Published in The Journal of Ethics 16 (2012): that counted against their behavior. This assumption imposes a moral competence requirement

Transcription

1 MORAL COMPETENCE, MORAL BLAME, AND PROTEST MATTHEW TALBERT WEST VIRGINIA UNIVERSITY Published in The Journal of Ethics 16 (2012): Many theorists assume that wrongdoers are open to serious moral blame only if they had the psychological, emotional, and motivational resources to respond to the moral considerations that counted against their behavior. This assumption imposes a moral competence requirement on blame, and agents who fulfill this requirement can respond to morally relevant features of their environments. 1 According to proponents of the moral competence requirement, only competent wrongdoers can reasonably be exposed to moral blame because only these agents are reasonably held to the expectation that they respond to moral considerations. This paper offers a counterpoint to theories of moral responsibility that emphasize the importance of moral competence for blameworthiness. I argue that if it is generally appropriate to blame wrongdoers, then it can sometimes be appropriate to blame wrongdoers with impaired moral competence even if they could not reasonably have been expected to respond to the considerations that counted against their actions. Possession of moral competence is not what separates agents who are open to blame from those who are not or so I shall argue. Since writers use blame and blameworthiness in different ways, I will say something about what I mean by these terms. Like many compatibilists, I assume that a person is blameworthy if she is a proper target for negative reactive attitudes like resentment. 2 Of course, 1 A requirement of this sort plays an important role in Benson (2001), Doris and Murphy (2007), Levy (2003) Wallace (1996), Watson (2004a), and Wolf (1990, 2003), among many other possible examples. Some writers (including some of those just mentioned) use the phrase normative competence in roughly the way I use moral competence. I chose the latter expression to emphasize my focus on specifically moral norms and considerations. 2 For the seminal account of blameworthiness in these terms, see Strawson (2003). 1

2 it is possible to judge that someone is a proper target for resentment without actually resenting that person, and even when we feel resentment, this fact may not find expression in our behavior. When I say that a person is blameworthy, I mean that attitudes like resentment are reasonably directed at that person, regardless of whether anyone experiences or expresses these attitudes. When I speak of blaming, I usually have in mind the actual expression of negative reactive attitudes, though it is no doubt true that in some cases the mere judgment that a person is a proper target for these reactions also constitutes a form of blame. 3 I also assume again following other compatibilists that a person is a proper target for negative reactive emotions when she treats others with unjustified ill will, contempt, or disregard. 4 In the opening sections of this paper, I argue that even if a wrongdoer is impaired in her ability to recognize and respond to certain moral considerations, her actions may still express the sort of ill will to which blame responds. Of course, the wrongdoers I have in mind are so unlikely to respond to the relevant moral considerations that it is not reasonable to demand that they do so. Since I argue that these agents are open to blame, I am committed to viewing the fairness of blame as independent of the reasonableness of demanding that an agent respond to moral reasons. I consider the claim that blame issues a moral demand in Section 5. In Section 6, I argue that the victims of impaired wrongdoers can reasonably protest their treatment in a way that counts as a form of moral blame even though it does not primarily express a moral demand. As just noted, I make use of views about moral responsibility that many compatibilists endorse. The overall argument of this paper is also inspired by common compatibilist stances, particularly the suggestion that the powers of action guidance that matter for responsibility do not depend on access to alternatives, and the suggestion that the addition of alternatives to a 3 For an important treatment of the issues discussed in this paragraph, see Angela Smith (2007). 4 Again, see Strawson (2003). 2

3 deterministic sequence does not, by itself, enhance responsibility. Compatibilist are, then, the primary intended audience for this paper: I suggest that strategies compatibilists rely on in other contexts have a largely unnoticed application to cases of morally impaired agents Frankfurt-Cases and the Moral Competence Requirement Proponents of the moral competence requirement often make their case with examples of agents who have difficulty responding to certain moral considerations. These difficulties typically result not from mental disorders, but from an agent s possession of flawed values, dispositions, and beliefs. Often, the presence of these objectionable states is explained by the fact that the agent in question is the product of a specific cultural context. 6 Neil Levy, for example, argues that there are good reasons to believe that cultural membership can be ethically disabling: that our moral education, which is systematically related to the culture to which we belong, can prevent us from grasping important moral facts (Levy 2003, 160). Levy speaks of ethical disability, but the kind of impairment he has in mind does not 5 I take moral responsibility to be compatible with determinism and with certain forms of moral impairment, but I do not think this commits me to a view about the compatibility of free will and determinism, or free will and moral impairment. I endorse, then, something like Fischer and Ravizza s semi-compatibilism. Semi-compatibilists hold that moral responsibility is compatible with causal determinism, even if causal determinism is incompatible with freedom to do otherwise (Fischer and Ravizza 1998, 53). A lot hangs here on what we mean by freedom to do otherwise. If what we have in mind is the freedom to, as Carl Ginet puts it, add to the given past in multiple ways, then free will may not be compatible with determinism (because if determinism is true, there would seem to be a single causally possible extension of the actual past) (Ginet 1990, 103). In reply, compatibilists can argue that Ginet s formulation does not capture the relevant notion of free will, or that, even on this conception, free will is compatible with determinism. Both responses are contentious. The semi-compatibilist finds it dialectically fruitful to leave these debates unresolved, arguing instead that the freedom to add to the given past in multiple ways is not necessary for moral responsibility regardless of whether it is compatible with determinism. Similarly, I would argue that, even if free will is the ability to add to the given past in multiple ways, and even if a morally impaired agent s free will is undermined, she may still be a proper target for blame because blameworthiness does not depend on this sort of free will. Still, the impaired agents I have in mind satisfy conceptions of freedom that are less metaphysically loaded than Ginet s. Among other things, the agents I consider should be understood to act as they see fit, to be moved by desires with which they identify, and to be such that they would have acted differently if they had taken themselves to have reason to do so. What the impairment of these agents comes to is that, when they do wrong, they could not reasonably have been expected to grasp the fact that they had good reason to do otherwise. 6 Another common, though often unstated, assumption is that the relevant states are, as Al Mele puts it, practically unsheddable. Values are practically unsheddable for an agent if, given her actual psychological constitution, and the way the world actually is, she cannot change the fact that she has the values in question (Mele 1995, 153). 3

