The philosophy of personal responsibility has ruined criminal justice and economic policy. It's time to move past blame.
In an article published shortly before his death, the political scientist James Q. Wilson took on the large question of free will and moral responsibility:
Does the fact that biology determines more of our thinking and conduct than we had previously imagined undermine the notion of free will? And does this possibility in turn undermine, if not entirely destroy, our ability to hold people accountable for their actions?
Wilson’s answer was an unequivocal no.
He has lots of company, which should come as a surprise given what scientific research into the determinants of human behavior has told us over the past four decades. Most of that research, as Wilson says, points to the same conclusion: our worldviews, aspirations, temperaments, conduct, and achievements—everything we conventionally think of as “us”—are in significant part determined by accidents of biology and circumstance. The study of the brain is in its infancy; as it advances, the evidence for determinism will surely grow.
One might have expected those developments to temper enthusiasm for blame mongering. Instead, the same four decades have been boom years for blame.
Retributive penal policy, which has produced incarceration rates of unprecedented proportions in the United States, has been at the forefront of the boom. But enthusiasm for blame is not confined to punishment. Changes in public policy more broadly—the slow dismantling of the social safety net, the push to privatize social security, the deregulation of banking, the health care wars, the refusal to bail out homeowners in the wake of the 2008 housing meltdown—have all been fueled by our collective sense that if things go badly for you, you’ve got no one to blame but yourself. Mortgage under water? You should have thought harder about whether you could really afford that house before you bought it. Trouble paying back your college loans? You should have looked more carefully at job prospects for sociology majors before you took out the loans. Unless of course “you” are “me,” in which case the situation tends to look a bit more complicated.
This has also been a boom time for blame in moral and political philosophy, partially in reaction to John Rawls’s A Theory of Justice (1971), which is widely credited with reviving these fields. Rawls focused not on personal responsibility but on ensuring fair conditions that would create opportunities for everyone to pursue their aims. Within a decade, however, Rawls’s theory was under attack from the left and right for giving insufficient attention to personal responsibility and associated attitudes toward blame. On the right, Robert Nozick’s 1974 Anarchy, State, and Utopia heralded a major libertarian revival, centered on individual rights and individual responsibility. On the left, Ronald Dworkin proposed an alternative to Rawls’s vision of liberal egalitarianism, one that brought personal responsibility into the egalitarian fold. On the one hand, Dworkin argued, our fate should not be shaped by “brute luck”—circumstances, whether social or biological, not subject to our control. But as to anything that results from our choices, blame away. As the philosopher G. A. Cohen said of Dworkin’s argument, it has “performed for egalitarianism the considerable service of incorporating within it the most powerful idea in the arsenal of the anti-egalitarian right: the idea of choice and responsibility.”
Why exactly are we trying so hard to make the world safe for blame? What have we gained and what have we lost in the effort? And is there an alternative?
• • •
The treatment of blame in moral and political philosophy closely tracks cultural and political sensibilities on the subject, and as a result will go far in answering these questions.
In the philosophical literature, arguments in praise of blame divide into two categories, distinguished according to whether free will is regarded as compatible with determinism. Compatibilists—as the name suggests—think the answer is yes: provided certain minimal conditions of voluntariness are met (you must not have been physically coerced into acting as you did, you must have the mental capacity to comprehend your actions, etc.), your actions are freely chosen, notwithstanding that they are predetermined. Incompatibilists think the answer is no: if a person’s actions are determined by antecedent conditions, such actions are not freely chosen.
Some incompatibilists, concluding that our actions are in fact predetermined, are reluctant to assign personal responsibility and blame. I will return to these “skeptical incompatibilists” later on. The category I want to focus on now are libertarian incompatibilists. Like skeptical incompatibilists, they believe that free will is incompatible with determinism. But they are libertarian incompatibilists because they reject determinism in favor of the view that we freely choose our actions. And, having stipulated that we are blameworthy if and only if we freely choose our actions, they conclude that we are blameworthy.
But what is the requisite sense of free will—of our actions not being determined by antecedent conditions—that makes someone blameworthy? And do we in fact have free will in that sense?
For the metaphysician, the theoretical possibility that one could have acted otherwise in some alternative world may suffice to establish free will. But if the question is whether we should hold a real-life Smith blameworthy in this world, one would think that the requisite sort of free will is not metaphysical but practical: When all is said and done, how plausible is it to think that Smith could have acted differently?
To take an all too frequent scenario, suppose that Smith grew up in a neighborhood where drug dealing was the most common form of gainful employment. He was raised by a single mother who was a cocaine addict, and by the time he was twelve was supporting his family by selling drugs. When he was seventeen, he got caught up in a drug deal gone bad, and in the altercation that ensued, he shot and killed the buyer.
How should we think about Smith’s level of moral responsibility? Is there some magical moment at which Smith was transformed from the victim of his circumstances to the author of his own story? If so, when was it? What can we realistically expect of someone who finds himself in Smith’s circumstances with Smith’s history and biological endowments? And what is to be gained—and what lost—by adopting social policies that expect more? Given the high stakes of public blame these days, one might hope that libertarian incompatibilists would take these questions seriously. But most have simply assumed that whatever kind and degree of freedom is required for moral responsibility, all of us, except for a small class of “abnormal” people, have it once we reach seventeen years of age.
The reality is that we are all at best compromised agents, whether by biology, social circumstance, or brute luck. The differences among us are differences of degree that do not admit of categorical division into the normal and the abnormal. A morally serious inquiry into the requisite meaning of free will needs to face some basic facts about this society—for starters, that in the United States parental income and education are the most powerful predictors of whether a three-year-old will end up in the boardroom or in prison; that most abusive parents were themselves victims of abuse and neglect; that the norms of one’s peer group when growing up are powerful determinants of behavior; and that traits of emotional reactivity and impulsiveness, which have a large genetic component, are among the more robust predictors of criminal behavior. Such an inquiry would also need to address what evidence would suffice to conclude that Smith could have behaved differently. Is it enough that someone in a similar situation once pulled herself up by her own bootstraps? That the average person does? And how can we be sure that the situations are in fact similar in relevant ways?
Libertarian incompatibilism, in short, hangs profoundly consequential judgments on the insubstantial hook of an abstract possibility.
Compatibilism, in contrast, dispenses with these uncomfortable questions about the existence of free will by dispensing with any robust requirement of free will. Even if conduct is determined by antecedent conditions, the compatibilist argument goes, people nonetheless are free in other ways that suffice to make them blameworthy for their actions.
The compatibilist position has been around for a long time, with the role of determinism played variously by fate, luck, the gods, God, and social and biological forces. Jonathan Edwards, the great 18th century New England preacher, arguably had the hardest compatibilist hand to play. His cards included the Calvinist doctrine of predestination (a take-no-prisoners version of determinism) and the Calvinist doctrine of sin (a take-no-prisoners version of personal responsibility). But he played the hand he was dealt. In his 1754 essay “Freedom of the Will,” he offered the following grand equivocation: even if we do not will as we will (that is, do not choose what we will to do), we do as we will, and the latter suffices to justify God’s dangling us like spiders over the pit of hell in the event that our actions do not entirely please Him. In short, what matters is not how we came to possess a sinful desire, but that we had it and acted on it.
To the modern reader, Edwards’s argument is likely to seem too clever by half and the entire compatibilist enterprise a little baffling. Why are we knocking ourselves out to make a deterministic world safe for blame? If we really believe someone could not have done other than she did, might we not want to take a different tack altogether?
But the majority of contemporary philosophers writing on the subject are compatibilists. And many have offered what is essentially Edwards’s grand equivocation, updated for modern sensibilities. Character or attitudes have replaced God as the forces that determine what we will, and the two halves of Edwards’s equivocation—willing as we will and doing as we will—go by different names. But the basic argument is the same. In T. M. Scanlon’s words:
The lack of freedom that would be entailed by a general causal determinism need not [rule out responsibility and blameworthiness]. Even if our attitudes and actions are fully explained by genetic and environmental factors, it is still true that we have these attitudes and that our actions express them.”
These days compatibilism is mostly the project of the left-liberal philosophical establishment, and, not surprisingly, has been given a kinder and gentler face. No more dangling over the pit of hell; indeed, in some versions, the consequences are no worse than the deliberate withdrawal of trust and friendship from those we believe have wronged us. But its indigestible core is unchanged: we are blameworthy for doing what we could not help but do.
That indigestible core is plainest to see when fate enters the scene not to determine what action we choose, but to determine its consequences—that is, when simple bad luck affects the outcome of our choices. Consider the following scenario. A bus driver is following his accustomed route, with all due care. A young child darts out in front of the bus. The driver, who does not see her and could not have seen her in time to stop, hits and kills the child. We may blame him for what he did, but in what sense is he blameworthy?
Focusing on such an unlucky outcome allows us to strip out two common distractions in discussions of compatibilism. The first is lingering doubt about free will. When a person makes a poor choice—say, the choice to drive recklessly—it is hard for us not to think that he really could have acted other than he did, if only he had tried harder. That thought often insinuates itself into compatibilist arguments, making the indigestible core go down more easily than it deserves to. In contrast, we have no difficulty believing that, having committed to a course of action, a person may—like the bus driver—have no control over the consequences.
The second distraction is special concern for antisocial conduct—that is, conduct that, whatever its consequences, we wish no one would engage in. Driving recklessly is one example. But, far from acting wrongly, the bus driver in our hypothetical scenario acted just as we would have him act. Someone had to drive the bus; he did the job and did it prudently. What more do we want from the guy? Why on earth should we blame him for doing what we would have had him do, just because things turned out badly?
