‘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’ (Walzer 1973: 168).

This Special Issue on the problem of ‘dirty hands’ marks and celebrates Michael Walzer’s seminal article ‘Political Action: The Problem of Dirty Hands’, which was published fifty years ago in the then newly created journal Philosophy and Public Affairs. The aim of this Special Issue is to take stock and explore the corpus of work on the dirty hands problem that has arisen over the last five decades principally inspired by Walzer’s writing. The articles explore common themes and tackle possible new avenues and interests for dirty hands theorists. They help us better understand the concept itself by exploring dirty hands theory’s contribution to a more nuanced and sophisticated understanding of ethical theories in general, and the problem of intractable moral dilemmas in particular. This introduction outlines the progress made to date in the examination of this controversial notion and points to possible productive future research. Finally, we will briefly comment on each of the nine papers in this Special Issue to highlight their contribution to the developments within this literature.

1 Michael Walzer’s ‘Political Action: The Problem of Dirty Hands’

Over the last 50 years anyone writing on the problem of dirty hands, be they supporters or critics, has needed to engage with the core claims made by Walzer in his ground-breaking article ‘Political Action: The Problem of Dirty Hands’. It was written as a fervent rejection of both the absolutist (deontological) and cost/benefit (consequentialist) accounts of how we ought to understand our responses to difficult moral dilemmas in politics. Walzer’s deep concern was that neither approach adequately articulated the ethical and emotional complexity facing persons who were seeking to act morally in politics.Footnote 1 Good and effective politicians, Walzer argues, are, when necessary, willing to paradoxically do wrong in order to do right. They will violate a cherished moral principle to ensure that in the face of a terrible moral dilemma, the action taken will result in the lesser evil. The pressures of deontological prohibitions clashing with fundamental consequentialist concerns, leave politicians facing a situation where, even when their actions are all things considered justified, they will become morally polluted for having so acted. This, in a nutshell, is the paradox of dirty hands.

Moreover, Walzer argues that we need and want our politicians to dirty their hands to protect us when we face great social and natural evils. We want politicians to be good but not too good since our lives and wellbeing are in their hands. Politicians need to do what is necessary to protect us even when this entails violating absolute moral prohibitions. What distinguishes dirty hands from a consequentialist end/means analysis is that the politicians recognise that they have committed a serious moral wrong, even if it was for manifestly moral reasons. The moral violation needs to be understood, acknowledged, and felt for what it is—the engaging in an action that morally pollutes the agent and requires an appropriate moral response, perhaps even punishment. It is only when politicians acknowledge their wrongdoing that we know them to be both good persons and fit to govern in an ethical reality that is unavoidably messy.

To understand the context in which Walzer published his seminal paper in 1973, we need to appreciate the heady academic environment of the time. Walzer and his philosophical colleagues had formed The Society for Ethical and Legal Philosophy (SELF), whose specific aim was to address pressing political issues in a philosophical manner. His colleagues consisted of many of the great analytical philosophers of the last 50 years such as John Rawls, Robert Nozick, Thomas Nagel, and Judith Jarvis Thomson. As can be anticipated, Walzer’s account of dirty hands was not welcomed by most philosophers. Except for Nagel,Footnote 2 most analytic philosophers considered Walzer’s essay to be deeply flawed and the product of confused and erroneous reasoning. It argued for the existence of an unavoidable moral paradox in political life, but most analytical philosophers dismissed its central claim—doing wrong to do right—as self-contradictory, incoherent and one which heaped confusion on an already difficult and confusing situation.Footnote 3

Walzer, however, rejected such analytical accounts of moral theory as too wedded to a monistic understanding of value whose reductionism resulted in an idealised and simplistic account of the texture of moral life with all its contradictions and difficulties. Walzer’s account of dirty hands flows from his deep conviction about how academics ought to engage in philosophical work. All arguments need to be sensitive to our lived experiences and avoid unnecessary abstraction. A useful analysis needs to avoid bizarre or weird hypotheticals and instead use concrete historical examples which reflect our complex world. Hence, Walzer tries to think concretely about political questions and eschews ‘singular, abstract, or foundational answers to them’ (Walzer and von Busekist 2020: 131). As a result, Walzer’s enormous corpus of academic work, his books and articles, are always political arguments rather than attempts at high theory.Footnote 4

What is more, Walzer has never sought to respond to the deep concerns of his analytic peers. In fact, despite the enormous influence of ‘Political Action’, he very rarely returned to the topic of dirty hands. When he did write again about this problem it was in the context of his work on Just War theory and in particular his account of ‘supreme emergencies’ (1977/2006, 2004).Footnote 5 Walzer argues that in a just war it might be necessary to get dirty hands in situations where the enemy poses an existential threat to the nation and engages in actions motivated by an evil ideology, one that would ‘shock the conscience of mankind’ (1977/2006: 107). Here dirty hands would typically involve the mass killing of innocents to prevent a greater evil. Such an act would be a deliberate violation of the jus in bello criterion which forbids harm to non-combatants in war. What is important to note is that this raises the bar for dirty hands so high that such actions can only be justifiable in extreme cases when the threat is nothing less than the ongoing survival of the community (2004: 44–50). Walzer’s paradigm case of such a threat is the Nazi attempt to conquer all of Europe while committing genocide and enslaving all persons and nations deemed to be inferior. But this threshold for dirty hands is a far cry from the examples used in ‘Political Action’ which involved the ticking bomb scenario and grubby deals with corrupt ward bosses. This is because Walzer understands warfare to require a much higher and demanding threshold since actions within war have a lethal effect on vast numbers of innocent civilians.Footnote 6 Nonetheless, Walzer insists that the ethical conundrum in war is the same as that for politicians facing dirty hands scenarios in peace time. As he eloquently states:

A morally strong leader is someone who understands why it is wrong to kill the innocent and refuses to do so, refuses again and again, until the heavens are about to fall. And then he becomes a moral criminal (like Albert Camus’s “just assassins”) who knows that he can’t do what he has to do – and finally does.’ (Walzer 2004: 45)

For Walzer the moral reality of a dirty hands scenario can be distilled into the recognition that in some situations we fight evil by doing evil and, importantly, we recognise that the good done does not negate or render irrelevant the evil that needed to be done. Good persons need to feel the appropriate moral emotions and pay a price for so acting. Neither a deontological nor consequentialist account can properly capture this paradoxical and unavoidable aspect of our moral reality.