4 entail that an agent is literally, and in every sense, unable to grasp a particular moral fact or consideration. According to Levy, the hypothesis that cultural membership can excuse someone from moral responsibility does not stand or fall with the inability thesis (Levy 2003, 156). What matters is not whether an agent is entirely unable to grasp a certain moral fact, but whether it is reasonable to expect her to grasp that fact, and to conform herself to moral norms. Paul Benson makes the same point this way: The main practical issue that dwells behind the inability thesis is not inability per se but rather whether we can reasonably hold persons to expectations that they acquire knowledge about the moral quality of certain actions or social practices. For even someone who has the raw competence to gain particular pieces of moral knowledge may be culturally positioned in such a way that attaining that knowledge would be so extremely difficult that we could not fairly hold her to the demand that she do so. 7 (Benson 2001, 614) One could insist, of course, that the agents Levy and Benson have in mind possess enough capacity for moral insight that it is reasonable to expect them to know right from wrong. Perhaps this is so, and if it is, then the agents in question may be blameworthy. However, I wish to concede that an agent s moral competence may be so impaired that it is unreasonable to expect her to see how she ought to behave, and yet I want to argue that such an agent may still be blameworthy. If this seems implausible, then I hope that reflecting on considerations advanced by Harry Frankfurt against the Principle of Alternate Possibilities (PAP) will begin to turn the intuitions of compatibilists, at least, in my direction. According to PAP, an agent is morally responsible for an action only if she could have done otherwise, but Frankfurt argues that the absence of alternatives does not always undermine 7 A compatibilist might make sense of the idea that it is reasonable to expect an agent to have correct moral understanding by appealing to what Fischer and Ravizza (1998, 53) call the dispositional or modal properties of the agent s actual behavior. The idea is that we gain insight into an agent s general psychological capacities by considering how the agent or, in Fischer and Ravizza s terms, the agent s reasons-responsive mechanism would have responded had she been exposed to counterfactual pressures and incentives. When an agent fails to respond appropriately to a given moral reason, we might say that she had the general capacity to respond to the reason and that it was reasonable to expect her to do so if she would have responded to that reason in the proper range of counterfactual scenarios. 4

5 moral responsibility. Suppose, Frankfurt suggests, that Black is ready and able to ensure that Jones performs some action, but that as things actually transpire, Black never has to show his hand because Jones 4, for reasons of his own, decides to perform and does perform the very action Black wants him to perform (Frankfurt 1969, 836). Frankfurt says that in such a case, Jones would be morally responsible for his action even if he could not have avoided it. Jones responsibility depends on the fact that he acted for reasons of his own, and we seem to be able to explain his behavior in this way regardless of whether he could have done otherwise. In a recent reflection on Frankfurt s work, John M. Fischer interprets the view this way: what we value in action for which an agent can legitimately be held morally responsible is not that he makes a certain sort of difference to the world, but rather that he expresses himself in a certain way. And this sort of self-expression does not require alternative possibilities. (Fischer 2006, 128) As Frankfurt s example suggests, when it comes to moral responsibility, a salient form of selfexpression occurs when an agent acts for her own reasons. And while there may be cases (e.g., instances of severe compulsion) in which an agent cannot do otherwise and cannot act for her own reasons, the fact that someone cannot do otherwise does not entail that she fails to govern her behavior in this way. 8 If governing one s behavior in the light of reasons, and the selfexpression consequent upon doing so, are central to moral responsibility, and if a person can govern her behavior in this way even if she cannot do otherwise, then moral responsibility may be compatible with a lack of access to behavioral alternatives. While Frankfurt s proposal has encountered a great deal of criticism, many compatibilists find an account along the general lines just sketched attractive. The point I want to make is that similar reasoning can be applied to cases of agents with impaired moral competence. Of course, these cases are different from Frankfurt s example in certain respects. Most importantly, the 8 Though I do not agree with it in all its particulars, John M. Fischer s influential concept of guidance control, introduced in Fischer (1994) is a helpful way of thinking about the sort of self-government at issue here. 5

