‘Dirty Hands’: Guilt and Regret in Moral Reasoning

Introduction

In this article, I attempt to answer two related objections to arguments about ‘dirty hands’ reasoning leveled by Nigel Biggar. The first is that ‘dirty hands’ reasoning is paradoxical. I argue that situations of dirty hands are better understood as open cases than as paradoxes. The second is that ‘dirt’ in ‘dirty hands’ should be prima facie understood as regret rather than guilt. ¹ I argue that moral emotions like guilt and regret should be understood teleologically, not simply analytically, and that guilt, more than regret, intrinsically connotes this sort of teleology: it moves the agent in a situation of dirty hands to pursue the means of being made clean.

Biggar’s work exemplifies the sophisticated nature of arguments about guilt versus regret, not least because he is concerned with the way the distinction relates to the internal coherence of the overall account of the moral life within which it operates. ² Biggar argues that moral reasoning along the lines of dirty hands requires a flattening out of the distinction between regret and guilt because this sort of reasoning trades on the paradox that, ‘[i]t can be right to do wrong, to get one's hands morally dirty, that it can be wrong to do right’. ³

Biggar resists the paradoxical way of reasoning and seeks to show how in a situation of moral compromise, the agent may still be non-culpable and, thus, should not be thought guilty, and should not feel guilty. ⁴ Rather, the agent's emotion should be categorized as regret. This means that Biggar appears to hold the position, though he does not put it in this exact way, that negative moral emotions can—and maybe should—be interpreted prima facie as regret, instead of prima facie as guilt.

I share a number of important concerns with Biggar. First, I am concerned about the coherence of the moral universe and the threat that appeals to ‘paradox’ in moral reasoning pose. Second, I am worried about those who are not guilty carrying an undue burden of guilt. Third, I am troubled about the way that guilt proliferates within our contemporary world such that we cannot see a way out of the morass of totalized complicity (in wrongs real and imagined) to a horizon of forgiveness, grace, and even beauty. ⁵ That granted, I want to defend the use of the term guilt to describe the negative moral emotions occasioned by action within situations of moral ambiguity.

My argument is that we best understand the distinction between guilt and regret within the problem of ‘dirty hands’ by locating the purpose or role of these negative moral emotions within a broader account of the moral life. If we ask what purpose they serve—to what end they are ordered—the most obvious answer is that moral emotions like guilt and regret suggest that we are in need of some sort of reconciliation—to God, to the law, to neighbor, or even to ourselves. ⁶ These emotions are not meant to be simply felt or experienced by the agent. They are not brute facts of the moral life—there to be stoically accepted. Rather, they are meant to move the agent to seek reconciliation that comes either in a word of forgiveness—that a wrong has been righted, or a word of vindication—that tells the agent that the deed was justified. To make the point using the image of ‘dirty hands’, these emotions are meant to compel the agent to seek out how they can be made clean. Understood this way, regret and guilt must be understood teleologically, not simply analytically.

Guilt, more than regret, intrinsically connotes this sort of teleology. It moves the agent to pursue the means of being made clean. Regret is a terminus of action; guilt is a spur to new action of repentance or repair or even vindication. Guilt motivates the agent to seek to be made right. ⁷ Moreover, guilt, more than regret, suggests a proper humility about our moral state. In situations of ambiguity, the agent genuinely does not know the morally justified course of action. The agent, thus, has good reason to be concerned with his moral state. Because in many of these situations, the moral ambiguity is compounded by the moral gravity of the choice—these can be life or death situations, or (e.g., torture cases) ones involving violations of human dignity—the agent has good reason to wonder if he is guilty of an especially grave moral evil. All things considered, then, I think that the category of guilt—in the prima facie way I indicated above—is worth defending.

This article is divided into two sections. In the first, I address Biggar's argument about the distinction between guilt and regret. During the discussion, I also briefly address Biggar's argument that arguments about dirty hands depend on paradox. I argue that situations of dirty hands are better understood as open cases than as paradoxes. In the second section I argue that since situations of ‘dirty hands’ are open cases, they require external adjudication after the action—and that the moral emotion of guilt or regret is meant to compel the agent to seek out that judgment. My argument throughout shows how Biggar's position is closer to the dirty hands position than he realizes.