The answer most compatibilists have given is: because that’s the way people are. We just do that sort of thing. Here is Thomas Nagel’s famous version of the argument:
It is tempting in [cases of decision under uncertainty] to feel that some decision must be possible, in the light of what is known at the time, which will make reproach unsuitable no matter how things turn out. But this is not true; when someone acts in such ways, he takes his life, or his moral position, into his hands, because how things turn out determines what he has done. . . . That these are genuine moral judgments rather than expressions of temporary attitude is evident from the fact that one can say in advance how the moral verdict will depend upon the results.
It may be predictable that we will blame others for the bad consequences of their prudent actions, although I think that response is less widespread and more amenable to reason than Nagel’s observation suggests. But the predictability of the response does not establish that it is a “genuine moral judgment” about the blameworthiness of the person as opposed to a pre-reflective emotional or psychological expression of upset at the consequences of what they have done. To establish the former requires a different sort of argument, one that I doubt can be made. If it can’t, then the claim that “how things turn out” determines the morality of “what one has done” simply raises hindsight bias to high moral principle.
Instead of defending the proposition that we are blameworthy for actions or consequences we could not control, many compatibilists have simply done away with the requirement of blameworthiness. More precisely, they have said, in essence, that our ordinary practices of blaming people settle who is blameworthy. This is at least suggested by Peter Strawson’s “Freedom and Resentment” (1962), a classic discussion of compatibilism that helped to launch its modern revival.
This account of blameworthiness leaves us with no vantage point from which to distinguish, say, our annoyance when our friend forgets to pick us up at the airport from the rage of a lynch mob demanding vengeance for a rape they believe, with no evidence, their quarry committed—unless of course the intensity of blame establishes the relative blameworthiness of its object, in which case the innocent victim of the lynching is much the greater sinner.
I do not mean to suggest that Strawson or his followers embrace any such perversities, or that they lack the resources to distinguish among instances of blaming. Their project is, at root, a humane one. The aim is to insist that we are not just thinking reeds, that part of what it means to be human—and to relate to others on human terms—is to react to terrible losses or antisocial acts with anger, blame, and other negative emotions. But it is a long way from that observation to the conclusion that such “reactive attitudes,” as Strawson called them, are at the core of our humanity and at the heart of our relationships of mutual recognition and respect. They can be part of our nature without being the better angels of it.
Earlier I mentioned a third position on the issues of determinism, free will, and moral responsibility: skeptical incompatibilism. The skeptical incompatibilist agrees with the libertarian that we are blameworthy for our actions only if we have free will in the requisite sense (the incompatibilist part). But, contra the libertarian, the skeptic concludes that we don’t have the requisite free will, or at least there is no persuasive evidence that we do. Although a minority view, skeptical incompatibilism has many eloquent defenders in contemporary moral philosophy. I have trouble seeing the case against it. At least, I have trouble seeing how libertarian incompatibilism or compatibilism could be regarded as serious contenders given the empirical challenges to the former and the normative challenges to the latter.
• • •
Why, then, have so many thoughtful people invested so much intellectual energy in making the world safe for blame? Here are some possible explanations.
(i) We can’t not believe in free will, and hence in moral responsibility, because each person’s daily experience of life is as an agent. Our experience is—to use Jonathan Edwards’s terms—that we do not merely do as we will, but that we also will as we will.
Edwards proved himself an astute psychologist as well as a brilliant metaphysician when he urged this argument on his fellow Calvinists. Wouldn’t people recoil from the idea that God would dangle them over the pit of hell for what they did, even though He made them do it? Edwards saw no cause for worry, because most people are never really going to focus on the “God made me do it” part:
The common people do not ascend up in their reflections and abstractions to the metaphysical sources, relations and dependencies of things, in order to form their notion of faultiness or blameworthiness. They do not wait till they have decided by their refinings, what first determines the Will. . . . The idea which the common people, through all ages and nations, have of faultiness . . . . [is] a person’s having his heart wrong, and doing wrong from his heart. And this is the sum total of the matter.
The fact that we are all instinctive libertarians has given libertarian incompatibilism a free pass on the empirical front. If those instincts are impossible to dislodge—if they are the firm deliverances of ordinary experience—then some accommodation must be made. But if the predisposition to blame is no more than an instinct and habit, the argument for accommodation is not a moral one.
(ii) Even if conduct is not blameworthy, blame is an indispensable tool to control antisocial behavior. This justification does not rest on the moral desert of the party we blame. It rests on the social benefits that flow to the rest of us from locking up the morally blameless and throwing away the key. Those who wish to rely on it have a moral obligation to show that such benefits are great enough to justify the costs we are imposing on the morally blameless, their families, and their communities. In the current American criminal justice system, or the current American version of giving every child an equal opportunity to succeed in life, it is preposterous to think we have come close to meeting that test.
More importantly there are tools of social control that are directed specifically at harm reduction. The point of such tools is not to coddle criminals, or to deny their accountability or volitional capacities. It is to reduce future harm at a tolerable cost to all of us, wrongdoers included, by influencing wrongdoers’ future choices through rehabilitation, more carefully calibrated deterrence, and, when necessary, isolation from society. There are serious disagreements about whether harm-reduction policies have worked in the past, though there are no serious disagreements about the failures of mass incarceration. But we have some evidence that interventions can work if they are evidence-based and carefully tailored to the problems we are trying to fix. Since, unlike retribution, such tools are designed for the purpose of harm reduction, we should hardly be surprised if they do a better job of it.
(iii) Blaming others is a way to show respect for them. This very Kantian argument is at the core of much of the contemporary academic literature in praise of blame.
In the hands of hardcore retributivists, the argument has a decidedly Dickensian cast. To quote one proponent, when we punish someone, we respect his “fundamental human right to be treated as a person” by “permitting him to “make the choices that will determine what happens to him” and then respecting his “right to be punished for what [he has] done.” Lord save us all from such respect.
In the hands of modern-day compatibilists, the stakes of the argument are much lower, and the delivery not so redolent of the Dickensian workhouse. Blaming others, Jay Wallace tells us, “is a way of taking to heart the values at the basis of morality” and of taking seriously “relations of mutual recognition.” Refusing to blame others, in contrast, “involves an attitude of superiority toward the person in question (something like the attitude of a parent toward a very young child) and thus represents a failure to take that person seriously as a participant in the relationship,” according to Scanlon.
A genuinely humane impulse is at work here. When we expect too little of others, we do in a certain sense fail to treat them as equals, and we also limit the kind of relationship we can have with them. But the argument presupposes that there are only two standpoints from which we can evaluate others: the subjective standpoint, in which we are enmeshed in a relationship and therefore in thrall to reactive attitudes such as blame; and the objective standpoint, from which we dispassionately evaluate others as fit objects for rehabilitation or instrumental social control, or as unfit candidates for friendship.
There are other possibilities that neither hold us hostage to reactive attitudes such as blame nor require us to view others from a position of moral superiority or indifference. We could begin by extending to others the interpretive generosity we would wish for ourselves were we standing in their shoes. Here is Erin Kelly’s eloquent account of what such a standpoint might entail:
While it seems to me that we are not morally required to enter into a wrongdoer’s perspective enough to appreciate the difficulty of the obstacles that led her to falter, the possibility of a compassionate recognition of the reasons for a person’s moral failures humanizes relationships and opens possibilities for understanding, forgiveness, and an honest reckoning with faults we might share.
Which kind of respect would you rather have?
In either the hardcore or softer versions, the “blaming you is how we show respect for you” argument runs into a serious PR problem when applied to bad actors whose moral agency is undeniably compromised: young children, the mentally ill, those in the throes of dementia, the severely retarded, and others who are commonly regarded as morally blameless. Retributivists and compatibilists have dealt with the problem by making an exception for these abnormal cases, acknowledging that the absence of meaningful moral agency renders such actors inappropriate objects of blame.
For the compatibilist, that concession is deeply problematic. Once compatibilism allows for the possibility that some forms of compromised moral agency excuse bad conduct, there is no logical stopping place short of incompatibilism. If a schizophrenic can introduce evidence that he is not a full moral agent, why not someone in the grips of a major depression, or impulsive anger, or drug addiction? A teenager growing up in gang territory, whose physical safety and social inclusion depends on choosing sides? Of course, the compatibilist may observe that we do commonly distinguish among different factors that compromise agency, allowing excuses in some cases (schizophrenia) but not in others (impulsive anger). But that observation, like Strawson’s view, merely describes current practice; it does not justify it.
For the libertarian incompatibilist, making an exception for the abnormal isn’t problematic in principle: we needn’t have free will always to have it some or even most of the time. It is, however, troublesome in practice. Nowhere is this clearer than in our current criminal justice system. Of the more than 2 million Americans currently incarcerated, 15 percent show symptoms of psychosis (delusions, hallucinations, etc.); another 25 to 40 percent have serious non-psychotic mental disorders. And this does not even get to the severe deprivation most prisoners faced growing up. But most libertarian incompatibilists see no reason to inquire into these or any other realities of our criminal justice system before concluding that we are finally giving criminals what they deserve.
(iv) Blame is here to stay, and if we can’t beat it, we might as well do what we can to civilize it.
Such fatalism is understandable. But there are lots of reasons to reject it.
First, while we experience many feelings toward others that contain some element of reproach, the feelings are more nuanced, more variable, and more mutable than the public face of our blame fest would suggest.