6 impaired agent s access to moral understanding is obstructed by actual internal factors (such as her values, dispositions, or moral beliefs), whereas the factors that limit Jones ability to do otherwise are external and counterfactual. Still, there is a crucial respect in which Jones and the morally impaired agent may be similar: Jones can act for his own reasons even if he cannot avoid a particular action, and perhaps the same is true of a morally impaired wrongdoer. After all, even if we concede that a wrongdoer s corrupt values seriously impair her ability to recognize the moral reasons that count against her behavior, this does not mean that she acts for no reason. I suggest, then, that even if a wrongdoer lacked reasonable access to certain moral considerations, she may still have acted because of other considerations that she took to count in favor of so acting. In the next section, I will argue that we may reasonably care and have a moral investment in why someone behaves as she does, even if we believe that certain possibilities of moral awareness, concern, and motivation, were not practically available to her. Blameworthiness may be preserved in such a case because something relevant to the attribution and evaluation of actions namely, the agent s judgments about what reasons she has contribute to her behavior. 2. Quality of Will and Moral Competence I have just argued that an insight from Frankfurt-style compatibilism may be applicable to cases of morally impaired agents, at least if they are able to guide their behavior by judgments about reasons. But why should it matter whether an impaired agent can govern her behavior in this way? My answer to this question appeals to P. F. Strawson s proposal that attitudes like 6

7 resentment are responses to the quality of others wills towards us, as manifested in their behavior: to their good or ill will or indifference or lack of concern (Strawson 2003, 83). 9 Strawson s approach to moral responsibility emphasizes both our concern for the attitudes and intentions towards us of other human beings (Strawson 2003, 75), and the way our reactions to others behavior depend on our evaluations of the dispositions and intentions that move them. Of special significance, for my purposes, is our tendency to judge that our needs and interests should have normative significance for other people, and our corresponding concern with whether others treat our needs and interests as giving them reasons to constrain their behavior in various ways. When others fail to recognize these reasons, we often perceive their actions as expressing disrespect, ill will, or at least an insufficient regard for our standing as beings whose needs and interests should have normative weight. I shall assume that when the actions of a morally competent agent issue from an unjustified failure to take others welfare as reason-generating, these actions often convey a quality of will that licenses blaming attitudes like resentment. 10 The problem is to explain how the actions of agents with impaired moral competence can express a similar quality of will. In order to address this issue, we should think about the kinds of impairments that proponents of the moral competence requirement take to undermine blameworthiness. Levy, for example, considers the case of a person raised in a slaveholding society in which racist notions 9 I should note that Strawson himself takes moral competence to be a condition on blameworthiness insofar as he thinks that those we blame should be capable of engaging with the negative reactive attitudes we direct at them: see Strawson (2003, 78-82). Since the morally impaired agents I am interested in may lack the ability to engage constructively with our reactive attitudes, Strawson would presumably regard them as being like the psychotic or the young child beyond the proper reach of blame. I consider the degree to which blame presupposes the ability to engage with the negative reactive attitudes in Sections 5 and 6 below. 10 This approach is indebted to T. M. Scanlon s development of Strawson s account of blame in the terms of contractualist moral theory. According to Scanlon, a judgment of moral blame asserts that the way in which an agent decided what to do was not in accord with standards which that agent either accepts or should accept insofar as he or she is concerned to justify his or her actions to others (Scanlon 2003, 365). I take it that we often find an action unjustifiable (and expressive of ill will or objectionable disregard) when an actor explicitly or implicitly judges our needs and interests to be normatively insignificant. 7

8 are widely accepted and reinforced. Levy notes that such a person might be exposed to alternative views... these views might be expressed by the slaves themselves (Levy 2003, 152). But Levy encourages us to bear in mind the facts concerning moral education : Our agent has grown up in a slave-owning society, and has been taught by everyone she most loves and respects her parents and other close relatives, her teachers, the authorities in her society that slaves are (say) subhuman. Now if a slave tells her otherwise, has she any reason to believe him? Surely she ought, rationally, to give the views which have been inculcated in her, and which are held by those who, it is acknowledged on all sides, are the wisest members of her culture, greater weight than the view of someone she at least suspects of being subhuman. (Levy 2003, 152) Part of Levy s point is that if it is not subjectively rational for a wrongdoer to question the values and beliefs that contribute to her wrongdoing, then her capacity for certain moral realizations is, for all practical purposes, seriously impaired. 11 It is therefore not fair to expect the wrongdoer to respond appropriately to the relevant moral considerations, or to blame her for not doing so. Susan Wolf considers similar cases in Freedom Within Reason. Wolf says that while people raised in a racist society like the slaveholding U. S. south may be capable of appreciating some basic moral attitudes and judgments, they may also exhibit gaps in their moral understanding that are most naturally explained by factors in their upbringings and their social environments (Wolf 1990, 121). Elsewhere, Wolf says that wrongdoers such as the slaveowners of the 1850s and the Nazis of the 1930s may have behaved in ways that are strongly encouraged by their society with the result that these wrongdoers may falsely believe that the ways in which they are acting are morally acceptable (Wolf 2003, 382). If we believe that these wrongdoers could not help but be mistaken about their values, then Wolf suggests that we do not blame them for the actions those values inspired (Wolf 2003, 382). 11 Elsewhere, Levy says: It is a necessary condition of agents possessing this capacity [for conforming to moral norms] and therefore of us reasonably expecting them to behave appropriately that conforming their behavior to normative standards is something they can do rationally (and not merely by chance or accident) (Levy 2009, 735). 8