Biggar on Regret

Nigel Biggar's writing exemplifies both the willingness to frankly address grave moral problems and the attempt to provide conceptual clarity to difficult problems in moral philosophy. ⁸ From an early article on the importance of casuistry, through an extended discussion of principles of moral reason like double effect in In Defence of War, to his most recent work that argues for the importance of the virtue of prudence in deliberations about rights, Biggar has dedicated the bulk of his career to helping the discipline of Christian ethics develop the skillset it needs to do its work. ⁹

Biggar's work on moral reasoning in In Defence of War has generated the most conversation and debate, and he has continued to develop central concepts from that text in his most recent work. ¹⁰ However, Biggar's most developed argument related to the problem of dirty hands, and to the importance of regret and guilt, is found in his 2018 article, ‘Compromise: What Makes it Bad?’. ¹¹ In this article, Biggar articulates eight criteria for distinguishing between good and bad compromises, defining compromise as ‘a decision that deliberately involves the loss of good—that is an evil. It involves either conceding an opportunity to maintain or promote a good or becoming a part-cause of evil’. ¹² The first criterion for distinguishing between good and bad compromises is that ‘the one who makes’ the good compromise should feel ‘agent-regret’—and that he should not feel guilt. ¹³

Each criterion is important in Biggar's analysis of compromise, but I think it fair to say that this first one bears more weight than the others, or at least that it plays an outsized role in his argument. Indeed, Biggar develops his account of ‘regret’ as a full argument against those who argue that a moral agent who has acted rightly might incur moral guilt. That is, Biggar's account of ‘agent-regret’ does not function simply as a criterion for distinguishing between good and bad compromises; it is a normative argument on its own.

Biggar writes, ‘Regret … is not the same as moral guilt. Sometimes, there is what Roman Catholic ethicists call ‘non-moral evil’. When it is morally right, the causing of evil is non-moral, cannot be culpable, and should not attract feelings of guilt’. ¹⁴ What Biggar means to avoid by deploying this distinction between guilt and agent-regret is the ‘paradox’ of claiming that ‘It can be right to do wrong’ by, for example, thinking that there are cases where we find it necessary to break an ‘absolute rule’ in order to do right. ¹⁵ While ‘objectively’ the compromiser ‘does no wrong’,

Subjectively … he might well agonise over his decision, question its necessity and his own motives—as a commendably sensitive conscience should. He might also deeply lament that he had to do what he did do—and so should we all. And he might have great difficulty in distinguishing between regret, lament and self-doubt on the one hand, and moral guilt on the other. ¹⁶

This is all because, as Biggar says, the action ‘did involve causing an evil; it involved deliberately causing it; it involved intentionally causing it. That is deeply regrettable and lamentable—yet it is not culpable and so does not merit guilt’. ¹⁷ While Biggar admits that the agent ‘might have great difficulty in distinguishing between regret … and moral guilt’, this difficulty should not be thought to tell against the distinction itself. ¹⁸ That is, after undertaking a right action—no matter the moral weight of the action itself—it is in fact the case that the agent is not morally guilty, the agent's moral emotions notwithstanding.

The example that Biggar uses here is the standard one about torture and a ticking-time-bomb scenario, and it is helpful not least because of how familiar it is. Biggar distinguishes between ‘“aggressive interrogation”, which intentionally inflicts pain on another human person, not out of sadism, but intending to preserve the innocent against grave injury, and as a necessary and proportionate last resort’, and ‘“torture”, which is wicked by definition’. ¹⁹ Let us grant that distinction for the sake of the argument. If a properly formed responsible agent (i.e., an agent ‘with a commendably sensitive conscience’) authorizes something like ‘aggressive interrogation’, the agent will inevitably feel some negative moral emotion as a result of his act. ²⁰ This is as it should be given the weight of the moral acts the agent has undertaken. According to Biggar, the agent might mistakenly feel guilty; what he should feel is, rather, regret.

