that Beth is burdened by it he should not resent her for abstaining from contributing. The same, however, should be said even if a norm is mutually beneficial but in ways that some parties reasonably do not recognize. For instance, Alice might observe that Bob does not follow norms requiring helping to drain the swamp. Bob may not help because his house is far away and, not understanding the relationship between swamps and disease-spreading mosquitos, thinks such norms exploit rather than benefit him. In such a case, Alice should see Bob as ignorant and misguided, even needing education and factual correction, but his failing to engage in behavior that reasonably seems to him exploitative does not warrant resentment.

Norms may be interpersonally justified in ways other than by mutual benefit, but in some way or other someone being held accountable to the demands of a norm must be thought to have sufficient reason to endorse the norm from her own beliefs and values. This may include endorsement based on the way the individual believes that the norm promotes the interests of other valued people, secures natural rights, or discourages behavior worthy of discouragement even if not against the interests of the individual or of people at all. For some people, indirect considerations like tradition or general popularity may provide sufficient reason to endorse a norm, and some may value a norm for its own sake. The key is that in resenting someone for violating a social norm we must think that the norm is justified to her somehow, while the particular content of the justification is not of much importance.

E. Social norms and the justification condition

Despite the variety of kinds of reasons that might justify a norm to an agent, the mere existence of a social norm will not guarantee that it is justified to any particular person. Strikingly, the mere existence of a norm does not guarantee that the norm is justified to anyone at all. A norm may exist even if it is unjustified for, and even consciously criticized by, every member of the population. This possibility arises most immediately from the fact that social norms require beliefs about the normative expectations of others, but not the actual having of those expectations. A social norm may thus exist based on widely shared false beliefs about people’s expectations, particularly if people conform with the norm to avoid sanctions or have incentives to be deceptive about their true views.^{Footnote 26} Furthermore, even if most members of a community in fact endorse a social norm, that norm may still be in excessive conflict with the beliefs and values of the accepting individual. In such cases, we might say that the norm is not justified, and its actual acceptance is a case of “ideology,” “false consciousness,” a mere taboo, or otherwise irrational. The mere existence of a social norm, then, does not necessarily support presuming that the justification condition is met.

Though the existence of a social norm will not itself ground a presumption of justification, it seems that there may be an appropriate presumption of justification for the system of moralized social norms that are well established and that have remained stable over an extended period. Consider, for instance, norms that have existed for a population across decades and generations, violations of which are typically met with resentment, and which have not seemed vulnerable to mass abandonment, defection, or criticism. Although there is nothing about such a norm that conceptually entails justification, there seem to be important tendencies for such norms to be justified. Long-standing norms will tend to be internalized, accepted as given, and thus seen as valuable for their own sake and shaping the way that members of the population understand their interests.^{Footnote 27} Furthermore, insofar as members of the population value tradition, conformity, the way the norm reflects group membership, group expectations, or the like, these provide additional reasons to endorse these entrenched norms regardless of their particular content. At least in these ways, it seems the existence of a norm itself tends to produce its own justification over time.

Beyond the justification that comes essentially from tradition, the benefits of a norm are likely to become more apparent over time. The benefits of a norm may manifest to people most clearly when the norm is complied with on many occasions, particularly in the case of norms that may be mutually beneficial only from multiple interactions. Many norms may seem in a single case to impose costs on some people in order to benefit others, but as roles are changed the agents who bore the costs will collect compensating benefits. Likewise, the benefits may emerge more from the overall predictability and coordination of action in the population, rather than particularly in any given instance. Even when the benefits of a norm are not readily apparent in operation, the teaching of the norm across time may be readily supplemented with explanations of what people take to be the justifications, particularly insofar as those teaching the norm are trying to promote compliance, and perceived justification contributes to compliance.^{Footnote 28} So, the various likely sources of justification for norms are likely to be better known, as they may be needed, as norms exist for extended periods.^{Footnote 29}

This accumulation and promulgation of justifications help address the justification condition insofar as the long-standing existence of a norm will tend to increase the probability that the justification condition will be met for members of the population. These considerations help support a presumption of justification for the demands grounded in the well-established system of moralized social norms. This is in strong contrast with demands grounded merely in what some particular agent believes are abstractly morally best, for there seems to be no reason to presume that those demands will be justified to others. Stable systems of social norms, then, can play an important role in our practices of mutual accountability by helping to support rational presumptions about not only what demands will be known to others, but also which demands will be justified to them.

This is of course not a guarantee that any given norm will be justified, and even where it does hold for most members of the population there may remain members for whom the norm is not justified. Even without such guarantees, however, a rational presumption of justification for the stable system of moralized social norms facilitates accountability relations even among strangers without special information about each other beyond their co-membership in the norm-governed population. When Alice sees a stranger, perhaps Bill, violate a long-standing norm, she may reasonably presume that he knew the relevant demand (since nearly everyone knows the norm) and that the norm is justified to him (since such norms tend to be justified to most members). While outside of norms the opacity of moral demands might make it too likely that Bill is merely ignorant of the demand or its appropriateness, the norms make such ignorance unlikely and thus leave ill will as a primary explanation for Bill’s behavior. These are of course merely presumptions, and Alice could gain new information that would require her to withdraw her resentment. She may, for instance, have to withdraw the presumption of justification if she knows that the norm has recently come under great criticism or is known to be opposed by a cultural minority group that Bill seems to be a part of, or if she learns that Bill has led an unusually isolated life that prevented him from learning the norm, or he is an eccentric philosopher with a strange moral theory from which he cannot join in the common endorsement of the norm. Though additional considerations such as these are possible, Alice may reasonably hold Bill accountable until she gains the information to defeat the presumption.^{Footnote 30}