2 The Notion of ‘Dirty Hands’

2.1 Demarcating the Problem of Dirty Hands

The term ‘dirty hands’ is not usually understood as a metaphor for the grubby and sometimes bloody moral dilemma facing good agents. The more general use refers to the unpleasant and squalid tasks which we deeply dislike doing. Working in abattoirs or cleaning sewers are examples of activities where we refer to people involved in dirty work. This is essentially a descriptive use of the term and makes no normative judgment about the actions or work in which the person is engaged. Yet another usage of the term is normative but has a purely condemnatory aim and exposes the moral misdeeds of agents, typically that of politicians. It is common for the general public, journalists and others to insist that all politicians are dirty. The claim here is that all politicians are immoral simpliciter, and this is what we should expect from those who enter this profession. Politicians will lie, cheat and worse for their own selfish purposes in order to enrich themselves and their cronies.

Sometimes, dirty hands are equated to the notion of ‘admirable immorality’.Footnote 7 While there are different ways to interpret this concept,Footnote 8 the primary concerns of an admirable immorality focuses on character traits an agent possesses which we admire yet are conceptually linked to a strong tendency towards immoral behaviour. For example, we might think that ruthlessness is a virtue for successful politicians yet also think that this trait leads ineluctably to immoral actions.

The notion of ‘dirty hands’ as used by Walzer, and which is the focus of this Special Issue, differs from the above meanings; for our purposes it refers to situations where good persons, in the face of unavoidable moral dilemmas, engage in immoral actions for moral reasons. They do this to bring about the lesser evil but, unlike consequentialists, also acknowledge the moral violation committed and properly feel the moral pollution for having so acted.

2.2 Two Models of the Problem of Dirty Hands

The problem of dirty hands evokes a metaphor which links immorality with uncleanliness. The realisation that political agents face intractable moral conflicts which, at times, require choices that result in betrayal, mendacity, coercion and even murder, was understood as far back as the ancient Greeks and is reflected in the tragic plays of Sophocles and Aeschylus.Footnote 9 These concerns became linked to early Christian notions of purity and sin, such that immoral behaviours became strongly associated with the metaphor of ‘dirty hands’. Perhaps the most iconic account of this link is that of Pontius Pilate washing his hands to remove the moral stain from condemning Christ to death for political reasons all the while believing him to be innocent.Footnote 10

While the terminology of ‘dirty hands’—used to refer to a particular kind of moral conflict in politics—only arose in the twentieth century, Parrish argues that discussions of it can be traced from Antiquity through the Renaissance and into Modernity (Parrish 2007). In the past, such debates referred to the well-recognised problem of balancing the provision and preservation of public goods with the means required to do so—means which could seriously undermine an agent’s moral virtue. This approach, which we will call the ‘Machiavellian Model’, can be found in contemporary accounts of dirty hands by theorists such as Stuart Hampshire and Bernard Williams.Footnote 11 These theorists take seriously Machiavelli’s core insight that success, and even survival, in political life often requires the rejection of core Christian moral values and to place one’s moral virtue in serious jeopardy.Footnote 12 It is worth noting here that the focus of the Machiavellian Model is on the unavoidable collision of public (or more specifically political) duties with private moral convictions. To successfully fulfil one’s obligations as a political servant will inevitably undermine one’s private virtue. It implies a form of value pluralism in as much as it recognises the incompossible and yet valid demands of both the public and private sphere. These conflicting values cannot be placed in a moral hierarchy, nor made subservient to a single supreme or overriding value. Consequently, there will always be situations when the clash between public and private obligations ensures that the realisation of one cherished value inevitably undermines the other.

The second model, which we call the ‘Walzerian Model’, arises out of a clash between competing approaches to understanding ethical action and their rival conceptions of the right and the good. While deontological reasoning focuses on an agent’s adherence to absolute moral principles (i.e. what is right) based on reason or revelation, consequentialists look to maximizing specific non-moral goods. Since all moral judgments typically involve both deontological and consequentialist considerations, it is impossible to always avoid serious and irresolvable conflicts between these approaches. Walzer’s characterisation of dirty hands in politics focuses on this problem. In some situations, a politician will encounter choices where whatever they decide to do will be ‘exactly the right thing to do in utilitarian terms’ yet leave the politician ‘guilty of a moral wrong’ (Walzer 1973: 161). The Walzerian Model, over the last 50 years, has been used as the predominant account of dirty hands by both critics and supporters. What is more, academics in the social sciences, humanities, and elsewhere, have made use of this model when trying to make sense of irresolvable moral dilemmas in fields as diverse as medicine, psychology, law, warfare, policing, and business.

3 50 Years of the Problem of Dirty Hands

This section outlines the most significant debates since the publication of ‘Political Action: The Problem of Dirty Hands’ and points to where the articles within this Special Issue contribute to them. We start by considering some external criticisms of the concept of dirty hands itself before moving on to internal debates; i.e. debates among dirty hands theorists who nevertheless agree that the concept itself is a viable and essential part of our moral vocabulary.