The agent's moral emotions are not fully trustworthy: he may feel guilty while being non-culpable. It is part of the ethicist's vocation to help the agent distinguish between guilt and regret. This task is partly pastoral since the ethicist's ability to draw this distinction in a manner that is convincing to the agent ‘could well be all that stands between his [the agent's] tender conscience and self-destruction’. ²¹ The task is also part of the job of the ethicist qua ethicist since drawing this distinction is what safeguards against the paradox; the distinction guarantees ‘the coherence and harmony of the moral universe’. ²²

Notice that in this example the agent is ‘sensitive’ enough to wish for his ‘tender conscience’ to be cleared. This sensitivity is itself evidence of the agent's character. As Michael Walzer puts it in his ticking-time-bomb case (which is the original version of what the now ubiquitous versions are all inheritors):

When he ordered the prisoner tortured, he committed a moral crime and he accepted a moral burden. Now he is a guilty man. His willingness to acknowledge and bear (and perhaps to repent and do penance for) his guilt is evidence, and it is the only evidence he can offer us, both that he is not too good for politics and that he is good enough. Here is the moral politician: it is by his dirty hands that we know him. If he were a moral man and nothing else, his hands would not be dirty; if he were a politician and nothing else, he would pretend that they were clean. ²³

Walzer's final caveat—‘if he were a politician and nothing else, he would pretend that they were clean’—is very instructive here. The willingness to ‘acknowledge and bear (and perhaps to repent and do penance for)’ guilt is an important indication that the agent is rightly formed. To put it in Biggar's terms, the feeling of guilt—and not only regret—is, here, evidence that the agent has a ‘commendably sensitive conscience’. In this situation there is literally no other evidence. Or, to again put it in Biggar's terms, an unwillingness to so acknowledge and bear guilt is at least soft evidence—and it may be good evidence—for the idea that the agent who authorized ‘aggressive interrogation’ did so ‘out of sadism’ (or some other motive that is equally impure, thus making the act ‘wicked by definition’). ²⁴

But let us assume that the agent has, as Biggar describes, a ‘commendably sensitive conscience’. And let us assume that the agent did authorize the aggressive interrogation ‘intending to preserve the innocent against grave injury, and as a necessary and proportionate last resort’, rather than out of sadism. And, finally, let us assume, with Biggar, that the agent's ‘tender conscience’ is moving him toward ‘self-destruction’. Since we have already granted, for the sake of the argument, that the distinction between aggressive interrogation and torture is valid, in this case, Biggar is right that this agent would not be guilty, and thus should not feel guilt. Biggar argues that the ethicist has a responsibility to relieve the agent of his burden of guilt. Because the ethicist has the ability to draw the distinctions between non-culpable aggressive interrogation and culpable torture, it is incumbent upon the ethicist to find a way of making those distinctions convincing to the agent who feels guilt. Presumably, Biggar means that the ethicist is to convince the agent that the emotion he feels is properly categorized as regret instead of guilt, not to convince the agent to no longer feel what he feels.

It would be misleading to say that Biggar's ethicist is playing the role of a priest by offering a kind of absolution. It seems, instead, that the ethicist is standing between the agent and the confessional, telling the agent that he does not need the priest, since he is not guilty. But this, too, is doubtless not exactly what Biggar—who is a priest—means. What, then, does he mean? What is the role of the ethicist? Of the priest? Of the moral community for which the agent's willingness to acknowledge and bear his guilt is evidence of his morality? Three points are worth noting here.

First, it is in this discussion that Biggar argues that ‘dirty hands’ reasoning depends on paradox. Biggar writes:

A paradigmatic case of ‘dirty hands’ is the so-called ‘ticking bomb’ scenario, where a politician authorises the torture of a terrorist, who is known to have accurate information about the location of city-centre bombs intended to kill masses of civilians. De Wijze's ‘dirty hands’ view is that torture is always and everywhere wrong, but that there might be rare cases—such as the ‘ticking bomb’ scenario—where it is the right thing to do. It can be right to do wrong, to get one’s hands morally dirty: that is the paradox. ²⁵

Biggar's critique is of the claim that it is possible or even necessary to do wrong to do right. ²⁶ We can treat the same case with a slightly different description that allows us to take seriously the challenges presented by it without giving in to paradox. If we grant that torture is absolutely prohibited while also granting that in certain situations an agent will be faced with what appear to him to be good—even overriding—reasons to torture, what we have is a case in which a particular agent will have what he thinks are good reasons for and against the action. The agent will not know if he has done right, and he will have good reason to think that he has done wrong. This is not a paradox; it is an open case.