Public reactions to wrongdoing have been studied most extensively in the context of crime. Researchers have found that peoples’ evaluations of serious wrongfulness vary significantly across social conditions and individuals. Tellingly, the more information people have about the context of the crime, the person who committed it, and the circumstances he or she came from, the more nuanced are their views of moral responsibility. Peoples’ intuitions about appropriate punishment are as likely to be responsive to future-directed utilitarian concerns as past-oriented desert-based ones. There is little consensus about the absolute levels of punishment appropriate for different forms of wrongdoing, and, according to a number of recent studies of public opinion about punishment, the same people who describe current punishment policy as insufficiently punitive recommend replacing it with policies that are significantly less punitive.
The same is true for most of us in the personal realm. Even as we experience anger toward those who have harmed us, we are capable of fellow feeling as well. (He had a bad day; this is an issue he has a very hard time with; etc.) Many people placed in the position of the parents whose child was killed by the blameless bus driver would be capable of not blaming him—indeed, of sympathizing with him, knowing that for the rest of his life he will reproach himself, as others will reproach him, for an outcome for which he was in no sense blameworthy. It doesn’t take a saint or an emotional paralytic to feel that way. What it takes is empathy: the capacity to look at someone else’s life as we hope others will look at ours.
The fact that we alter our judgments of blameworthiness as we acquire greater knowledge of the person and the context in which she acted should put to rest any thought that our blaming practices are naturally immutable, or even recalcitrant. An hour listening to the average lifer in prison or the average at-risk teen talk about his or her circumstances, and most Americans would never view those groups in the same way again. Unfortunately, most of us will never spend that hour. Everything we know about people outside our social circles—assuming we know about them at all—is mediated by others (politicians, pundits, the media) who have every incentive to provide whatever information will elicit the emotional response they are looking for (anger, blame, sympathy, sorrow, etc.). It is hard to break out of that echo chamber, but it is possible.
Which brings me to the second reason to reject the fatalistic claim that blame, as we currently practice it, is not going away. Change always seems impossible—until it doesn’t. After 40 years of policies that have relentlessly ratcheted up punishment, the direction has shifted slightly in the last few years. New York and Massachusetts repealed their mandatory minimum sentences for drug offenses. California repealed the most egregious elements of its three-strikes law. The changes in New York and Massachusetts were spurred by budgetary crises and worked out by Republican and Democratic legislators in a manner that gave both groups political cover. In California, change came via a 70 percent majority in a popular referendum. The primary motivation voters cited for scaling back the three-strikes law was not money but rather a belief that the law was unfair. Both developments are encouraging, in different ways—the first, because it suggests the possibility of détente in the political arms race to prove which party is tougher on crime, the second because it suggests that political grandstanding on the subject may finally be losing its audience.
The final reason for cautious optimism is that we have gotten nothing from our 40-year blame fest except the guilty pleasure of reproaching others for acts that, but for the grace of God, or luck, or social or biological forces, we might well have committed ourselves. Our schools are broken, a new generation of kids has been lost, our prisons are crammed with petty offenders whose lives we have ruined in the name of a war on drugs that has been a total failure. And judging from the current mood of the country, the guilty pleasure of blaming others has not proved all that pleasurable.
I doubt there will be a groundswell of support any time soon for the view that others may not, after all, be to blame for the mess they (and we) are in. But the fact that we have gotten so little in return for our blame mongering at least opens up the possibility that people would be receptive to a new approach. The next time something goes terribly wrong, suppose that instead of immediately asking who is to blame, we were to ask: How can we fix this problem? Fixing problems is costly. But as we have learned from the past 40 years, so is not fixing them. In the long run, most of us stand to gain by changing the national attitude toward blame. Doing so won’t magically transform the world. But it will increase the odds of a better life for many, if not most, of us. That seems like a more-than-even trade for giving up a sense of self-righteousness that none of us has earned.
Editors' Note: This forum appeared in the July/August 2013 print issue.
Photo: Nelson Vargas
1. What is Blame?
To begin, note that almost all philosophical discussions of blame ignore (or mention only to set aside) the form of blame sometimes characterized as causal or explanatory responsibility (Kenner 1967; Hart 1968; Beardsley 1969). It is this notion of blame that is at stake when we say that Hurricane Hugo is to blame for the destruction of Charleston's harbor, or that the cat is to blame for knocking over the vase. Theorists contrast this sense of “blame” with the sort of interpersonal blame that, for example, one gives up when one forgives. (As Pamela Hieronymi has pointed out (2001), forgiveness in fact requires not giving up one's judgment that the other person is explanatorily to blame.) But just what the relation is between causal blame and interpersonal blame is an important question that has not been well-explored. Nevertheless, in this entry the focus will be on blame as a response to moral agents on the basis of their wrong, bad, or otherwise objectionable actions or characters.
A taxonomy of theories of blame could be organized along a number of different dimensions, depending on one's purposes. Consider just two of these possible dimensions. First, we could categorize theories of blame according to the content of blaming attitudes. On this way of dividing up things, the view put forth by Pamela Hieronymi (2004)—which holds that the force of blame is located in judgments of ill will—would be categorized with that of R. Jay Wallace (1994, 2011)—which holds that blame is an emotional response to ill will displayed in others' action. Alternatively, we could categorize theories of blame according to those psychological states or dispositions that are identified with blame. This second way of dividing things up emphasizes an important difference between Hieronymi's and Wallace's theories: for Hieronymi, blame can be identified primarily with a judgment, but for Wallace, blame is primarily an emotion.
Of course, any scheme for categorizing theories has its advantages and disadvantages, and by selecting one of these taxonomies, one necessarily emphasizes certain aspects of blame while ignoring other aspects of blame that might be equally important. So in some sense, any preference towards one way of cutting things up over another would be arbitrary. Nevertheless, the taxonomy to follow is consonant with much of the literature in classifying theories according to the activity or mental state (or the function of the activity or mental state) that is identified with the blame. According to this way of carving things up, we get four categories: cognitive, emotional, conative, and functional accounts of blame.
1.1 Cognitive Theories of Blame
Cognitive theories of blame share the idea that blame is fundamentally a judgment or evaluation that we make about an agent in light of her attitudes or her actions. One of the earliest cognitive theories of blame is due to J. J. C. Smart (1961), who develops his analysis of blame indirectly, since he begins by distinguishing between praise and dispraise (rather than with the more natural distinction between praise and blame). According to Smart, to praise or dispraise an individual is simply to grade her as a member of a particular kind. And as Smart says, this sort of grading is no different than the sort of grading involved in judging one apple to be better than the others on the grocer's stand. Crucially, Smart notes that though you might dispraise a young philosopher for his poor writing in a letter of recommendation, you are not thereby blaming him for it. Thus for Smart, blame is distinct from dispraise. Unlike dispraise, blame involves more than merely grading someone's actions or character (morally), since blame carries with it the implication that the person is responsible for her action or character. Blame, then, is a negative evaluative judgment that implies responsibility.
In a similar vein, Gary Watson (1996) has suggested that there is an evaluative form of blame connected with what he calls the “aretaic perspective”. To blame someone in this way is to judge that she has failed with respect to some standard of excellence (arête). It is also to insist that the agent is responsible for her action in the sense that the action is attributable to the agent—it represents her evaluative standpoint, her practical identity, what she “stands for” (Watson 1996). And like Smart, Watson recognizes that it is possible to make such a judgment dispassionately. Thus, on the grading and evaluative theories of blame developed by Smart and Watson, there is nothing about blame that requires a blamer to be conatively or emotionally exercised in any way. However, unlike Smart—who identifies blame with a form of grading that implies moral responsibility—Watson does not take aretaic blame to be a general analysis of blame. Rather, for Watson, aretaic blame is but one way among many that we blame others for their actions.
Unsurprisingly, many have resisted thinking of blame as a form of grading. For example, T. M. Scanlon (1986) objects to Smart's overly descriptive characterization of moral grading, which, by Scanlon's lights, fails to account for the characteristic force of blame. For although the force of blame is purportedly tied to the (often painful) normative burdens that one accrues when one is blamed, simply being graded poorly is not in itself such a burden. So blame as grading plus responsibility seems insufficient. Nevertheless, a number of contemporary accounts of blame retain the core idea in Smart's (and Watson's) account that blame is a kind of evaluative judgment. But what sort of evaluative judgment will do? Well, it can't just be any old judgment that describes the whole person as such (as is the case in Smart's grading theory). As a result, many theorists have identified blame with judgments that essentially implicate how our moral or practical selves were involved in the production of action. Only this, one might think, can explain the special force of blame. After all, as Scanlon puts it, “given that most people care about” their moral selves, judgments that implicate these aspects of a person are not, as Scanlon puts it, “mere descriptions” (1986: 170). Thus, for example, Michael Zimmerman (1988) and Ishtiyaque Haji (1998) have argued that to blame someone is to judge that in virtue of their attitudes, actions, or character, they have a stain on their moral self or a black mark in their moral ledger. As Zimmerman puts it, when we blame someone, we judge
that there is a “discredit” or “debit” in his ledger … that his “moral standing” has been “diminished”. (Zimmerman 1988: 38)
One need not endorse the idea of a moral ledger in order to hold a cognitive theory of blame, however. T. M. Scanlon (1986) and Pamela Hieronymi (2004) both articulate cognitive accounts of blame, where the judgment in question is a judgment that someone has shown us (or another) ill will. Since we care deeply about other people's judgments about the quality of our wills, this judgment can also carry the distinctive force of blame. (See also Kekes 2009 for a view of blame that slants toward the cognitive side of things.)