3.1 External Criticisms

The most persistent, widely held, and potentially damaging criticism of the concept of dirty hands is that it is conceptually incoherent. There are two main arguments supporting this claim: the ‘meta ethical argument’ that argues against the existence of genuine moral dilemmas and the ‘all-things considered argument’ which rejects the claim that an action can be morally justified (even obligatory) yet nonetheless also morally wrong.Footnote 13 These arguments are seen as powerful and persuasive by both deontological and consequentialist moral theorists who, therefore, summarily reject the very possibility of dirty hands scenarios.Footnote 14 There has been considerable pushback against this dismissal of dirty hands in a number of ways, but especially by those theorists who endorse, what Christopher Gowans calls, ‘The Remainder Thesis’.Footnote 15 It maintains that even when we make a justified moral judgment about how to act in dilemma situations, such action can nevertheless involve moral wrongdoing, leaving residual moral pollution for having so acted. The central argument of those defending the notion of dirty hands is that both the meta ethical and all-things considered arguments are based on an overly simplistic understanding of moral choices, values and emotionsFootnote 16 as well as an adherence to a mistaken view of what is required of efficacious ethical theories.

There are two further commonly held criticisms of dirty hands which focus on the deleterious consequences for moral theory that result from taking such a notion seriously. Firstly, to accept the existence of dirty hands will lead us inexorably down a slippery slope towards condoning immoral actions. It offers evil persons a convenient excuse for their terrible deeds by letting them apply dirty hands rationalisations to any kind of political decision. Secondly, if we accept that dirty hands scenarios are possible, we face considerable difficulties in aligning this notion with a cluster of other important normative concepts, such as guilt, punishment, regret, and accountability (to mention a few). For example, the standard philosophical justifications for punishment of wrongdoers become problematic when applied to persons involved in genuine dirty hands scenarios. Recent work by dirty hands advocates has sought to understand how to integrate this paradoxical aspect of our moral reality within the normative concepts used to make sense of ordinary wrongdoing. We return to these issues below when outlining the internal debates among dirty hands theorists.

3.2 Internal Debates

3.2.1 The ‘Dirt’ in Dirty Hands

If the concept of dirty hands is possible, does it denote a sui generis ethical perspective which is different from other forms of unavoidable moral conflict? Those who endorse the ‘Remainder Thesis’ offer different explanations for why some circumstances make it inevitable for good persons to become morally compromised. For example, Gowans (1994) refers to cases of ‘inescapable moral wrongdoing’, Gardner (2007) to ‘justified moral wrongdoing’, and Tessman (2014) to ‘unavoidable moral failure’. None of these views necessarily endorse a dirty hands position per se but they do endorse the preconditions for their existence. It is therefore important to explain how dirty hands can be distinguished from other, more ordinary instances, of unavoidable moral wrongdoing. As Kramer notes, ‘every problem of dirty hands is a moral conflict, but not every moral conflict is a problem of dirty hands’ (2018: 187). What is it then that constitutes the unique ‘dirt’ in dirty hands? There is considerable agreement in the literature that dirty hands ‘do not simply involve mere bads and harms. […] Rather, they involve betrayals and violations of people, principles or values’ (Stocker 1990: 25). This, however, is not considered sufficient to set apart dirty hands and other moral conflicts and there is some disagreement about what is additionally required.Footnote 17 Stocker and de Wijze argue that the ‘dirt’ in dirty hands arises due to the ‘immoral circumstances created by other persons (or organisations of persons) within which an agent finds herself’ (de Wijze 2007: 12). Nick disagrees and argues that this account is too narrow and that, instead, dirty hands scenarios always involve the violation of a ‘core moral value’ (Nick 2021: 199). Kramer offers yet another account where dirty hands are unique because an agent needs to choose how to act in ‘a moral conflict that involves the prospect of evil conduct’ (2018: 189). Tessman, in her essay for this Special Issue, contributes to this debate by offering yet another way to think about what makes dirty hands unique.

Another issue which brooks disagreement among dirty hands theorists is whether it is possible to keep our hands clean once we are confronted with a dirty hands conflict. In the interview in this Special Issue Walzer argues that while politicians who decide not to torture even when this has a high probability of preventing a bomb from killing innocent people renege on their political responsibility to protect their citizens, they do keep their hands clean.Footnote 18 Nick, however, has argued against this view since it relies on drawing an implausible divide between the actions that we take or the principles that we adhere to and the outcomes of doing so. If dirty hands involve a kind of violation of a moral principle or value, once confronted with a dirty hands situation, whatever course of action is taken ‘will always inevitably involve such a violation’ (2019a: 938). Dirty hands for her are therefore symmetrical; when faced with a dirty-handed choice it is already too late to keep our hands clean.Footnote 19

3.2.2 Moral Emotions and Dirty Hands

When agents do wrong to do right, how should they feel about getting dirt on their hands? Put differently, what is the appropriate moral emotion that those with dirty hands ought to experience? Remorse and shame are fitting emotions for moral wrongdoing, but they are inappropriate for dirty hands cases given that the immorality was justified, perhaps even morally obligatory, because it brought about a lesser evil.Footnote 20 Recent discussions in the dirty hands literature explore different possibilities such as Bernard Williams’ (1976) distinction between regret, agent-regret and remorse. Williams argues that persons who are not guilty of moral wrongdoing, but whose actions resulted in terrible outcomes, ought to feel ‘agent-regret’.Footnote 21 This emotion, which is distinct from remorse or the regret of a spectator, recognises the causal responsibility for great harms and acknowledges that ‘in the story of one’s life there is an authority exercised by what one has done, and not merely by what one has intentionally done’ (2008: 69).