Second, it seems that Biggar's implicit argument is about the prima facie interpretation of negative moral emotions. Given the ‘tender conscience’ and possible ‘self-destruction’ of the agent, better to assume regret than guilt. Otherwise, the burdens of the moral life would be simply too heavy to bear. Soldiers, for instance, cannot afford to be taking continuous moral inventory. They would not be able to perform their jobs. The ethicist or priest may become convinced that the agent is guilty—and so that he should feel guilt, but they should begin with assumption that the agent is not guilty—and so that he should feel regret.

Third, Biggar's account of the ethicist's role in these cases is an implicit argument that resolving cases like these requires external judgment. I will return to this point in the final section.

Crucial for my argument at this point is the end to which feelings of moral residue should be ordered. Biggar does not tell us exactly, but we can draw inferences from his explicit arguments. ²⁷ Along with the implicit argument in Biggar's account of the role of the ethicist, about the importance of the moral community in helping the agent deal with moral residue, there are two explicit arguments that suggest possible ends to which moral emotion is ordered. These arguments are two sides of the same coin. The first is that these emotions can help us see that the set of circumstances that bring about the feelings of guilt or regret should be used to clarify the moral rules we use to do moral deliberation. The second is the inverse of the first. If the moral emotions experienced in a particular case demonstrate that the case does, in fact, come up against an inviolable moral rule then the agent is required to cease acting altogether, and surrender the situation to God's care.

The first inference comes from Biggar's approval of Michael Stocker’s thought. Stocker writes, ‘that there “remains [within these situations of compromises] … a disvalue even within that justified perhaps obligatory whole—a disvalue which is still there to be noted and regretted”’. ²⁸ Biggar then adds his own conclusion: ‘Noted and regretted, indeed; but not repented of’. ²⁹ The negative moral emotion that follows a rightly-undertaken compromise should not be ordered to repentance. Rather, Biggar argues, ‘we should be moved to reconsider the absolute rule, the kind of action it is meant to prohibit, and why the action in this case does not belong to that kind’. ³⁰ Regret, then, far from motivating us to repent of our action, can, rather, indicate that our rules themselves are not fine-grained enough to help us decide on a right course of action.

Here, Biggar is continuing to draw on the case of torture. Biggar writes,

‘aggressive interrogation’ should remain illegal, to prevent institutional habituation and corruption, but it can still be moral, and courts should prepare themselves to recognize it, somehow, as such. If I am correct about this, then the politician who authorises torture-as-enhanced-interrogation does no wrong, and he does not dirty his hands objectively. ³¹

The agent properly feels regret in this situation because the objective state of affairs is regrettable, not because the agent's moral state is polluted. This regrettable state of affairs does, however, include the action of the agent and the consequences of the act which caused ‘an evil; it involved deliberately causing it; it involved intentionally causing it’. ³² But it does not mean that the agent should have done otherwise, since this was the right course of action.

Biggar succinctly states:

if there are cases where it is morally right to torture, then these are cases when torturing is not morally wrong. Consequently, we should be moved to reconsider the absolute rule, the kind of action it intends to prohibit, and why the action in this case does not belong to that kind. In other words, such cases should provoke us to tighten the scope of the rule, so as to exclude them. ³³

One purpose of moral emotions, then, is to make us aware that an absolute rule may be too broad, and that the rule should be made either narrower or less absolute. The rule, in other words, should be rewritten in light of the regrettable case. ³⁴ And while the rule is being rewritten, the agent should be willing to ‘note’ and ‘regret’ the act and its consequences without repenting of them. ³⁵

It is unclear to me why such revision of absolute moral rules should not inform legal norms. If these acts should be accepted as moral but ‘should remain illegal, to prevent institutional habituation and corruption’, does this actually assist decision-makers (e.g., in a ticking-time-bomb scenario) in making moral decisions? According to Biggar, it falls to the courts to ‘prepare themselves to recognize it [the act], somehow, as such’. But surely this makes forward-looking casuistry subservient to backward-looking judgment. This seems to violate the intent of Biggar's argument. ³⁶