Despite a number of supporters, there are many who are less sanguine about the prospects of a purely cognitive account of blame. One potential problem for cognitive accounts is that they risk conflating blaming with judging blameworthy (Kenner 1967; Coates and Tognazzini 2012). After all, it seems quite possible to judge, for example, that another has displayed ill will in his action or that he has a mark against his moral ledger (and so, judge that he is blameworthy), without actually blaming that individual. The co-conspirator's recognition of the wrongness of his partner's criminal activity might, in fact, underlie his admiration for his partner's skillful execution of a heinous crime that most of us couldn't stomach. The fact that the same judgment could elicit such different responses (repulsion and resentment in those of us who are committed to the values of morality and admiration in those who are not) suggests that the judgment alone cannot constitute blame. More recently, Hanna Pickard (forthcoming) has argued because it is possible to blame others irrationally (i.e., to blame others even when we know that they are not really blameworthy for their actions), the judgment that another is blameworthy, or that she has shown ill will or disregard in her action is not necessary for blame. It looks, then, that judgments of the sort discussed above are neither necessary nor sufficient for blame.
A further problem for cognitive accounts is one suggested by Gary Watson (1987). According to Watson, attempts to identify or reduce blame to its cognitive components (recall that although Watson thinks that aretaic blame takes the form of judgments, it does not exhaust the phenomenon of blaming) make it seem
as though in blaming we were mainly moral clerks, recording moral faults … from a detached and austerely “objective” standpoint. (1987 as reprinted in Watson 2004: 226–27)
But surely blame issues from the perspective of a participant in human relationships, one in which we are not merely observing the moral order but are actively involved in a moral community. These sorts of considerations form the basis for emotional theories of blame.
1.2 Emotional Theories of Blame
Despite the fact that P. F. Strawson's “Freedom and Resentment” (1962) contains little sustained discussion of blame as such, many take it to be the contemporary genesis of emotional theories of blame. According to Strawson, our status as morally responsible agents is grounded in the non-detached attitudes and emotions that are (in part) constitutive of ordinary interpersonal relationships. Regarding others as morally responsible agents, for Strawson, is not a matter of judgment but of emotional response. (Of course, according to cognitivist theories of the emotions, emotional states are identified with a suite of judgments (Solomon 1993; Nussbaum 2001). Thus, if the cognitivist theory of emotions were correct, there would be no fundamental differences between cognitive and emotional theories of blame. However, cognitivist theories of the emotions are very controversial, and so we will set this important question aside in order to better understand extant emotional theories.)
R. Jay Wallace (1994) developed this idea into an account of “holding responsible” according to which we hold others morally responsible just in case we experience resentment, indignation, or (in the self-regarding case) guilt as a response to their actions, or judge that such a response would be appropriate. Thus, for Wallace, (a specific subset of) Strawson's “reactive attitudes” are essentially implicated in the stance we take up when we hold others responsible. But though it is possible to take up the stance of “holding responsible” without being emotionally exercised, Wallace stresses (and reiterates this in Wallace 2011) that to actually blame an agent, one must be exercised emotionally.
Of course, Strawson and Wallace are hardly alone in endorsing emotional theories of blame. While these “Strawsonian” accounts of blame focus on the reactive attitudes (particularly resentment, indignation, and guilt), other emotional theories of blame are more inclusive. Susan Wolf (2011), for example, defends an account of blame that emphasizes anger. Macalester Bell (2013a, b) argues for a “hostile attitudes” account of blame that includes the attitude of contempt as a blaming attitude. Consequently, what holds emotional theories of blame together is not widespread agreement over which emotions constitute blame. Rather, it is a shared commitment to thinking that to blame is to respond to others' actions with a negative emotion.
Though it's very plausible that we blame others by responding to their actions with anger, resentment, indignation, or even contempt, there are a number of objections to emotional theories of blame. George Sher (2006) argues that emotional responses are unnecessary for blame. For example, Sher argues that we can blame a loved one without feeling negative emotional reactions. So too, we can blame villains from whom we are temporally distant without any emotional response. The thought here is simply that it is possible to blame Nero for the burning of Rome, even though we do not feel any resentment or indignation towards Nero for his cruelty. In response, defenders of emotional theories might simply argue that despite appearances, without the emotions, one is simply not blaming Nero but instead merely judging blameworthy (see Wallace 1994, 2011). Alternatively, a defender of emotional theories could argue (plausibly, but by no means uncontroversially) that one can be in an emotional state even if one does not experience any felt affect. On this latter point, Elisa A. Hurley and Coleen Macnamara (2010) have argued that on the correct theory of reactive emotions, it is possible to resent or be indignant without experiencing the affective components paradigmatically associated with these emotions.
A second objection to emotional theories of blame might be called the “force objection”. Pamela Hieronymi develops this objection by noting that
an affective accompaniment of a judgment would be a certain unpleasant emotional disturbance … but, the force of blame seems deeper, more serious or weightier. (Hieronymi 2004: 121)
Thus, the normative force of blame must be grounded in the cognitive elements of blaming emotions, since it is these elements that are responsive to and reflect our concern for morality. But if the force of blame is grounded in the cognitive elements of the emotion, then why wouldn't a judgment with the same content constitute an instance of blame? It seems that while emotions might be concomitant with blame, it is the cognitive element—one that can be present even if the blamer is not emotionally exercised—and not the emotion itself that constitutes one's blame.
In response to this sort of objection, Wallace (2011) argues that the reactive emotions are not superfluous add-ons to the judgment, but instead they serve to change the meaning of the judgment, imbuing the judgment with the sort of expressive significance that is characteristic of blame and that would otherwise be lacking from a mere judgment.
1.3 Conative Theories of Blame
Conative theories of blame emphasize motivational elements, like desires and intentions, as essential to blame. Two of the most developed extant theories of blame—those due to George Sher (2006) and T. M. Scanlon (2008, 2013)—fall in this category. And though we will focus on these two theories, other conative theories are possible.
1.3.1 Dispositions Around a Belief-Desire Pair
As already mentioned above, George Sher (2006) is skeptical of emotional theories of blame. However, he is also skeptical about accounts of blame that are merely cognitive. There is more to blame than a mere judgment that an agent has acted wrongly, but one need not be emotionally exercised in order to blame. Sher, thus, is looking for a happy medium between these two widely accepted alternatives.
According to Sher, what must be added to judgments of wrongness is a backwards-looking desire “that the person in question not have performed his past bad act” (2006: 112). But it's not enough that the blamer simply wish that the bad action not have happened; the desire must be one that issues from the blamer's general commitment to morality, since what we really want is that the wrongdoer not have “exercised his own decision-making capacities in a certain way”, and that “he have responded, or that he be disposed to respond, to what we consider a compelling moral reason” (2006: 105). Thus, on the resultant view, when the cognitive component of judging blameworthy is accompanied by this desire, which reflects our general commitment to morality, then we are said to be blaming. (See also Arpaly 2006 and Arpaly & Schroeder 2014 for a similar view, according to which blame requires having a conative orientation “against the wrong or bad” (Arpaly & Schroeder 2014: 161).) Moreover, Sher argues that the belief-desire pair in question is itself the basis of those affective and behavioral dispositions that are commonly associated with blame. For example, a blamer's disposition to feel hostile attitudes like anger towards the agent and to also reprimand, rebuke, and seek apology are to be explained by the presence of the belief-desire pair.
Despite the elegance of Sher's view, it has generated a number of critical replies. Pamela Hieronymi (2008) objects to the link between the belief-desire pair and attendant affective and behavioral dispositions. To her mind, the link is too weak; though she accepts Sher's claim that the belief-desire pair is essentially implicated in one's general commitment to morality, she does not think he has adequately shown that the characteristic dispositions are implicated in the same way. After all, “surely our commitment to morality could be affirmed or clarified in ways that do not involve hostile behavior or reproach” (2008: 25). But if this is correct, then it looks like blame's characteristic dispositions need not be present—even in those who are genuinely and sincerely committed to moral norms. As a result, Hieronymi concludes that Sher has failed to show that blame—which must involve such dispositions—is essentially tied to a more general commitment to morality.
A second objection to Sher's view is due to Angela Smith (2008). Smith rejects Sher's claim that only by adding a desire component to the belief that someone has acted wrongly or badly, can we be said to blame someone. To defend this, she invites us to consider an ordinary case of blame, say the blame we feel for a politician who leads us into a disastrous war. While we no doubt desire that the politician hadn't led us into the war because we are generally committed to morality (and we therefore don't enjoy the suffering of innocents), it is not clear how this desire is itself part of our blame. By Smith's lights, the desire component of the belief-desire pair, like the attendant affective and behavioral dispositions, seems to be something that is above and beyond blame itself. In more recent work, Smith has also argued that in some cases, say in “the reactions of a mother whose son is blameworthy for [a] crime” (2013: 35), the relevant belief-desire pair might be present without blame. Other challenges to Sher's theory include the worry that it is too “sanitized” because it compromises psychological realism by “stripping away [blame's] unsavory features” (McGeer 2013: 166).