De Wijze, however, has argued that Williams’ three emotional possibilities do not fit with dirty hands scenarios since ‘to feel mere regret about this state of affairs would fail to do justice to the serious moral violation […] while to feel remorse would falsely suggest that [the agent] had no moral justification for [their] actions. Agent-regret will not do either since it is not merely the fact of [their] causal role in the event that is problematic’ (2004: 464). Instead, de Wijze argues for, what he terms, ‘tragic-remorse’, which is the appropriate response by good persons to a justified violation of cherished moral values necessitated by the evil actions and projects of others. There is, appropriately, remorse for the moral violation but also acknowledgement of the tragic circumstances which necessitated the act and the resulting moral pollution for the agent.Footnote 22 The experience of tragic-remorse urges the agent to justify her actions to relevant parties and repair, as far as possible, the harm she has caused.Footnote 23

A different approach to understanding the emotional toil involved in dirty hands situations is through the language of moral distress and injury. Tigard (2019) argues that an agent experiences moral distress when they want to act morally but are unable to do so. As a result, the agent may feel a variety of emotions such as frustration, anger, anxiety, guilt, and regret. But importantly, she also feels torn about experiencing these emotions and continuously queries whether they are truly fitting for the choice forced upon her. This inner turmoil results in moral distress and precisely captures the emotional experience caused by dirty-handed actions. Wiinikka-Lydon takes a different tack and argues that dirty-handed agents can be understood as experiencing a kind of moral injury. Moral injury is the experience of guilt, shame and suffering that occurs as a result of the traumatic experience of committing a moral wrong while attempting to carry out one’s moral or professional duties. As a result, ‘the individual experiences […] a moral fall. They feel that they are no longer able to successfully strive toward certain expectations of what it is to be a good person’ (2018: 359).Footnote 24

The way in which agents ought to feel about their dirty hands will be problematised by Lisa Tessman and Joe Wiinikka-Lydon in this Special Issue.

3.2.3 Dirty Hands as the Ends/Means Problem in Politics

One significant strand of theorising about dirty hands results from insights taken from Niccolo Machiavelli’s work (1532/2003) and is often referred to as the ‘ends/means problem in politics’ (Gowans 1994: 228–234; Hampshire 1991: 162–168). According to this account, dirty hands are exclusive to the political domain since it is here that we find the unavoidable clash between private moral scruples and political necessity that gives rise to the problem in the first place. To wit, dirty hands problems are uniquely political problems. This approach assumes a sharp distinction between a personal and political (or public) morality, the former concerned with maintaining individual moral goodness by acting within clearly defined deontological constraints, while the latter seeks to ensure the safety and well-being of citizens in the face of an immoral and dangerous world (Hampshire 1983: 121–124; Niehbur 1932/2013: 231–278). As Huxley points out in his biography of Father Joseph who sought and failed to reconcile his religious principles with his position as adviser to Cardinal Richelieu: ‘To be a good man according to God is one thing; to be a good man according to men is quite another’ (1941/1994: 146).

This ends/means interpretation of dirty hands relies on a specific view of what is required of those who seek to be successful when engaged in realpolitik. Given that politics is properly understood as requiring actions which good persons need to reject qua private citizens, the constraints of morality must take second place to ensuring the survival of the state and safety of citizens.Footnote 25 Tillyris endorses this insight and offers his ‘dynamic model’ of dirty hands (2015). He rejects Walzer’s conception of dirty hands as deeply problematic since it assumes that moral violations in politics (especially in well-functioning democracies) are intermittent dysfunctional occurrences in which agents do wrong in order to do right and after which they should seek to regain their moral standing (for example, through public acknowledgement and atonement). The dynamic model, in contrast, focuses on the virtues needed of politicians facing the constitutive dirtiness of politics and who, rather than being able to regain their moral standing, will have to become accustomed to their dirt.

However, this Machiavellian ends/means problem in politics has been challenged. Firstly, there are strong reasons to believe that in a democracy the need for a politics that takes deontological constraints seriously is paramount for a decent society. Secondly, understanding dirty hands as a distinctly political problem fails to account for the ways in which some moral dilemmas in private life can be rightly seen as cases of dirty hands. In short, the concept of ‘dirty hands’ needs to be understood as an important part of our moral vocabulary that has application outside the sphere of politics (de Wijze 2007, 2022).

In this Special Issue, the essays by Giorgini, Zaibert, and Finlay continue this debate offering arguments for and against understanding dirty hands scenarios as ends/means dilemmas in politics.

3.2.4 Dirty Hands and Political Virtues

Even if it is agreed that dirty hands scenarios are possible in all spheres of human interaction, it is widely acknowledged that they are found in their most challenging form in the domain of politics. Given the moral complexity of dirty hands scenarios in this sphere, an important question is whether politicians require a specific set of political virtues that would enable them to be effective in their role, while at the same time allowing them to retain their moral goodness. Thomas Nagel (1993) argues that getting one’s hands dirty will require politicians to exhibit a certain degree of ruthlessness. William Galston, in turn, has suggested that politicians require a degree of ‘toughness’—understood as an Aristotelian mean ‘between squeamishness, wishfulness and innocence (one the one hand), and callousness, cynicism and calculation, on the other’ (1991: 176). Building on a Machiavellian rather than Aristotelian understanding of virtue, Tillyris (2016) has argued that political leaders need to cultivate virtù—i.e. the set of characteristics that enables them to guarantee the continued survival and success of the political community—which will at times require the practice of what, in other areas of life, would constitute a vice. These virtues should be understood as distinctly political virtues because they are supposed to allow politicians to do what is necessary qua politician without, simultaneously, them losing sight of the moral costs of their actions.