This conclusion is flipped on its head with the second argument. Like the first, this one considers absolute moral rules; unlike the first it comes in the next section of Biggar's argument that is about ‘common moral sense’ rather than regret exactly. But the argument follows the same logic. Where the first conclusion is based on Biggar's explication of torture, this one comes in the course of a conversation about pacifists and ‘just warriors’. Biggar writes that, like pacifists,

‘just warriors’ also recognise that it belongs to the specific dignity of human beings to bear responsibility for shaping the world as best they can within the terms set by absolute moral rules. And if they cannot defend or promote goods within those terms, then, together with their pacifist brethren, they too must bend their knees and pray God to do what they would, but may not. ³⁷

In the first instance, the moral emotion of regret leads to the reworking of an absolute moral rule. In the second, the presence of the negative moral emotion is a sign that an absolute moral rule has been, or more precisely in this instance, is about to be, violated. ³⁸ This would mean that the emotion would be rightly categorized as guilt rather than regret. The difference between the two is the basic difference between violating and not violating an absolute moral rule. As Biggar puts it, ‘A good compromise complies with absolute moral rules; a bad one violates them. Where the line between good and bad compromises is drawn then depends on which moral rules are viewed as absolute—and that, of course, is controversial’. ³⁹

Biggar demonstrates the controversial nature of drawing lines between rules that are revisable and rules that ‘are viewed as absolute’ by giving two examples that violate the rule against intentionally killing the innocent. The first is the example of ‘the intentional British bombing of German city-centres and residential areas in 1940–41, in order to terrorise the civilian population’. Here Biggar agrees with the ‘conservative Catholic’ position that judges this act ‘immoral’ and he disagrees with Michael Walzer who argues that it was justified as an ‘extreme emergency’. ⁴⁰ The second example, along with Biggar's conclusion about it, is worth quoting at length.

In 1944 the British were engaged in the desperate defence of Termoli, a town on the Adriatic coast of Italy, against German counter-attack. At one point German artillery shells a lorry-load of British soldiers, scoring a direct hit and dismembering everyone in it. A surviving British officer, who was not in the truck, picks himself up off the ground and finds himself facing a young boy screaming in the middle of the street, his parents killed and his intestines spilling out in front of him. The officer takes out his pistol, catches the poor boy, and, in a deliberate act of mercy-killing, shoots him dead. This was a clear case of the intentional killing of the innocent. Conservative Catholic moralists would regard it as a bad compromise: the pursuit of a good end—the relief of the boy's dreadful and terminal suffering—by the immoral means of violating an absolute moral rule. They might soften their judgement by adding that the officer's action was venial or forgivable, but, in their view, the action itself remained morally wrong and should not have been performed. I find myself unable to agree with them. Instead, the tragic case instructs me to revise the rule by restricting its scope, and to say that it does not apply in cases that fall outside of civil society, where the intentional killing of the innocent is in the innocent's interests, and where the innocent's consent may be presumed. ⁴¹

What Biggar does not provide in either example is what the agent should feel after performing the act. Indeed, the argument is written in such a way as to presume that the agent discovers beforehand that the rule is inviolable and he must refrain from acting. But this is clearly not what happens with Churchill and mass bombings. What should Churchill have felt after authorizing the bombings? What should the pilots flying the bombers have felt after performing the act? Are there substantive differences between Churchill's moral state as the one authorizing the attack and the moral state of the pilots as the ones carrying it out—and should one feel guilt, while the other should feel regret? ⁴²

These are no small questions. Should the officer performing the mercy-killing feel guilty? Should he feel regret? It is very plausible that he feels neither and, instead, feels hatred toward the enemy and moral satisfaction with the service that he rendered to the suffering boy. It seems especially implausible to describe Churchill as filled with regret, much less guilt, about the mass bombings. One lesson here is that moral emotions are not entirely predictable, and they depend in large part on prior moral formation. This is why the guilt of the politician in Walzer's example is evidence—rather than constitutive—of the politician's morality.

More importantly, to what end should those moral emotions be ordered? If we may presume that for Biggar the officer who performs a mercy-killing should not feel guilty, though if he has a properly sensitive conscience he likely will—and maybe even should—feel regret, what should the officer do with his emotions? Likewise, what should the pilots and Churchill do with theirs, if they do in fact have them? Who is required to repent and who is actively encouraged not to? Where do we draw the line between revising an absolute rule and honoring it?