1.3.2 Attitude Adjustment in Response to Impairment
T. M. Scanlon (2008) has recently developed an account of blame that represents something of a shift from his earlier, more cognitive, account (see Scanlon 1986). In developing this new account, Scanlon's initial motivation is similar to that of Sher, since Scanlon thinks that an adequate account of blame must fit somewhere between a mere judgment that another has acted in some objectionable way and a sanction (of which expressed reactive emotions are but one paradigm case). But unlike Sher, Scanlon does not think that we can avoid these (putatively) unattractive alternatives simply by supplementing the belief that another has acted wrongly or badly with a desire that they did not so act. Indeed, for Scanlon, the belief that another has acted wrongly is not part of blame at all. Instead, the cognitive component of blame is provided by a judgment that another has acted in a way that impairs meaningful interpersonal relations; this is a judgment of blameworthiness. (For Scanlon, blame is a response to the meaning of someone's actions, rather than the permissibility of those actions. For more on this distinction, see the first three chapters of Scanlon 2008.) But this judgment itself is insufficient for blame (for largely the reasons that Sher takes the belief that another has acted wrongly or badly to be insufficient for blame), so in addition to judging that the agent is blameworthy, blame requires you “to take your relationship with him or her to be modified in a way that [a judgment of blameworthiness] holds to be appropriate” (2008: 128–29). In other words, blaming someone involves not just the belief that he has acted in a way that impairs your relationship with him, but also, that you take yourself to have reasons to revise your intentions and attitudes towards him, and accordingly that you revise these intentions and attitudes on the basis of such reasons.
Like Sher, then, Scanlon has provided an initially plausible account of what it is to blame. But also like Sher, his account has been widely criticized. The most common line of criticism is best summed up by R. Jay Wallace's (2011) slogan that Scanlon's account “leaves the blame out of blame”. More precisely, Wallace argues that
blame has a quality of opprobrium that is not captured by the considerations about the normative significance of impaired relationships that are at the center of Scanlon's approach. (2011: 349; see also Mason 2011)
Susan Wolf (2011) has also argued that in some cases, such as the case of a hot-headed but ultimately loving family, it seems that you can blame another without taking yourself to have impairments in your relationship or attendant reasons to revise your intentions or attitudes towards that person. The characteristic features of Scanlon's interpretation of blame, then, seem to be unnecessary. More recently, Sher (2013) has argued that Scanlon's emphasis on relationships is problematic. After all, many cases of wrongdoing involve strangers—e.g., in most car thefts, the victim does not know the criminal. Nevertheless, it still seems that it is possible to blame those with whom we have no standing relationship. So blame cannot essentially implicate interpersonal relationships. Scanlon, in response (2008, 2013), insists that all rational agents stand in the “moral relationship” to one another. However, whether this kind of relationship is sufficient to explain the blame of strangers is unclear. And indeed, as Sher points out, even if there is some relationship between a victim and the stranger who victimizes her, it's not clear that this relationship plays any role at all in grounding her blame.
1.4 Functional Theories of Blame
Functional accounts of blame are analogous to functionalist theories of mental states or properties. Instead of identifying blame with any particular attitude (like a judgment or emotion) or combination of attitudes (like a belief-desire pair), functional accounts of blame identify blame by its functional role. This way of proceeding leaves open the particular attitude or combination of attitudes that constitute blame. In this way, functional accounts can be more flexible.
According to one functional account of blame, the function of blame is protest. In other words, what we're doing when we blame others is protesting their actions or character. But this, of course, means that perhaps any number of attitudes or combination of attitudes could be present in blame. Pamela Hieronymi (2001), Matthew Talbert (2012), and Victoria McGeer (2013) argue that reactive attitudes like resentment (and the expressions of these attitudes) serve as powerful forms of protest. Angela Smith (2013), on the other hand, argues that when we modify our attitudes and intentions as Scanlon envisions, but do so as a form of protest, then we are actually blaming. In other words, for Smith, it's not enough that we modify our attitudes and intentions; the modification in question must serve a particular function, namely that of protest, to count as an instance of blame. And in order to count as a protest, it need not involve any particular emotional state. (See Franklin 2013 and Houston 1992 for more on the way in which blame allows us to stand up for our values.)
Of course, there might be other functions of blame: to express or communicate condemnation or disapproval, for example. Michael McKenna (2012, 2013) has argued for such an account. In fact, he has claimed that blame is conversational, and thus functions to continue a conversation started by the blamee's wrongful action. In particular, McKenna claims that the reactive attitudes and their expressions serve this function. Antony Duff has proposed a similar understanding of the aim of blame, according to which it is
an attempt to communicate to the wrong-doer a moral understanding of his wrong-doing; to bring him to recognise his guilt and repent what he has done. (1986: 70)
(See also Macnamara 2011 on the communicative nature of holding responsible more generally.) In a similar vein, Christopher Bennett (2013) has argued that blame functions symbolically to express our disapproval. Like Smith, he develops this account by supplementing Scanlon's theory of blame, but it seems that one could adapt Bennett's expressive apparatus even outside of a Scanlonian framework.
Because of their relative newness to the scene, there is not much criticism of these views. But there are at least two sources of concern for those theories that take protest to be the function of blame. First, it's not clear that protest is independent of blame, such that one could specify what it is to protest without appealing to blaming attitudes. But if this is so, it's not clear that appealing to the notion of protest will help us clarify the nature of blame. Second, protest seems paradigmatically expressed. Indeed, it's hard to make sense of unexpressed protest. Do workers protest unfair labor conditions simply through their beliefs or attitudes? Or must they make such beliefs and attitudes known? And if it is the latter, then it's not clear that protest could be the function of blame. After all, not all blame is expressed.
2. When is Blame Appropriate?
How this question gets answered will depend on four variables: (i) the person who is being blamed, (ii) the person who is blaming, (iii) the nature of the blaming interaction, and (iv) the precise sense of “appropriate” at issue. A better way to put the question, then, is perhaps to ask when it is appropriate for X to blame Y, keeping in mind that emphasizing different aspects of this relationship will produce a different contrast, and hence a different question. (Also keep in mind, of course, that X and Y might be one and the same person.) Moreover, it's important to recognize the connections between the questions of what blame is and when blame is appropriate, since the propriety conditions of a judgment are plausibly thought to be distinct from the propriety conditions of overt forms of blame like rebuke. Consequently, because we are not here endorsing a particular theory of blame, our characterization of the norms in question will operate at a level of abstraction that floats free of substantive commitments concerning the nature of blame.
Begin by considering potentially relevant facts about the person who is being blamed. A natural answer to the question of when blame is appropriate is to say that blame is only appropriate when the person blamed is in fact blameworthy. This may sound at first like an unhelpful tautology—after all, what could it mean to be worthy of blame if not simply that you can be appropriately blamed?—but the emphasis on worthiness is meant to draw attention to the fact that it's only appropriate to blame a person when she has earned it or deserves it. That is, only when certain facts about her are in place. Which facts? What does one have to do to earn blame?
2.1.1 Moral Agency
Being to blame (i.e., causally responsible; see Beardsley 1969 and Kenner 1967) is not sufficient for being blameworthy because often, the best or most salient casual explanation doesn't even involve a moral agent at all. Earthquakes and mosquitoes can be to blame for various negative outcomes, but neither can be blameworthy because neither can, as Gary Watson puts it, “act effectively and competently in moral matters” (2013a: 3322). Only certain creatures are even candidates for blame in the first place, and though it is a matter of some controversy which precise capacities are required, the list certainly includes the capacities for reflection, deliberation, decision-making, and self-determination. But earthquakes and mosquitoes are the easy cases; the harder cases are children and psychopaths, individuals who haven't (or haven't yet) developed an understanding of or an appreciation for moral norms. These individuals, it seems, can still act in morally significant ways—indeed, in ways we would naturally describe as cruel and even evil—but whether they can earn moral blame (as opposed merely to giving us good reason to protect ourselves from them) is a vexed question (see Watson 2011 for insightful discussion). But regardless of how one answers that question, it is widely accepted that potentially blameworthy agents must be capable of reflecting upon, reasoning about, and executing a decision about how to behave. If someone lacks these capacities, he is exempted from blame.
In addition to having the general capacity for practical reasoning, however, it is often thought that an individual is appropriately blamed only if he has (and, on the occasion, exercises) free will. The excuse “I couldn't help it” or “I was forced to do it” is often enough to render blame inappropriate, so it's a natural thought that someone can only be blamed for those things that he could have helped, or wasn't forced into—in other words, for those things that he chose of his own free will. (Of course, this is primarily a condition applied to actions for which one is thought to be blameworthy. Taking seriously the possibility that we can be blameworthy for our attitudes as well might naturally lead one to downplay the importance of free will. See, for example, Smith 2005.) Typically, free will is thought of as a sort of control: as the ability to control (by selecting) which of two possible futures obtains, for example, or as the ability to control (by guiding) one's actions in light of one's considered judgments about what one ought to do. (See van Inwagen 1983; Fischer 1994; Watson 2003.) And of course it's a difficult question whether control of the right sort is compatible with determinism; hence it's a difficult question whether blame would ever be appropriate in a deterministic world. There are less sweeping threats to freedom, however. We are all vulnerable to coercion, manipulation, situational pressures, and varying degrees of temptation or compulsion, and the extent to which these factors rob us of our freedom is the extent to which we may be excused (though not exempted) from blame.
2.1.3 Moral Responsibility
If you add the capacity for practical reasoning to the right sort of capacities for control (which will likely include not just volitional capacities but cognitive capacities, too), you end up with a morally responsible agent—that is to say, an individual who has the capacities that render her a sensible target of blame (see Fischer & Ravizza 1998). If she exercises those capacities in order to bring about some negative outcome, then she is morally responsible for that action—that is to say, she is a sensible target of blame for that action (she is neither exempted nor excused from our blaming practices).