9 Wolf s well-known discussion of JoJo makes the same point. JoJo is the son of a ruthless dictator, and after internalizing his father s values he is unable to recognize his own brutal behavior as immoral (Wolf 2003, ). Like the Nazi and the slave-owner, JoJo is a victim of his environment, but what calls the blameworthiness of these actors into question is their impaired ability to grasp certain moral considerations. According to Wolf, the reason for withholding blame from [i.e., slave-owners and Nazis] is at bottom the same as the reason for withholding it from JoJo. Like JoJo, they are, at the deepest level, unable cognitively and normatively to recognize and appreciate the world for what it is (Wolf 2003, 383). John Doris and Dominic Murphy have recently developed the moral competence requirement in a unique way. They argue that soldiers (even those who commit war crimes) typically occupy excusing conditions and are therefore not morally responsible for their conduct (Doris and Murphy 2007, 26). The proposed basis for excuse is that soldiers are very often exposed to situational pressures that undermine their normative competence, which is a complex capacity enabling its possessor to appreciate ethical considerations, ascertain information relevant to particular ethical judgments, and identify behavior implementing their ethical judgments (Doris and Murphy 2007, 30). Doris and Murphy suggest that the situational pressures to which soldiers are exposed can alter their standards of appropriate behavior such that atrocious behavior can come to seem morally appropriate (Doris and Murphy 2007, 38). Since distorted moral perceptions make it very difficult for soldiers to assess and respond appropriately to moral considerations, Doris and Murphy argue that soldiers are very often excused for their behavior. 12 As I noted in the last section, we could respond to the preceding examples by insisting that someone like JoJo really can grasp the moral significance of his behavior. However, I shall 12 For an extended criticism of Doris and Murphy s view along the lines suggested in this paper, see Talbert (2009). 9

10 assume that things are as Levy, Wolf, and Doris and Murphy suppose. I grant, then, that the wrongdoers just described cannot reasonably be expected to see that they behave badly. Given this assumption, do these wrongdoers express blame-grounding qualities of will through their actions? We might think not. After all, these agents believe through no fault of their own, let us assume that their actions are permissible. And sometimes, even when an actor does harm, his will is not bad if he thinks he acts permissibly. Note, for example, how we excuse behavior by saying things like, She thought she was doing the right thing. Part of the point of this excuse is that even when a person s actions have unwelcome consequences, she can avoid blame if we think she was trying (perhaps clumsily) to help rather than to harm, and to make things better rather than to make things worse. But this excuse is not very plausible in the cases under discussion. Consider the antebellum slave-owner: even if he regards his actions as permissible, he can hardly be said to have had his slave s interests primarily in mind. Slave-owners, I take it, were not generally like people trying to enact their good intentions toward others, but going about it clumsily. 13 A more plausible hypothesis is that slave-owners were (as Levy suggests) aware that their slaves objected to the treatment they received, but that the slave-owners found it permissible to disregard these objections. Presumably, what accounted for many slave-owners finding this disregard permissible was their judgment that African slaves did not have the standing to raise normatively significant 13 Slave-owners may have held a paternalistic attitude toward slaves that they regarded as incapable of looking after their own interests, but this attitude is compatible with offensive racial contempt. Paternalism is a central topic of Eugene Genovese s classic study of American slavery, Roll, Jordan, Roll. Genovese claims that, Southern paternalism... had little to do with Ole Massa s ostensible benevolence, kindness, and good cheer. It grew out of the necessity to discipline and morally justify a system of exploitation. It did encourage kindness and affection, but it simultaneously encouraged cruelty and hatred (Genovese 1974, 4). Making a related point, J. L. A. Garcia notes that the racist who condescendingly and deliberately deprives others of education and autonomy, even if ultimately well intentioned, acts with the instrumental intention of stunting and infantilizing the beneficiary she victimizes. So we should not assume that immoral paternalism involves nothing that offends against the virtues of goodwill (Garcia 2001, 142 n. 22). 10