Distinguishing between an absolute rule (e.g., against torture) that requires us to not undertake an action, and to instead pray to God, and a putatively absolute rule that can admit of fine-grained exceptions (e.g., that between torture and ‘aggressive interrogation’) is a very difficult task. In any given situation where this kind of decision is required, it is also likely to be the case that the agent does not have time to reflect on the case at length. He will simply have to act in one way or the other. Once he has acted, what is required? On Biggar's telling so far, he should not feel guilty if he has properly distinguished between the absolute rule and the putatively absolute rule that can admit exceptions. But we may reasonably assume that the agent likely did not have the training in moral philosophy to draw that distinction, and in any case, did not have the time to engage in the work of moral reasoning required to reach a justification for his act before undertaking it. What is required after this act?

The End of Moral Emotions

I am very hesitant to agree with Biggar that if we rightly undertake ‘aggressive interrogation’ and feel some negative moral emotion as a consequence, that we should use that negative emotion to fund a re-reading of the rule against torture, in order to make the rule precise enough to excuse the kind of act we have just undertaken. This seems to me to open wide the gate that leads to self-justification and even self-righteousness.

Recall that, for Biggar, the experience of a negative moral emotion that is occasioned by violating an absolute moral rule can mean one of two things. It can mean, on the one hand, that the moral rule should be reworked in light of the experience that brought about the negative moral emotion. Or, on the other, it means that the agent must cease acting and ‘pray God’ to accomplish the good or right end that the agent desires but cannot pursue because the absolute rule prohibits the agent from acting. Earlier, I referred to these two options as different sides of the same coin, because they both have to do with discerning whether an absolute rule prohibits an action in a given circumstance. Biggar explicitly points out that discerning which rules are, in fact, absolute, and thus binding, and which are not, is ‘controversial’. ⁴³

There is also another, implicit, way in which these two seemingly contradictory choices are actually two sides of the same coin. Both require the agent to surrender the act to external judgment. In the first instance (Biggar's example is the case of torture or ‘aggressive interrogation’) Biggar argues that the moral rule should be reworked while the legal prohibition against the action should continue without adjustment. According to Biggar, this means that ‘“aggressive interrogation” should remain illegal, to prevent institutional habituation and corruption, but it can still be moral, and courts should prepare themselves to recognise it, somehow, as such’. ⁴⁴ In the second instance, the agent surrenders the situation to God in prayer since the agent is prohibited from acting by the absolute rule. As I pointed out above, this does not tell us what the agent should do when he has already acted in the latter case, but it stands to reason that the agent should repent.

In both cases, the agent rightly surrenders his act—or desired outcome—to external judgment. In the first case, he does this by being held to account by the courts. It seems that by asking the courts to ‘somehow’ ‘recognise’ the agent's act as moral, Biggar expects the courts to pardon the agent for the act of aggressive interrogation. ⁴⁵ Court rulings strictly on the agent's adherence to the law will likely be reluctant to engage in recognition of this sort (and if they do, such recognition effectively means that the courts, rather than the legislators, are making aggressive interrogation legal). ⁴⁶ It seems to me at least as likely that the courts will find aggressive interrogators guilty of the crime of torture. It is this possibility that places the weight of the choice to torture or not where it properly belongs: on the agent. If the agent thinks that the courts are morally obligated to pardon him, he can act without concern for the rule that he violates. He might even be incentivized to hide his action because he has already ruled as the judge in his own case, and does not want to tempt the courts to get it wrong. If, however, the agent knows that he may be found guilty of a crime he is likely to demonstrate proper hesitation before undertaking the act. And if he is genuinely concerned about his moral state, he should open himself up to judgment rather than try to hide from it.