There are further subtleties here, but they are inessential to the main point, which is simply that most theorists think that it is appropriate for X to blame Y only if Y has certain capacities for control, practical reasoning, moral understanding, etc., and exercised them on the occasion in question. (One of the subtleties is that even if an agent satisfies all the relevant control conditions, she may still fail to be responsible if she fails to meet an independent epistemic condition. Non-culpable ignorance of the consequences of one's actions seems to excuse as much as lack of control. See Ginet 2000 and Mele 2011.) Likewise, most theorists think that if Y has and exercises these capacities, then Y is blameworthy—that is, Y has earned blame. But this is all compatible with saying that blame is nevertheless all-things-considered inappropriate. To see why, let's turn now to facts about the person who is blaming.
2.2 Moral Standing
Even if some agent is blameworthy, it's not the case (or, at least, not always the case) that everyone can blame him. As Roger Wertheimer notes,
some matters—like other folks' intimate intrafamilial relations—may be none of your business, not your affair, no (proper) concern of yours, so, whatever your evidence and emotions, it is not your place to bear ill will. (1998: 499)
G. A. Cohen (2006: 118) echoes the sentiment from a different perspective:
[Moral] admonition may be sound, and in place, but some may be poorly placed to offer it. When a person replies to a critic by saying: “Where do you get off criticizing me for that?”, she is not denying (or, of course, affirming) the inherent soundness of the critic's criticism. She is denying her critic's right to make that criticism, in a posture of judgment.
The general idea here is that there may be facts about the person who is blaming that undermine her standing to blame. It's not her place, she isn't well positioned, she doesn't have the authority, and so on. (Marilyn Friedman (2013: 272) puts the point nicely by saying that not everyone is blamerworthy.) But there are several different ways in which one's standing might be undermined.
At least sometimes, we blame others with the aim of getting them to see the error of their ways and change their behavior in the future. One sure way to fail at this is to be guilty of the very same (or a relevantly similar) transgression as the one you are condemning. The hypocritical blamer is perhaps the paradigm example of someone who lacks the moral standing to blame. And it's not just that hypocritical blame won't be effective—though that's also true (see Dworkin 2000)—it's that it seems positively inappropriate (and would remain so even if it were effective). Even the genuinely blameworthy agent, it seems, can respond to the hypocritical blamer by saying, “Look who's talking”, and this retort has a “silencing” effect (see Cohen 2012).
But what precisely is the problem with hypocritical blame? The answer to this question will likely depend on the nature of blame. R. Jay Wallace, for example, who advocates a Strawsonian account of blame, explains the problem with hypocrisy by an appeal to the underlying commitments of the reactive attitudes. For Wallace (2011: 326), “blame carries with it a kind of practical commitment to critical self-scrutiny”, a commitment that the hypocritical blamer fails to live up to. Given that “we all have an interest in being protected from the kind of social disapproval and opprobrium that are involved in blame”, the hypocritical blamer—as long as he isn't also blaming himself, in which case he might not count as hypocritical—treats his own interest in avoiding blame as more important than the interest of the target of his blame. As Wallace puts it (2011: 328): “This offends against a presumption in favor of the equal standing of persons that I take to be fundamental to moral thought”. Thus, for Wallace, the problem with hypocritical blame is that it is morally wrong (and thus inappropriate even if the target is blameworthy).
For T. M. Scanlon, on the other hand, who takes blame to be an adjustment of attitudes in response to an impaired relationship, the problem with hypocritical blame is that it distorts the facts. The adjustment of attitudes and expectations that constitutes blame purports to be a response to some way in which the person blamed has impaired his or her relationship with the blamer. But, as Scanlon says (2008: 177),
There is something false in [the hypocritical blamer's] suggesting that it is [the blamee's] unwillingness to act in ways that indicate untrustworthiness that impairs [their] moral relationship.
Rather, according to Scanlon, the past attitudes and actions of the hypocritical blamer already impaired the relevant relationship. Thus, for Scanlon, we might say that the problem with hypocritical blame is that it is unfitting (or untrue to the facts).
Other accounts of the way in which hypocritical blame is inappropriate are, of course, possible. G. A. Cohen (2006), for example, seems to favor an account that considers blame as a speech-act, and hypocrisy is then one way in which the illocutionary force of the speech-act can be compromised. One might also take a page from Stephen Darwall (2006), and maintain that in blaming, we are attempting to give the blamee a distinctively second-personal reason to apologize and/or change her behavior, and that hypocritical blame fails to communicate this distinctive reason. (See also Radzik 2011.)
Work on this topic is still very much in its infancy, and there are many more questions in need of exploration. (See Szabados and Soifer 2004 for a book length treatment of the ethics of hypocrisy.) For example, need someone be guilty of the very same transgression in order to be a hypocrite in the relevant sense? And how can an erstwhile hypocrite regain his lost standing? Is it sufficient that the blamer who is guilty of a similar transgression feels guilty about his own transgression and resolve to do better in the future? But even in that case, isn't there still something inappropriate or otherwise untoward about the blame?
To charge someone with blaming hypocritically is to allege that he is blaming (or, at least, pretending to blame) in response to transgressions similar to those (or perhaps type-identical to those) that he himself has committed in the past. A somewhat related charge, but worth distinguishing, is the claim that the blamer is somehow objectionably involved in the very act that he is, at this very moment, condemning. This is to charge the blamer with complicity, and such a charge might take many forms. G. A. Cohen (2006: 126) gives a nice sample: “you ordered me to do it, you asked me to do it, you forced me to do it, you left me with no reasonable alternative, you gave me the means to do it”. The superior officer who orders a subordinate to do something morally reprehensible is not in a position to blame the subordinate for carrying out the order, even if civilians are. And this is not necessarily because the superior officer has himself done similar things in the past (perhaps he hasn't), but instead because he is too closely involved in the very act he purports to condemn.
Of course, there are complications here, as well. For example, we can distinguish between permissible involvement in someone else's transgression and impermissible involvement in someone else's transgression, and we can ask whether both or only one of these ways of being involved undermines one's standing to blame. Patrick Todd (2012) argues that only impermissible involvement undermines one's standing to blame, and to illustrate his point he imagines two Nazi commanders, one of whom is committed to the Nazi cause and the other of whom is using his position of power as an attempt to undermine the regime. If, in order to keep up appearances, the latter commander issues an order for a subordinate to perform an action that is morally impermissible, does he lose his standing to blame the subordinate? Todd claims that he doesn't, and thus that there is an important distinction between complicity that undermines standing and complicity that doesn't.
Even if a blamer isn't a hypocrite and isn't involved in the action she is condemning, her blame can nevertheless be inappropriate if the wrong in question is just none of her business. Linda Radzik (2011: 582) gives a nice description of our common moral attitudes toward these situations:
For example, a group of co-workers will often insist that the romantic infidelity of one of their members is none of their business (and then feel a bit ashamed when they continue to gossip about it). Neighbors and teachers hesitate to interfere with a parent's treatment of her child although they judge the treatment to be wrongful, unless the wrong reaches a certain level of severity. Even within close relationships, we are sometimes uncertain whether we should express our negative moral judgment of a friend's behavior. True, the hesitancy to sanction in these cases is sometimes based on laziness, self-interest, cowardice or uncertainty about the moral judgments at issue, none of which contradict the claim that we have the standing to sanction. But, at other times, our hesitancy seems to be based on the sense that it would be wrong to sanction. We say, “It isn't my place to interfere even though I can see what she is doing is wrong”. We do not feel entitled to sanction every wrongdoer for every wrong.
And the thought, of course, is that we do not feel entitled to sanction every wrongdoer for every wrong because we aren't so entitled. Radzik describes three situations in which only a limited group of individuals have the standing to blame: (1) cases where an agent wrongs herself, (2) cases where the wrong is “committed within special relationships, such as romantic partnerships, familial bonds, and friendships” (2011: 593), and (3) cases where third-party blame “would interfere with the victim's ability to find vindication in the aftermath of wrongdoing” (2011: 597).
It's a good question exactly why the standing to blame would be restricted only to certain individuals in these sorts of case. Perhaps there are norms of privacy at play (Smith 2007; Nagel 1998), or perhaps there's an illuminating analogy to be made here with the notion of standing in the law (Sabini and Silver 1982), or perhaps if we see blame as a response which presupposes that the person blamed is in some way accountable to the members of her moral community, then we can distinguish between several (overlapping) moral communities, only some of which any one person belongs to, and thus only some of which underwrite one's standing to blame (Duff 2010: 126). If we adopt Scanlon's recent account of blame (2008), then perhaps we can say that some wrongs are none of our business because they don't impair any of our relationships, and hence don't render appropriate any blame-constituting modifications in those relationships.
2.2.4 Moral Fragility
Moral luck (in all its forms) provides another perspective from which to see how the standing to blame might be undermined. Consider Gary Watson's (1987) influential discussion of Robert Harris, who is at once an unequivocally cruel murderer and also, in a real sense, a victim of his tragic formative circumstances. It's a legitimate question, given his history, whether Harris is even the sort of creature who is a sensible target of blame—that is, whether Harris is even a morally responsible agent in the first place—but even if we grant that he is, there's another potential obstacle to blame at work here. Watson expresses it like this (2004: 245):
The fact that Harris's cruelty is an intelligible response to his circumstances gives a foothold not only for sympathy, but for the thought that if I had been subjected to such circumstances, I might well have become as vile. What is unsettling is the thought that one's moral self is such a fragile thing. One tends to think of one's moral sensibilities as going deeper than that (though it is not clear what this means). This thought induces not only an ontological shudder, but a sense of equality with the other: I too am a potential sinner.