What these views have in common is the claim that politics is a distinct sphere of endeavours and to succeed in them requires an appropriate set of virtues.Footnote 26 Furthermore, as Susan Mendus points out, given the nature of politics with its unique demands, constraints and responsibilities, being ruthless or tough or possessing virtù does not mean that politicians are necessarily ‘morally worse than the rest of us’ (2009: 51). They can remain good persons despite their moral violations.

Wiinikka-Lydon in this Special Issue examines the idea of what he calls ‘dirty virtue’ to address this debate.

3.2.5 Democratic Dirty Hands

While we have seen that dirty hands arise with particular urgency in the domain of politics, what remains to be seen is how they can enable and threaten democratic politics in particular. Ramsay (2000), Shugarman (2000) and Sutherland (1995), however, have argued that dirty-handed actions and policies are incompatible, both in principle and in practice, with genuine democratic politics. Such actions, they argue, will ultimately undermine the cherished democratic ends that, paradoxically, are used to justify the use of the morally problematic measures in the first place. Democratic politicians therefore cannot get their hands dirty but instead commit wrongdoing simpliciter. Both de Wijze (2018) and Nick (2019b) have argued that this line of reasoning is mistaken as it relies on flawed assumptions about both dirty hands theory and the ability of democracies to resolve the occurrence of irresolvable moral conflicts through open deliberation and the rule of law. This last point has recently received more uptake in the literature as writers have begun to explore how dirty hands are, indeed, part and parcel of quotidian politics (Hall 2022; Sarra 2022; Tillyris 2017).

Another strand of discussion concerning democratic dirty hands focuses on the need of democratic politicians to publicly reckon with their dirty hands.Footnote 27 This has been argued against most prominently by Richard Bellamy (2010) who posits that dirty-handed politicians ought to wear ‘clean gloves’ and not admit their actions to the public. Bellamy argues that, since it is impossible to eliminate conflict altogether from politics, political actions will necessarily fall short of the ideals of liberal democracies. However, because the liberal democratic project is a morally important one, politicians need to ensure that citizens continue to subscribe to it, even if its ideals and promise can never be entirely fulfilled. When a democratic politician decides to wear clean gloves, however, they are faced with a secondary dirty hands problem because they have violated an important moral value in preventing their citizens from holding them to account. In the words of Thompson such politicians then ‘have doubly dirty hands’ (1987: 32).

This raises the thorny issue of whether, and to what extent, citizens in a democracy share in the dirt of their political leaders. There are two broad stances that have been taken on this in the literature. On the first understanding, citizens have authorised political leaders to act in their name. This view has been developed by David Archard (2013) and Miriam Thalos (2018) respectively. Both of them understand citizens to be the ultimate authors of political action, but they disagree about what this means for the level of responsibility they ultimately hold. On the second account, citizens have consented to political leaders doing whatever is necessary to act in their best interest and hence share in the dirt committed on their behalf.Footnote 28 The most sustained defence of this view is given by de Wijze who argues that because ‘citizens rightly expect their politicians to protect them from harm even in situations where there is no morally cost free action or policy’ (2018: 140) available, we can understand them to have implicitly consented to their political leader’s dirty hands. For de Wijze, political leaders will, however, shoulder a greater degree of responsibility because they are the ones who ultimately choose to commit the dirty-handed action in question. Citizens, on the other hand, are merely complicit accessories in the act.Footnote 29

In this Special Issue, Kirby’s essay examines this debate and offers a novel way to think about the responsibilities of democratic citizens for the dirty hands of their politicians.

3.2.6 Punishing Dirty Hands

In his ‘Political Action: The Problem of Dirty Hands’, Walzer contends that a person who gets dirty hands ‘commits a determinate crime, and […] must pay a determinate penalty’ (1973: 178). Furthermore, we cannot leave the punishment up to the agent herself; instead, the punishment needs to be socially recognised, limited and proportionate. However, there is considerable disagreement among dirty hands advocates concerning the plausibility of these claims. Most fundamentally, what could be the justification for the punishment of those with dirty hands since their actions were, all things considered, the right thing to do?

De Wijze (2013) has argued that our standard justifications for punishment—whether consequentialist, retributivist or communicative—cannot accommodate the nuanced and paradoxical nature of dirty hands scenarios. In response, he has offered a number of alternative grounds on which we could potentially justify punishing dirty-handed agents instead. Fausto Corvino (2015), further building on de Wijze’s work, has sought to understand how to justify punishing dirty-handed agents in more complex dirty hands cases that occur as a result of collective action failures.

However, there is strong disagreement over whether punishing the dirty can be justified. Levy (2007) argues that punishing those with dirty hands is both counter-productive and immoral. He posits that politicians who get dirty hands are not blameworthy despite acknowledging that the actions taken committed moral violations. Levy argues that blameworthiness, and with it the justification for punishment, necessarily presupposes that the agent had a genuine choice to act morally. Since this is not the case in genuine dirty hands scenarios, the agent ought not to be punished. Additionally, Meisels argues that ‘punishing (the agent) for what we ourselves would have wanted (the agent) to do is no longer an irony or a paradox; it is simply wrong’ (2008: 173).