Such is the case, it seems to me, with the agent in the second case as well. Here, the only court to which the agent can appeal is the court of Heaven. Let us continue with the case of aggressive interrogation. Assume that the agent thinks he is in a situation in which the fate of thousands of innocent lives is in his hands. He has done his best to reason well about the situation, and concludes that either he must accept the deaths of the innocent based on the absolute rule against torture or, as a ‘necessary and proportionate last resort’, and ‘not out of sadism’, he must ‘intentionally inflict pain on another human person’. ⁴⁷ If the agent decides that he must accept the deaths of the innocent because of the absolute rule, what should he feel? According to Biggar's analysis, if the agent engages in the interrogation, he is not guilty and thus should feel regret rather than guilt—and, crucially then, should not repent. Biggar's judgment here is at least as contested as the Catholic position against mercy-killing that he references later. ⁴⁸ Does Biggar think that the agent who refuses to engage in the aggressive interrogation is guilty and, thus, should feel guilty and must repent? If so, how do we know? In matters that are, as Biggar tells us, this controversial, are we simply supposed to trust our own moral reasoning, and thus refuse to be held to account for the controversial acts we have undertaken?

The moral emotions the agent feels are meant to compel him to seek out and surrender himself and his deed to the moral community. The community is tasked with the activity of judgment that brings the incomplete work of moral reason to completion. This is congruent with Biggar's emphasis on the courts, the role of the ethicist, and the need to surrender some actions to God. Note, though, that this appears to make Biggar more open to the possibility of the agent's being found guilty than his argument seems to suggest. And this raises the fundamental question of guilt and regret anew. If an agent can be found guilty, either before the courts or before God, then the agent should feel guilty for the act under judgment. This has the effect of reducing Biggar's argument to the point that we should feel guilty when we are guilty and we should not feel guilty (thought we may feel something else) when we are not guilty. This flat way of putting it, however, does not account for the nuance of Biggar's position, especially as it is developed across his corpus, nor can it account for the ambiguities of situations of ‘dirty hands’.

Both the ambiguity of these situations and Biggar's account of the difficulty of reasoning morally in these situations show that the problem is simply that we do not always know whether our acts are morally justified before we undertake them. What is needed is judgment after. ⁴⁹ We may, as Biggar has put it elsewhere, have confidence about our choices without giving in to presumption about our own moral state. ⁵⁰ Even such confidence may, however, be misguided because the moral gravity of the case is too weighty for us to bear alone, or because we are sometimes wrong about our own morality, or because we sometimes change our minds about whether a moral act was justified. ⁵¹

Biggar's suggestion about the role of the courts and of God implies that moral knowledge of an act requires the act to receive some sort of external and retrospective judgment. ⁵² This need does not invalidate the work of moral reasoning that seeks to secure moral justification before the act. Rather it suggests that this sort of moral reasoning is even weightier than we might assume, since it is completed or perfected in the work of judgment—judgment that feelings of guilt should cause us to seek out. Here we see the importance of Biggar's argument about the role of the ethicist. The ethicist, like the courts, serves the agent by reflecting with him on the morality of his act. This suggests that the role of courts, the role of the ethicist, and the role of God are formally similar in Biggar's account, in that they are all external to the agent and they all work retrospectively to bring resolution to the open case.

If this is the case, then the necessity for retrospective judgment does not invalidate the work of moral reasoning—making of dirty hands nothing more than a sloppy appeal to paradox. Rather, retrospective judgment is the continuation and conclusion of moral reasoning. Negative moral emotions that arise from a situation of dirty hands should thus be understood as leading the agent toward, or even compelling the agent to seek out, this kind of judgment. The residue is an indication that judgment is needed. As such, it seems to me that guilt, at least as much as regret, is an important concept to retain—and that there are good reasons to prefer a prima facie interpretation of moral residue as guilt to a prima facie interpretation of those emotions as regret.

While regret describes the emotion in relation to a state of affairs, guilt describes the emotion that indicates to the agent that he needs to receive judgment. Regret, as the prima facie account of moral residue, forecloses on judgment, because it assumes that the case has already been brought to conclusion. But if judgment is what is needed in order to bring the work of moral reasoning to its conclusion, then to foreclose too early on the guilt that motivates us to seek judgment induces us to shield ourselves from judgment. And that is to leave our most difficult cases unresolved—even unaddressed—and, thus, to tolerate a rift in the moral universe. Since this is exactly what criticisms of dirty hands are meant to avoid, it seems to me that critics of dirty hands—including Biggar—should prefer a prima facie interpretation of moral residue as guilt.