The obstacle to blame that Watson is describing here is not the thought that Harris might not be blameworthy (though he might not be), but rather the thought expressed well by the common thought that “There but for the grace of God go I”. It's a humbling perspective to take on one's agency, and one that may “taint one's own view of one's moral self as an achievement” (2004: 246), and make one feel that “indignation on one's part would be self-righteous and indulgent” (2004: 254). For want of a better term, we might say that this is a worry about subjunctive hypocrisy, since it certainly has a similar flavor to the hypocrisy worry discussed above. The thought is something like this: “If I were as bad as him, I'd have no standing to blame him. But the difference between us is simply a matter of luck, and surely my good moral luck can't serve as the basis for my moral standing to blame. So I lack the standing to blame even though I've never done the terrible things in question.”
2.3 Blaming Procedures
When X has the moral standing to blame Y, we might say that Y is within X's blaming jurisdiction (Coates & Tognazzini 2012). But even if Y is within X's jurisdiction, it may still be inappropriate for X to blame Y due to a number of other factors specific to the way in which X goes about blaming Y. If we think of blame as a “move” made through moral space, or as a contribution to a moral conversation (see McKenna 2012), then questions of standing address whether X is even playing the right game, or whether X even has a voice with which to speak. But there are still norms governing which moves X can make or what X can say. Call these procedural norms.
Analogous to the common thought that the punishment must fit the crime, it seems plausible to suppose that the blame must, in some sense, fit the transgression. Perhaps it's legitimate to be annoyed at your friend for forgetting your birthday one year, but you can't (at least in the absence of some special context) vow never to speak to him again as a result of that one lapse. What will count as an appropriately proportional blaming response to a transgression will no doubt vary with different relationships and different transgressions, but there will likely always be some responses that take the transgression too seriously, and some that don't take it seriously enough.
What counts as a proportional blaming response won't depend just on the nature of the transgression, though; it will likely also depend on the way the wrongdoer has responded to his own transgression. As Angela Smith (2007: 482) puts it:
If someone has an objectionable attitude toward me, for example, but is already reproaching herself for it and making efforts to change, then I may judge that I have no reason to adopt or express any blaming attitudes toward her at all. Her own self-reproach shows me that she already recognizes that I have moral standing and deserve better treatment, and therefore I may no longer see her attitude as posing a challenge to me or my status. In cases of this sort, the faulty attitude is still attributable to the agent and she is open to legitimate moral criticism for it; but the agent is already responding appropriately to this fact and therefore there may be no grounds for further criticism on the part of others.
A judgment about what counts as a proportional blaming response must also be tempered by facts about what temptations or circumstances we can reasonably expect human beings to resist. (The “expect” here is normative rather than predictive; see Wallace 1994). Erin Kelly makes this point well (2013: 258):
Our sense that a particular person is unfairly burdened by contingencies the rest of us rarely encounter triggers a judgment that compassion is appropriate and that the circumstances that call for our compassion temper a person's worthiness of blame.
Kelly puts the point in terms of blameworthiness, but in our terminology she is saying that even if a person has exercised his or her capacities of moral reasoning to do something wrong (and thus even if the person is blameworthy, in our sense), it may nevertheless be inappropriate to blame them if the transgression in question was a natural consequence of circumstances that would push anyone to the breaking point. Perhaps this last point is similar to the worry about subjunctive hypocrisy (see section 2.2.4). If it is, then it serves as a nice reminder that any proposed taxonomy is going to involve categories that blur and overlap, at least to some degree.
Imaginary philosophical examples are always told by an omniscient narrator, but of course real-life cases of blame are never like that, and we have to rely on our fallible judgments about the obscure motivations of other human beings. Sometimes we are confident that someone has done wrong; other times we let our anger hamper our imagination and our generosity in searching for possible excuses for apparent wrongdoing. Having too quick a temper is itself something for which one can be open to criticism, and what makes a temper count as too quick is often that it outstrips the evidence for wrongdoing. The realm of interpersonal blame is not perfectly analogous to the realm of legal responsibility, of course, so “beyond a reasonable doubt” may be too demanding a requirement, but nevertheless there is some epistemic standard that must be met before blame is appropriate, even if the potential target of blame is in fact blameworthy and inside of the potential blamer's jurisdiction.
In fact, Gideon Rosen (2004) has used epistemic considerations to argue for a general sort of skepticism about moral responsibility:
What must you think in order to judge that Bill, for example, is responsible for lying to his wife? You must think that at the time of action, either he knew that he had decisive reason not to lie, or if he did not know this, that his ignorance was the upshot of some prior bad action done in full knowledge of every pertinent fact or norm. You must think, in other words, that his bad action either is, or derives from, an episode of genuine, full-strength akrasia.
I suggest that given the opacity of the mind—of other minds and even of one's own mind—it is almost always unreasonable to place significant confidence in such a judgment.
Rosen's skepticism here relies on arguments presented earlier in the article for the conclusion that the epistemic requirements on moral responsibility are quite stringent (in order for ignorance to be culpable, it must eventually trace back to clear-eyed akratic action; see also Levy 2011), but for our purposes the important point is simply that “the opacity of mind”, as Rosen puts it, can make it hard to tell when someone is genuinely blameworthy, and thus can render unjustified the judgment of wrongdoing on which appropriate blame rests. Of course, one need not accept Rosen's skeptical conclusion to agree that blamers must not jump to conclusions about wrongdoing.
2.4 Varieties of Appropriateness
The ambiguity of the word “appropriate” is useful if one is simply trying to enumerate all the possible ways that blaming can go wrong, but a full understanding of the ethics of blame will require more precision. Moreover, digging into the normative details will help to ensure that theorists aren't talking past each other when they discuss the “appropriateness”, “legitimacy”, “justice”, “fittingness”, or “rationality” of blame.
Take the case of hypocritical blame, for example: there is widespread agreement that hypocritical blame is inappropriate (though see Bell 2013a for a dissenting view). R. Jay Wallace (2011) takes this to mean that hypocritical blame is morally objectionable, and thus his investigation leads him to identify the moral principle that the hypocritical blamer flouts. But it is clear that G. A. Cohen, on the other hand, is interested in a different sense of appropriateness altogether (2006: 119–120, n. 10):
My topic is not when it's morally permissible or obligatory to condemn, and it is not part of my view that it is always bad or wrong for someone who is not in a position to condemn to condemn…I think one can say: “He has no right to condemn, but let us hope he does condemn”, and maybe even “but he ought to do so …”.
Cohen frames his investigation more in terms of when someone has the “right” to blame, and he thinks that someone might lack that right even while that person is morally obligated to act as though he does.
T. M. Scanlon, too, seems to be interested in a non-moral sense of appropriateness. As described above, his way of explaining how hypocrisy undermines standing is to say that hypocritical blame is unfitting, not true to the facts as blame represents them. It may be that unfitting blame is also morally inappropriate, of course, but if so that would require an argument. (On the distance between fitting and appropriate, see D'Arms & Jacobson 2000.)
There are many other senses in which an instance of blame might be inappropriate: perhaps it fails to motivate its target, or perhaps it fails to communicate a second-personal reason, or perhaps it displays a vice. Gary Watson (2013b), for example, has offered an analysis of the vice of judgmentalism, which he understands as either of two related faults: (1) lacking interpretive generosity or (2) being too unaccepting of the faults of others. Of course, it's controversial whether blame is (merely) a judgment, but in any case it is clear that certain responses to wrongdoing (perhaps blame is among them) can be vicious even if they don't violate the principles of interpersonal morality. (Pettigrove 2012 contains a similar discussion, but about the alleged virtue of meekness.)
2.5 Varieties of Blame
Because the nature of blame is controversial, we have tended in this section simply to use the word “blame”, but it's important to remember that one's views about the ethics of blame will very much depend on what one takes blame to be. For instance, although hypocritical moral address (to use Wallace's term) seems clearly to undermine standing, it is less clear whether merely harboring a hypocritical blaming judgment is likewise inappropriate, or inappropriate in quite the same way. In general, the difference between expressed and unexpressed blame seems likely to be an important distinction to keep in mind when theorizing about the ways in which blame can go wrong.
Moreover, blame seems to be just one among many ways that we respond to wrongdoing, and it's still an open question just how blame relates to activities like holding responsible, demanding answers, punishing, and so on. (For some attempted taxonomies, see Macnamara 2011; Shoemaker 2011; Smith 2012; Tognazzini forthcoming.) So, answers to the above questions about the ethics of blame will not automatically double as answers to analogous questions about the ethics of these other ways of interacting.
- Arpaly, N., 2006, Merit, Meaning, and Human Bondage: An Essay on Free Will, Princeton: Princeton University Press.
- Arpaly, N. and T. Schroeder, 2014, In Praise of Desire, New York: Oxford University Press.
- Beardsley, E., 1969, “A Plea For Deserts”, American Philosophical Quarterly, 6: 33–42.
- Bell, M., 2013a, “The Standing to Blame: A Critique”, in Coates & Tognazzini 2013: 263–281.
- –––, 2013b, Hard Feelings: The Moral Psychology of Contempt, New York: Oxford University Press.
- Bennett, C., 2013, “The Expressive Function of Blame”, in Coates & Tognazzini 2013: 66–83.
- Brandt, R. B., 1959, “Moral Assessments of Actions and Persons”, Ethical Theory, Englewood Cliff, NJ: Prentice-Hall.