11 objections to how they were treated. I take this judgment about slaves standing to involve a morally significant contempt for the slaves regardless of whether the slave-owners were capable of forming morally preferable judgments. I conclude, then, that even if a slave-owner is unshakably confident that her behavior is permissible, her behavior can still express contempt for those she mistreats if it issues from the judgment that slaves have a status that permits their interests and suffering to be disregarded. We have, then, some reason to deny the claim that impaired moral competence undermines blameworthiness. Even if we concede that a morally impaired wrongdoer cannot be expected to respond appropriately to moral considerations, he may display blame-grounding contempt for his victims insofar as he guides his behavior by the (perhaps implicit) judgment that their suffering is no reason to refrain from an action. 14 Similarly, even if the values that blind a wrongdoer to moral realities were imposed upon him by external forces, this does not mean that his actions do not express objectionable contempt for those he mistreats. We should consider, furthermore, what it means to deny that someone like JoJo, or a slave-owner, is blameworthy for his actions. If a person is not blameworthy, then it is unfair to target that person with the negative attitudes involved in blame. Indeed, if JoJo and the slaveowner are not blameworthy, then it is not only unfair for us to blame them, it is also inappropriate for their victims to do so. This is a sobering thought. If it seems excessive to insist that a victim of slavery, or of torture at the hands of a dictator, should forswear blame, then we may be tempted by a compromise. We could say that blame on the part of the victim is understandable (so he should not be censured for it), but that, strictly speaking, blame is unreasonable in this case, given the 14 As Scanlon notes: [a] person who is unable to see why the fact that his action would injure me should count against it still holds that this doesn t count against it (Scanlon 1998, 288). 11

12 constraints that apply to blame. 15 However, victims of torture and slavery might plausibly respond that to count their blaming attitudes as merely understandable (and not fully reasonable) fails to take the full measure of the wrongs done to them. Sometimes it really is unreasonable to blame people for their behavior: if a person was forced by coercion or mind control to injure you, then blame would be out of place because unwilling or involuntary actions do not normally express the qualities of will that justify blame. The cases at hand, however, are not like this: JoJo does not involuntarily or unwillingly injure his victims, nor does he do so for reasons that we should expect his victims to accept as justifying his actions. But to say that it is, strictly speaking, unreasonable for JoJo s victims to blame him suggests that his victims would be more reasonable if they responded to their injuries as if they were morally analogous to injuries produced by involuntary or justifiable behavior. And this invites JoJo s victims to agree to a serious mischaracterization of their treatment. 16 Of course, we may insist that JoJo s actions constitute serious and offensive moral wrongs, and that this is not true of involuntary or justifiable behavior. But this insistence rings hollow if we do not allow that it is reasonable to respond to JoJo s actions as if they are serious and offensive moral wrongs. If we take the blaming responses off the table, I fear we risk losing our grip on the sort of interpersonally significant wrongdoing of which JoJo is supposed to be 15 Doris and Murphy consider something like this in their discussion of strict liability for the reactive attitudes (Doris and Murphy 2007, 54). 16 I would say something similar about attempts to apply Gary Watson s distinction between the attributability and accountability faces of responsibility in this context (Watson 2004b). With this distinction in hand, one might say that JoJo is blameworthy in the limited sense that his wrong actions are attributable to him, but that it is inappropriate to hold JoJo accountable for his actions by targeting him with negative reactive attitudes. But this fails to acknowledge the moral quality of the actions we are attributing to JoJo. Suppose that after calm deliberation, JoJo judges there to be no good reason to refrain from arbitrary arrest and torture. JoJo s victims might be obliged to withhold resentment if his actions were not malicious or if they were not attributable to him, but if these actions are attributable to JoJo, and if they are as deliberate and disdainful as I suggest, then his victims seem to have grounds for resenting him. For additional discussion and criticism of Watson s distinction, see Angela Smith (2008, ). 12

13 guilty. 17 From the perspective of JoJo s victims, what is morally significant about his behavior is the way it expresses the kind of ill will that we often mark with blame. Neither the fact that JoJo regards his behavior as permissible, nor the fact that he behaves as well as can be expected, changes the character of JoJo s behavior from the perspective of his victims because these facts do nothing to make JoJo s actions seem less deliberate or less motivated by disregard for his victims suffering. Thus, neither of these factors do much, I think, to undermine the legitimacy of blame in JoJo s case. 3. Does Moral Competence Explain Ill Will? So far, I have argued that there may be reason to attribute blame-grounding qualities of will to some impaired agents. We would have additional support for this conclusion if it turned out that, in some cases, even if a wrongdoer is morally competent, her competence is not what explains why her actions express ill will. To flesh out this suggestion, we can imagine two similar agents. First, consider Jessica, who is similar to the agents described by Levy and Wolf: she is a committed racist, but given the facts about her culture and upbringing, it is not reasonable to expect Jessica to recognize that racism is objectionable. This is not to say that she could not recognize this fact under any counterfactual circumstances. It is just that, through no fault of her own, Jessica was raised in a homogenous social context that strongly encouraged acceptance of dominant social values and beliefs, and now that she has internalized these values and beliefs, she has very limited practical access to successfully questioning them. Jessica s upbringing leaves her impaired in the way that Levy, Wolf, and many others take to undermine blameworthiness. But advocates of the moral competence requirement need 17 Scanlon makes a related point: if we give up the idea that an agent can be properly condemned for his action, then it seems that we must also withdraw the claim, on his victim s behalf, that they were entitled not to be treated in the way that he treated them (Scanlon 1998, 332). Thanks to Angela Smith for pointing me to this passage. 13