Cristina Roadevin, who is in broad agreement with Levy and Meisels, proposes that in the place of punishment, agents should receive a particular form of forgiveness, what she calls ‘no-fault forgiveness’ (2019: 123). No-fault forgiveness expresses both ‘our moral disapproval of the wrong action but also our recognition that these agents have done something good, perhaps worth praising or admiring’ (2019: 124). Peter Digeser, on the other hand, has argued that forgiveness is not an appropriate response to dirty hands situations because there is a significant danger that it could make it too easy and convenient for politicians to portray immoral actions as politically necessary. Instead, he argues that an angle of repose—‘a kind of suspicion of and distancing from governmental action’ (1998: 714)—would be more appropriate than forgiving in dirty hands cases.

In this Special Issue, Leo Zaibert examines the relationship between punishment and dirty hands.

3.2.7 Dirty Hands, Political Violence, and War

Some of the paradigmatic examples of dirty hands scenarios have been closely linked to political violence and war. In this Special Issue, Christopher Finlay argues that dirty hands arise from ‘the need to engage with certain empirical phenomena that frequently generate peculiar normative problems for political action’; the most important of these phenomena being the practice of violence. The ‘dirt’ of dirty hands results from the evil of violence itself, which undermines the noble or worthwhile ends it seeks to ensure. Even in cases where violence is used to protect legitimate political and moral systems, the evil of violence can never be wholly cancelled by the good that is achieved.

Turning to the matter of warfare, there is disagreement about whether dirty hands scenarios are possible in this context. The standard view is that if political leaders and soldiers follow the rules of just war—the ad Bellum and in Bello principles—then they do not commit moral wrongs and hence the notion of dirty hands is not applicable here. However, Michael Neu (2013) argues that this binary account of war as either just or unjust fails to reflect the tragic dimension inherent in all such conflicts. Understanding justified wars as cases of dirty hands provides a nuanced understanding of why violence on this scale can never be committed without creating a moral remainder. In this Special Issue, Walzer rejects Neu’s position but allows for the possibility of dirty hands if the state faces a threat which constitutes a supreme emergency. In such circumstances, he argues, the violation of in Bello principles that protect non- combatants can justifiably be violated and the situation be described as a dirty hands problem.

There have been other applications of dirty hands theory to situations of appalling violence. Firstly, there is the much-vexed question of whether torture should ever be morally justified in order to save lives. This is usually discussed with reference to the (in)famous ‘Ticking Bomb Scenario’, introduced by Walzer as a paradigm example of a dirty hands scenario in politics. There is rightly a great deal of concern that, given torture is mala in se, it should never be practiced by decent law governed societies.Footnote 30 Justifying torture as a dirty hands scenario, even to save many lives, leads down a slippery slope towards moral corruption and degradation. It would require a society to institutionalise torture, thereby corrupting and coarsening the legal system and law enforcement organisations.Footnote 31The Ticking Bomb scenario is also critiqued as a fantasy thought experiment that is based on deeply faulty assumptions and has little purchase on how such scenarios would actually play out.Footnote 32

Yet despite these powerful arguments, political leaders are duty bound to do what is necessary to protect citizens from the great evils of arbitrary violence, terrorism, war, genocide and annihilation. Many dirty hands theorists argue that to refuse to torture when so doing would prevent mass murder is deeply counter intuitive and relies on an overly rigid adherence to absolutist principles. Dirty hands justifications, they argue, occupy the moral middle ground between outright rejection and the easy consequentialist acceptance of torture based on utility or cost/benefit calculations.

Secondly, there has been much disagreement over whether a dirty hands scenario is the best way to characterise the use of targeted killings in asymmetric conflicts. De Wijze (2009) contends that justified targeted killings are dirty hands scenarios rather than simply another form of legitimate warfare.Footnote 33 They always involve a tragic element and leave a moral remainder. In contrast, Jones and Parrish (2016) see this justificatory approach as deeply flawed.They argue that if a policy of targeted killing can be justified, it must be so either in terms of the exacting standards of just war theory or in the context of a law enforcement ethic that is applicable to police in their actions against civilians. Meisels, in her essay in this Special Issue, rejects the very idea that targeted killing needs a dirty hands lens for it to be justified. She argues that under certain circumstances targeted killing can be a legitimate and morally justified military response to one’s enemies.

3.2.8 Dirty Hands and Resistance to Injustice

As we have seen, much of the literature on dirty hands in the context of political conflict, violence and war has concentrated on the issues faced by those in political leadership roles. There have, however, been some instances in which dirty hands theorists have questioned this focus and asked to what extent dirty hands can also be a necessary means of resistance and disobedience to be used by those who are not in positions of political power. This idea of dirty hands can be traced back to Sartre’s (1948/1989) eponymous play in which Hugo, a young resistance fighter grapples with his task to assassinate Hoederer, a political leader he sees as a class traitor. Similar themes can also be found in Albert Camus’ (1949/2013) play The Just in which a group of revolutionaries plan and carry out the assassination of a Grand Duke, and Berthold Brecht’s (1930/2001) play The Measures Taken in which a group of four revolutionaries are forced to kill one of their comrades to ensure the success of their mission.

Despite the relative prominence of this type of scenario in literature, it has only received limited attention from dirty hands theorists so far. Patrick Taylor Smith has examined the case of revolutions problematised in the plays above. He argues that any revolution is inherently risky because ‘it disrupts relationships that are often of real value with the speculative hope that they will be able to replace these relationships with a new system that will be, in the aggregate, superior’ (2018: 204). And in order to bring about the revolutionary vision, revolutionaries will have to unilaterally seize power and employ morally dubious and coercive means along the way. This, for Smith, is the distinct political dilemma that constitutes the dirty hands of revolutionaries.