- Coates, D. J. and N. A. Tognazzini, 2012, “The Nature and Ethics of Blame”, Philosophy Compass, 7: 197–207.
- ––– (eds.), 2013, Blame: Its Nature and Norms, New York: Oxford University Press.
- Cohen, G. A., 2006, “Casting the First Stone: Who Can, and Who Can't, Condemn the Terrorists?” Royal Institute of Philosophy Supplement, 58: 113–136.
- –––, 2012, “Ways of Silencing Critics”, in his Finding Oneself in the Other, Princeton: Princeton University Press, pp. 134–142.
- D'Arms, J. & D. Jacobson, 2000, “The Moralistic Fallacy: On the ‘Appropriateness’ of Emotions”, Philosophy and Phenomenological Research, 61: 65–90.
- Darwall, S., 2006, The Second-Person Standpoint: Morality, Respect, and Accountability, Cambridge, MA: Harvard University Press.
- Duff, R. A., 1986, Trials and Punishments. Cambridge: Cambridge University Press.
- –––, 2010, “Blame, Moral Standing and the Legitimacy of the Criminal Trial”, Ratio, 23: 123–140.
- Dworkin, G., 2000, “Morally Speaking”, in Edna Ullmann-Margalit (ed.), Reasoning Practically, New York: Oxford University Press, pp. 182–188.
- Fischer, J. M., 1994, The Metaphysics of Free Will: An Essay on Control, Oxford: Blackwell.
- Fischer, J. M. and M. Ravizza, 1998, Responsibility and Control: A Theory of Moral Responsibility, Cambridge: Cambridge University Press.
- Franklin, C. E., 2013, “Valuing Blame”, in Coates & Tognazzini 2013: 207–223.
- Friedman, M., 2013, “How to Blame People Responsibly”, The Journal of Value Inquiry, 47: 271–284.
- Gibbard, A., 1990, Wise Feelings, Apt Choices: A Theory of Normative Judgment, Cambridge: Harvard University Press.
- Ginet, C., 2000, “The Epistemic Requirements for Moral Responsibility”, Nous, 34: 267–277.
- Haji, I., 1998, Moral Appraisability: Puzzles, Proposals, and Perplexities, New York: Oxford University Press.
- Hart, H. L. A., 1968, Punishment and Responsibility, Oxford: Oxford University Press.
- Hieronymi, P., 2001, “Articulating an Uncompromising Forgiveness”, Philosophy and Phenomenological Research, 62: 529–555.
- –––, 2004, “The Force and Fairness of Blame”, Philosophical Perspectives, 18(1): 115–148.
- –––, 2008, “Sher's Defense of Blame”, Philosophical Studies, 137(1): 19–30.
- Houston, B., 1992, “In Praise of Blame”, Hypatia, 7: 128–147.
- Hurley, E. A. and C. Macnamara, 2010, “Beyond Belief: Toward a Theory of the Reactive Attitudes”, Philosophical Papers, 39(3): 373–399.
- Kekes, J., 2009, “Blame versus Forgiveness”, The Monist, 92: 488–506.
- Kelly, E., 2013, “What is an Excuse?”, in Coates & Tognazzini 2013: 244–262.
- Kenner, L., 1967, “On Blaming”, Mind, 76(302): 238–249.
- Levy, N., 2011, Hard Luck, New York: Oxford University Press.
- Macnamara, C., 2011, “Holding Others Responsible”, Philosophical Studies, 152: 81–102.
- Mason, M., 2011, “Blame: Taking it Seriously”, Philosophy and Phenomenological Research, 83: 473–481.
- McGeer, V., 2013, “Civilizing Blame”, in Coates & Tognazzini 2013: 162–188.
- McKenna, M., 2012, Conversation and Responsibility, New York: Oxford University Press.
- –––, 2013, “Directed Blame and Conversation”, in Coates & Tognazzini 2013: 119–140.
- Mele, A., 2011, “Moral Responsibility for Actions: Epistemic and Freedom Conditions”, Philosophical Explorations, 13: 101–111.
- Nagel, T., 1998, “Concealment and Exposure”, Philosophy and Public Affairs, 27: 3–30.
- Nussbaum, M. C. 2001, Upheavals of Thought: The Intelligence of Emotions, Cambridge: Cambridge University Press.
- Pettigrove, G., 2012, “Meekness and ‘Moral’ Anger”, Ethics, 122: 341–370.
- Pickard, H., forthcoming, “Irrational Blame”, Analysis, first published online September 5, 2013, doi:10.1093/analys/ant075
- Radzik, L., 2011, “On Minding Your Own Business: Differentiating Accountability Relations within the Moral Community”, Social Theory and Practice, 37: 574–598.
- Rosen, G., 2004, “Skepticism about Moral Responsibility”, Philosophical Perspectives, 18: 295–313.
- Sabini, J. and M. Silver, 1982, Moralities of Everyday Life, Oxford: Oxford University Press.
- Scanlon, T. M., 1986, “The Significance of Choice”, The Tanner Lectures on Human Values VIII, Salt Lake City: University of Utah Press.
- –––, 2008, Moral Dimensions: Permissibility, Meaning, Blame, Cambridge, MA: Harvard University Press.
- –––, 2013, “Interpreting Blame”, in Coates & Tognazzini 2013: 84–99.
- Sher, G., 2006, In Praise of Blame, Oxford: Oxford University Press.
- –––, 2013, “Wrongdoing and Relationships: The Problem of the Stranger”, in Coates & Tognazzini 2013: 49–65.
- Shoemaker, D., 2011, “Attributability, Answerability, and Accountability: Toward a Wider Theory of Moral Responsibility”, Ethics, 121: 602–632.
- Smart, J. J. C., 1961, “Free Will, Praise and Blame”, Mind, 70(279): 291–306.
- Smith, A. M., 2005, “Responsibility for Attitudes”, Ethics, 115(2): 236–271.
- –––, 2007, “On Being Responsible and Holding Responsible”, The Journal of Ethics, 2: 465–484.
- –––, 2008, “Character, Blameworthiness, and Blame: Comments on George Sher's In Praise of Blame”, Philosophical Studies, 137(1): 31–39.
- –––, 2012, “Attributability, Answerability, and Accountability: In Defense of a Unified Account”, Ethics, 122: 575–589.
- –––, 2013, “Moral Blame and Moral Protest”, in Coates & Tognazzini 2013: 27–48.
- Solomon, R. C., 1993, The Passions: Emotions and the Meaning of Life, Indianapolis: Hackett Publishing Company.
- Strawson, P. F., 1962, “Freedom and Resentment”, Proceedings of the British Academy, 48: 1–25, reprinted in Watson 2003.
- Szabados, B. and E. Soifer, 2004, Hypocrisy: Ethical Investigations, Broadview Press.
- Talbert, M., 2012, “Moral Competence, Moral Blame, and Protest”, The Journal of Ethics, 16: 89–109.
- Todd, P., 2012, “Manipulation and Moral Standing: An Argument for Incompatibilism”, Philosophers' Imprint, 12 (March) [Todd 2012 available online].
- Tognazzini, N. A., forthcoming, “The Strains of Involvement”, in Clarke, McKenna, and Smith (eds.), The Nature of Moral Responsibility, New York: Oxford University Press.
- van Inwagen, P., 1983, An Essay on Free Will, Oxford: Clarendon Press.
- Wallace, R. J., 1994, Responsibility and the Moral Sentiments, Cambridge, MA: Harvard University Press.
- –––, 2010, “Hypocrisy, Moral Address, and the Equal Standing of Persons”, Philosophy & Public Affairs, 38: 307–341.
- –––, 2011, “Dispassionate Opprobrium: On Blame and the Reactive Sentiments”, in Wallace, Kumar, and Freeman (eds.), Reasons and Recognition: Essays on the Philosophy of T. M. Scanlon, New York: Oxford University Press, pp. 348–372.
- Watson, G., 1987, “Responsibility and the Limits of Evil: Variations on a Strawsonian Theme”, in F. Schoeman (ed.), Responsibility, Character, and the Emotions: Essays in Moral Psychology, Cambridge: Cambridge University Press, pp. 256–286.
- –––, 1996, “Two Faces of Responsibility”, Philosophical Topics, 24(2): 227–248.
- –––, (ed.), 2003, Free Will, 2nd ed, Oxford: Oxford University Press.
- –––, 2004, Agency and Answerability, Oxford: Clarendon Press.
- –––, 2011, “The Trouble with Psychopaths”, in Wallace, Kumar, and Freeman (eds.), Reasons and Recognition, New York: Oxford University Press, pp. 307–331.
- –––, 2013a, “Moral Agency”, in H, LaFollette (ed.), The International Encyclopedia of Ethics, Oxford: Blackwell, pp. 3322–3333.
- –––, 2013b, “Standing in Judgment”, in Coates & Tognazzini 2013: 282–302.
- Wertheimer, R., 1998, “Constraining Condemning”, Ethics, 108: 489–501.
- Wolf, S., 2011, “Blame, Italian Style”, in Wallace, Kumar, and Freeman (eds.), Reasons and Recognition: Essay on the Philosophy of T. M. Scanlon, New York: Oxford University Press, pp. 332–347.
- Zimmerman, M., 1988, An Essay on Moral Responsibility, Totowa, NJ: Rowman & Littlefield.
Our sincere thanks to John Martin Fischer, Coleen Macnamara, Angela Smith, and Gary Watson for all of their help thinking about moral responsibility and blame over the past several years, and to the American Council of Learned Societies and The College of William & Mary for financial assistance during the research for this entry.