14 not maintain that all wrongdoers are blameless. A wrongdoer may be blameworthy if, for example, her culture and upbringing are significantly more conducive to correct moral values and beliefs than Jessica s is. Let us consider such an agent. Suppose that James, like Jessica, is a committed racist. However, while James family and friends are also mostly racists, James larger social context is not one in which racism is promoted. In fact, James is aware that most prominent figures in his society condemn racism. Since the factors that inhibit Jessica s access to correct moral understanding are much less pronounced in James case, he may be open to certain moral expectations that would be less reasonably imposed on Jessica. I assume that at least some proponents of the moral competence requirement allow for a dichotomy along the lines just described: some wrongdoers are blameworthy, and others are not, and one thing that explains the difference is that blameworthy wrongdoers satisfy more of the general conditions that allow most agents to form correct moral judgments. Note, however, that even if James has more plausible access to correct judgments than Jessica does, this does not entail that his actual racist judgments are more objectionable than Jessica s. Of course, unlike Jessica, James can be assumed to judge (implicitly at least) that the dominant views about race in his society are wrong, but this does not mean that James treats members of the targeted race more contemptuously, or with greater malice, than Jessica does. We may think that because James makes his racist judgments in the face of social criticism, he must be more deeply committed to his racism, but this also does not follow: Jessica may be just as intransigently racist as James, even if everyone in Jessica s society shares her values. And even if James is a more unshakably committed racist than Jessica is, this does not necessarily mean that his racist views are more vicious than Jessica s, or that he is more contemptuous of members of the targeted race than she is. It is possible that both Jessica and James act on 14

15 offensive judgments like those people don t have the same standing we do or those people don t have any right to complain about being treated this way. And if this is so, I suggest that it would be reasonable for those targeted by James s and Jessica s judgments and actions to be similarly offended by them, and to see them as licensing similar blaming attitudes. It is not incumbent upon the victims in these cases to view James actions as revealing a more blameworthy sort of contempt for their interests just because his judgments and actions contrast with the values generally endorsed in his society. 18 Of course, James may in fact be more contemptuous of those he mistreats than Jessica is. My point is that the difference in their relative degrees of contempt is not likely to be explained just by the fact that James has access to better moral judgments and modes of behavior. The fact that James could have reached better moral conclusions, need not, by itself, endow his actions with any special malice, or change the significance of these actions from the perspective of his victims. Instead, and as with Jessica, what makes James s actions significant for those he mistreats may be just the actual values, judgments, and choices that lead to his actions. This comparison between James and Jessica bears a relation to Frankfurt s Black-Jones 18 On my view, blame is appropriate when our actions manifest an unjustifiably contemptuous orientation towards others; I have particularly focused on the way our actions can express blame-grounding judgments about the normative significance of others welfare. Thus, there is likely to be more to Jessica s and James racism than their possession of false beliefs about a certain race. Such beliefs may play a relatively superficial role in explaining what is objectionable about racism. For one thing, racist beliefs need not be firmly held. Kwame Anthony Appiah says that it would be odd to call someone racist if she easily gave her views upon the presentation of appropriate evidence, but even if we disagree with Appiah about whether such a person counts as racist, he is certainly right to say that [r]eal live racists often exhibit a distorted rationality, and that their beliefs display a disturbing resistance to contrary evidence (Appiah 1990, 8). I suspect that often what makes the racist, and her beliefs, so offensive are the underlying attitudes of contempt that hold these beliefs in place. Thus, I am attracted to J. L. A. Garcia s suggestion that [a]ctions and beliefs are racist in virtue of their coming from racism in the desires, wishes and intentions of individuals (Garcia 1996, 11). More generally, Garcia characterizes racism as a vicious kind of racially based disregard for the welfare of certain people (Garcia 1996, 6). I view Jessica and James as being moved by the kind of disregard Garcia has in mind: they have racist beliefs, but their racism also (and perhaps more fundamentally) involves an orientation characterized by contempt, ill will, and antipathy towards certain others on account of their race. Still, I do not necessarily endorse Garcia s broader view that beliefs are inessential to racism. I want merely to emphasize the role that ill will can play in some instances of racism, and this emphasis is compatible with accepting, for example, Tommie Shelby s claim that racism is best construed as an ideology, as a system of beliefs that functions to establish or reinforce structures of social oppression (Shelby 2002, 415). 15