So, what does that mean for those facing injustice and oppression? de Wijze has attempted to ‘delineate the normative constraints that good and moral persons ought to respect when fighting injustice’ (2012: 151). He argues that we need to understand how we can resist such injustice while staying within ‘limits beyond the normal moral boundary’ (2012: 169). For him, these limits are circumscribed by the principles of proportionality and reasonable success as well as by the agent’s motives for action. It is worthwhile noting, however, that de Wijze explicitly focuses on those in leadership positions (whether it be for nations, organisations, or movements) for the purposes of this analysis. Tillyris, in this Special Issue, points out that this is not sufficient and that we need to start exploring the applicability of the concept of dirty hands to those who engage in day-to-day resistance to oppressive regimes and systems.

3.3 Applying Dirty Hands to Professional Practice

While most of the work done on the problem of dirty hands concerns theoretical issues, the concept has been applied by practitioners in different disciplines to characterise and intuitively capture the difficult moral dilemmas that are faced when carrying out their duties. Leslie Griffin (1995), for example, notes that lawyers are duty bound to protect the interests of their clients, yet they also are duty bound to uphold the integrity of the law. When this conflict arises, lawyers face the problem of legal dirty hands. Iris Domselaar (2017) endorses a similar view and argues that tragic legal choices are also an indispensable accompaniment to any proper understanding of judicial adjudication. Judges then, like lawyers, can face legal dirty hands scenarios which best reflect the tragic choices that need to be made.Footnote 34

In the medical profession there has been a similar concern with moral conflicts and their pernicious effects on medical practitioners. Daniel Tigard (2019) highlights the problem of ‘moral distress’; a kind of psychological disequilibrium that arises when medical practitioners feel they have been prevented from pursuing a morally valuable course of action. Medical practitioners can find themselves feeling deep frustration, anger and anxiety for what they did, yet also believe it was morally justified and necessary to so act. Here, a good person’s natural feelings of wrongdoing—expressed as guilt, shame and regret—coexist with the contradictory assessment that there was no moral wrongdoing. Deborah Zion (2004) explores a related aspect of the dirty hands problem facing medical practitioners. She highlights the difficult clash between the obligations which arise from their professional medical ethic and the injustices of the broader social system within which they operate. Situations arise where the choice for doctors might be either not to participate (hence provide no treatment) or to offer treatment that could be considered unethical. Even if medical practitioners opt for the lesser evil, this compromises their moral goodness and renders them as both perpetrators and victims of evil.

Law enforcement officers also face tragic moral choices and, again, the concept of dirty hands has been evoked to properly characterise such circumstances. Kleinig (2002), Klockars (1980), Miller (2007) and van Halderen and Kolthoff (2016) examine situations in which police act in ways that violate the rights of individuals, but for noble causes such as community safety or to obtain information that will save lives. The focus here is on the moral dilemmas faced by law enforcement officers who seek to avert or minimise the evil done by criminals but cannot do this without acting in an immoral manner themselves. For example, they may be faced with a situation in which they have to violate the rights of a violent criminal to obtain information that will free a hostage or enable some other laudable end.Footnote 35 These situations are often referred to as either the ‘Dirty Harry’ problem or cases of ‘Noble Cause Corruption’.

The sphere of business activities also forces difficult moral dilemmas on those who wield power. Kaptein and Wempe (2002: 159–195) argue that running a business will at times require dirtying ones hands as there will be an unavoidable and irresolvable conflict between the rights and interests of the business’ collective stakeholders and the continuity of the business entity itself. The kind of situations they have in mind are when a business has to ‘consider firing employees to increase returns or when a decision has to be made about whether environmental investments that exceed the law, are justified if it is at the expense of profits’ (2002: 168). When this occurs, they argue, business leaders are forced to choose the lesser of two evils and get their hands dirty in doing so.

Finally, the concept of dirty hands has been used to examine, explain and justify the actions and policies of both governments and nongovernmental organisations, especially in situations where there are longstanding, deep and trenchant conflicts. Dixon (2002) explores the bloody conflict in Northern Ireland and how the British government needed to get dirty hands in order to facilitate a peace deal. Garrett (1996) and Sanders and Grint (2019) offer a lesson in leadership ethics by examining the dirty hands dilemmas facing local resistance activists and the allied forces in their fight against Nazi tyranny during WWII. Lenze and Bakker (2014) argue that Obama’s use of targeted killing in the war on terror was a case of dirty hands, while Grint (2016) explores 21 days of difficult political leadership in the United Kingdom through the prism of a dirty hands lens. More recently, Lewitt (2022) has applied a dirty hands framework to analyse UK policy-making during the COVID-19 pandemic. The problem of dirty hands can also be used to inform the analysis of actions taken by nongovernmental organisations; for example, Rubenstein (2015: 87–114) uses the case of Rwandan refugee camps in Zaire to argue that NGOs will frequently have to grapple with, what she calls, ‘splattered hands’.

3.4 Further Avenues of Research

Given the considerable scholarship on the concept of dirty hands that has built up over the last fifty years, in which directions might future work in this area be headed? There are two broad categories in which this further study could take place. The first concerns additional theoretical work on the concept itself and its place within our understanding of ethical theory. The concept of dirty hands feeds into a meta-ethical debate concerning our fundamental assumptions about the nature and purpose of moral theories. This includes the disagreement about the very possibility of dirty hands, the role of emotions in our understanding of our moral reality, and the scope of the ethical which, many argue, extends beyond merely what is action-guiding. The concept of dirty hands offers further arguments for those who seek to challenge the dominance of the standard monistic ethical theories of the last few hundred years. Furthermore, the concept of dirty hands strengthens the recent arguments for exploring the efficacy of virtue ethics as an alternative to deontological and consequentialist moral theories.