16 example: both cases focus our attention on the actual sources of action, rather than on counterfactuals associated with what actually happened. Frankfurt s example is a case in which deleting access to alternatives from the story of an agent s action is supposed to be irrelevant to the agent s moral responsibility. Similarly, with James case, I am suggesting that adding access to different psychological alternatives can be irrelevant to moral responsibility because the presence of psychological alternatives need not contribute to his actual quality of will, and may have nothing to do with why his actions are wrong and morally offensive. And if access to psychological alternatives does not explain why James actions express ill will, then Jessica s relative lack of access to alternatives need not entail that her actions fail to express ill will. But what of the fact that Jessica is a victim of misfortune? If Jessica had found herself in a better social context, she would have had increased access to certain forms of moral awareness and motivation, and thus to better modes of behavior. Such an expansion of opportunities for Jessica what some might see as an expansion of her freedom would be better for her, and for those around her, if it led her to fewer instances of wrongdoing. Part of what makes Jessica unfortunate is that, by hypothesis, she is not culpable for her cultural surroundings. Of course, Jessica may have made choices in her life that contributed to her possession of racist values, but, given her social context, it was overwhelmingly likely that she would make those choices, and, at any rate, she did not know that acquiring racist values is typically wrong. This last point about ignorance is important. As Holly Smith notes, [i]gnorance of the nature of one s act is the pre-eminent example of an excuse that forestalls blame (Smith 1983, 543). For example, if a doctor gives a patient the wrong treatment, she may be blameless if she was pursuing what she thought was the medically best option. Of course, our judgment of blameworthiness will likely depend on whether we think the doctor was culpable for her 16

17 mistaken view about what treatment was best. Suppose that the doctor earlier engaged in what Smith calls a benighting act (or omission) that failed to improve, or actually worsened, her cognitive position, and that this led her to pursue the wrong treatment (Smith 1983, 547). If we think the doctor was culpable for the benighting act, then we will likely regard her ignorance as culpable, and perhaps we will blame her for the consequences of her ignorance. 19 Smith s focus is on circumstantial rather than normative ignorance (Smith 1983, 544). Thus, in the above example, the doctor is unaware that, in the circumstances, treatment T is the wrong treatment, but she is aware that giving a patient an incorrect treatment is usually wrong. However, Gideon Rosen and Michael Zimmerman have recently both applied an approach like Smith s to cases of normative ignorance. Rosen and Zimmerman argue that for (normative or circumstantial) ignorance to be culpable, it must arise from an instance of knowing wrongdoing. 20 Additionally, both Rosen and Zimmerman suggest that many garden-variety unwitting wrongdoers may not be blameworthy because their moral ignorance is not appropriately causally related to a prior instance of knowing wrongdoing I say perhaps because Smith s aim is to distinguish the various positions one might take on this issue rather than to argue that a particular approach is correct. 20 Here is a relevant passage from Zimmerman: If one is culpable for nonignorant behavior, then of course, one s culpability involves lack of ignorance. If, in contrast, one is culpable for ignorant behavior, then one is culpable for the ignorance to which this behavior may be traced.... But one is never in direct control of whether one is ignorant.... Indirect culpability for something presupposes direct culpability for something else. Whatever this something else is, it cannot be ignorant behavior, because then the argument would apply all over again to this behavior. Hence all culpability can be traced to culpability that involves lack of ignorance, that is, that involves a belief on the agent s part that he or she is doing something morally wrong. (Zimmerman 1997, 418) Here is a similar passage from Rosen: Recall our observation that responsibility for action done from ignorance is invariably a matter of derivative responsibility: One is responsible for the act done from ignorance only if one is independently responsible for something else. On the present broad conception of action done from ignorance, this entails that the only possible locus of original responsibility is an akratic act. In weakness begins responsibility. Our first sin must be a knowing sin a sin done in full knowledge of every pertinent fact or principle. (Rosen 2004, 307) 21 Rosen s aim, more than Zimmerman s, is to urge epistemic humility: in most cases, we will not know whether a wrongdoer satisfies the conditions on blame (Rosen 2004, 295). Still, the tenor of his discussion suggests that Rosen would agree that there are fewer instances of genuinely blameworthy wrongdoing than ordinarily supposed. 17

18 On Rosen s and Zimmerman s accounts, it may well be that neither Jessica nor James are blameworthy because neither of them knowingly impaired themselves in a way that led to their false moral beliefs and bad actions. By contrast, William FitzPatrick argues that ignorance may be culpable even if it does not arise from knowing wrongdoing. On FitzPatrick s view, it is enough for culpability that an unwitting wrongdoer could reasonably have been expected to avoid, or to remedy, her ignorance (FitzPatrick 2008, 603). For FitzPatrick, while someone in Jessica s position is probably blameless, James may well be blameworthy because his social context makes it reasonable to expect him to recognize the error of his ways. 22 Contrary to the authors just cited, I reject the premise that a morally ignorant wrongdoer is blameworthy only if her ignorance is culpable. Even if an unwitting wrongdoer s ignorance is not her fault, moral blame may be appropriate if her actions express contempt for those she injures. As I suggested in the last section, it is quite possible both that a wrongdoer regards her behavior as permissible, and that, in an important sense, she is not well-intentioned. If someone like Jessica treats others in ways to which she knows they object, then the fact that she regards this as permissible does not alter the fact that her actions express the judgment that these others do not merit much consideration, that their interests do not matter, that their objections can be overlooked. And if one is injured by a wrongdoer who is moved by judgments of this sort, then the attitudes and responses involved in moral blame are appropriate regardless of whether the wrongdoer is at fault for her moral ignorance. 22 See FitzPatrick s discussion of Mr. Potter for an example of a wrongdoer who is situated similarly to James, and whom FitzPatrick regards as probably blameworthy (FitzPatrick 2008, 603). 18

19 4. When is Blame Unfair? When someone does wrong, does it matter whether she could have avoided wrongdoing? One reason for thinking that this does matter is that if a person cannot avoid wrongdoing, she will also have trouble avoiding blame, and it may seem unfair to blame a person who cannot avoid being blamed. This suggests a way of supporting the moral competence requirement: perhaps we should not blame a person who cannot respond appropriately to the considerations against doing wrong because it is unfair to blame those who have significant difficulty avoiding blame. 23 There are, however, different ways in which blame can be difficult to avoid, and the way in which this is difficult for morally impaired agents is not necessarily problematic. In a case of compulsion or mind control, someone might be unable to refrain from an action because her behavior arises independently of her decisions about how to behave. Blame would be inappropriate here because such actions are unlikely to express the attitudes and judgments to which blame responds. However, even if an impaired wrongdoer could not have refrained from a blameworthy action by exercising proper moral judgment, this does not mean that her behavior is independent of her deliberations, or that she would have been compelled to commit her act regardless of how she decided to behave. So, while the committed racist s distorted values may mean that she cannot recognize the force of certain moral considerations, this does not mean that she would not refrain from the action if she took herself to have good reason to do so. The morally impaired racist may have great difficulty avoiding blame for treating certain people contemptuously, but if this is only because her impairment involves a commitment to treating these people contemptuously, then it is not at all clear that blame is unfair in such a case. 23 Wallace develops this sort of argument in Wallace (1996, ). 19

20 Pamela Hieronymi has recently stressed the way in which blaming attitudes like resentment follow on the judgment that another person has shown us ill will. According to Hieronymi, these attitudes are simply reactions to that is, they simply mark or acknowledge the importance of a display of ill will or disregard (Hieronymi 2004, 135). If what arouses these blaming responses is typically the judgment that another has shown us ill will, then, as Hieronymi concludes, adopting the reactive attitudes could be rendered unfair only by considerations that bear on the content of the judgments they reveal (Hieronymi 2004, 133). But, again, the fact that a committed racist was not able to avoid, or to rid himself of, the values that lead him to treat others contemptuously does not necessarily undermine the judgment that he expresses ill will and contempt for others through his actions. I conclude that, at least on the sort of quality-of-will account of blame defended here, the fairness of blame is often independent of whether blame could have been avoided. The fairness of blame has to do, rather, with whether the person we blame revealed through his behavior qualities of will that support blaming responses. Thus, it is not necessarily unfair to blame an impaired agent just because her impairment disposes her to see reasons in favor of committing blameworthy acts. Of course, a proponent of the moral competence requirement could argue that it is unfair to blame impaired agents because their lack of competence means that their actions cannot express blame-grounding qualities of will, but I have explained in the preceding sections why I think such qualities of will can be attributed to impaired agents. 5. Is Blame a Form of Moral Communication? So far, I have tried to undermine the moral competence requirement by arguing that ill will does not depend on moral competence. But the moral competence requirement can also be motivated by interpreting blame as an attempt at moral communication that makes sense only in 20

21 the cases of agents who can be reached by such communication. Gary Watson, for instance, develops the Strawsonian picture of blame by focusing on the way in which negative reactive attitudes express a moral demand, a demand for reasonable regard (Watson 2004a, 229). According to Watson, To be intelligible, demanding presumes understanding on the part of the object of the demand. The reactive attitudes are incipiently forms of communication, which makes sense only on the assumption that the other can comprehend the message. 24 (Watson 2004a, 230) If the point of the reactive attitudes is to express moral demands, then Watson is correct when he says that [i]n a certain sense, blaming and praising those with diminished moral understanding loses its point (Watson 2004a, 230). Suppose that Watson is right, and that there is little point in issuing a moral demand to an agent who cannot comprehend that demand. This objection may apply more generally than Watson supposes, and it may even affect his own view. After all, the reactive attitudes typically arise after an unjustified injury has been done, and at that point it is a bit late to demand that a person show us due regard. 25 While a demand for respect may be, as Watson suggests, unintelligible when imposed on an agent who is incapable of conforming to that demand, there is also something strange about presenting this demand to an agent who has already ignored it: both agents are (in different ways) beyond the reach of this demand. Of course, this is not a serious problem for Watson s view. When he says the reactive attitudes express a demand for respect, Watson does not mean that the demand is issued for the first time when these reactive attitudes arise. Instead, Watson takes the demand for reasonable 24 R. Jay Wallace develops a similar position, claiming that insofar as the reactive emotions are bound up with moral obligations we accept, they will only be fully intelligible as forms of expression when addressed to people who are capable of grasping moral reasons (Wallace 1996, 164). 25 An obvious reply to this line of argument is that blame aims at securing future compliance with moral demands or at inspiring a wrongdoer to apologize or otherwise make amends. I consider this perspective in the next section. 21