Secondly, the practical application and impact of dirty hands theory on a range of pressing and important issues is likely to continue at pace. Consider the moral dilemmas which we are facing in the attempt to deal with the problem of rapid climate change. If nothing is done, we face a potential catastrophe with rising temperatures, extreme weather conditions and the terrible socio-political consequences that result from droughts, floods, mass migration, and wars over dwindling resources. Yet, to tackle climate change requires choices which will include using coercive means and may diminish the current quality of life for billions of people. Undoubtedly, the nations and citizens who are socio-economically most disadvantaged will suffer disproportionately and it may not be possible, given our current political and social reality, to significantly mitigate this suffering. Here, leaders across the world will face terrible moral dilemmas and dirty hands choices. If we believe that morality needs to play a part in these future policy changes, a dirty hands framework would offer a nuanced and realistic moral framework for judging our politicians given the moral problems they will face.

Another area in which dirty hands decisions will certainly arise is the increasing use and the growing sophistication of artificial intelligence (AI). These advances could give rise to deeply intractable conflicts, for example, in the use of future technologies in genetics, new weapon systems, and the transformation of our industrial base as machines can increasingly replace people. Should decisions made by AI systems be completely autonomous and independent from human decision making? If so, should they be programmed to make ethical choices that reflect a dirty hands reality? And given that moral emotions are central to the concept of dirty hands (and ethical thinking in general), are AI systems properly able to make the correct moral decisions in the face of intractable moral conflicts?Footnote 36 While there has been some work on coding the equivalent of emotions into autonomous systems (Arkin et al. 2012), this cannot truly approximate the complex function that emotions perform in dirty hands situations. As a result, do the insights from the problem of dirty hands make us rightly very apprehensive about any autonomous AI system operating without human oversight?Footnote 37

Another topic on which dirty hands theory seems apposite concerns the process of transitional justice and the provision of reparations (Nick 2022). The progress from a deeply unjust conflictual society to a decent and just one is never an easy or morally cost-free process. For example, the use of Truth and Reconciliation commissions, such as the one set up in South Africa after the end of the Apartheid years, resulted in many serious crimes not being prosecuted. This facilitated a process of reconciliation and peace after a long and bloody conflict but it left victims of state sponsored racial violence without justice. Again, a dirty hands analysis could be useful in understanding this morally messy process of conflict resolution and the tragic conflicts that arise from it.

Finally, the problem of dirty hands has been explored within a Western set of theoretical assumptions about the nature of ethics, the existence of moral conflicts, the relevance of moral emotions, theories of human agency and so on. An interesting and important further project would be to explore other moral traditions to establish if corresponding concerns exist in them. If they do, how do these moral traditions deal with them and what can we learn from this?Footnote 38

3.5 Articles in this Special Issue

This Special Issue starts with a short article by Michael Walzer who revisits his seminal 1973 paper on dirty hands. This is followed by an interview of Walzer by Stephen de Wijze in which they explore some of the issues raised in his articles and the impact that Walzer’s ideas on dirty hands have had over the past fifty years.

The second article by Giovanni Giorgini presents an alternative account of dirty hands that departs from Walzer’s original formulation and is instead distinctly Machiavellian. Giorgini argues that dirty hands should not be understood as a conflict between different moral considerations, but rather as a clash between the incompatible demands of two distinct spheres of action: morality and politics.

The third article by Leo Zaibert also presents an alternative account of dirty hands from Walzer’s but is located on the opposite end of the spectrum to Giorgini’s. By connecting dirty hands to some of his previous thought on punishment, Zaibert argues that dirty hands theory has focussed too much on the realm of politics at the expense of seeing that the concept of dirty hands is a distinctly moral problem.

Lisa Tessman’s essay offers an innovative account of how we ought to conceptualise the moral conflict at the core of dirty hands situations. She argues that dirty hands involve an unintelligible choice between agent-relative and agent-neutral reasons that results in unthinkable and tragic loss.

Joseph Wiinikka-Lydon’s article explores what kind of character is needed to make the right decisions in dirty hands scenarios. He proposes that it requires a particular kind of virtue—what he calls ‘dirty virtue’—a crucial part of which is the experience of pain and suffering in acknowledgment of the moral wrong committed when getting one’s hands dirty.

While the first five articles focus on theoretical aspects of the concept of dirty hands, the final four articles examine the application of dirty hands to the themes of democracy, violence, war and resistance.

Nikolas Kirby explores the relationship between democratic office holders, institutions, and citizens. His aim is to examine the extent to which the dirt and responsibility for dirty-handed actions attaches to each of them. Kirby argues that while those elected to represent citizens get dirty hands, the mechanisms of democratic authorisation operate in a way that keeps the hands of citizens clean.

Christopher Finlay’s article, using insights from Max Weber, focuses on the use of violence as a necessary means for effective political action. While there are alternative reasons for agents to dirty their hands, he argues that the need to employ violence explains why this concept is particularly prominent and pressing in the sphere of politics. This is so not only because the state holds a monopoly on the legitimate exercise of violence, but also because violence is often justified supra-morally in the name of protecting a state’s very system of morality and law.

Tamar Meisels’ essay considers the issue of dirty hands in the specific context of warfare and argues that targeted killings and the use of drone strikes are not instances of dirty hands. She argues that when such means are used in line with the principles of a just war, there is no wrongdoing involved and hence talk of such actions being cases of dirty hands is incorrect.

Finally, Demetris Tillyris argues in his essay that dirty hands theory has focussed too much on the actions of those in power and has done so at the expense of exploring the dirty hands situations faced by the powerless; those who engage in acts of everyday resistance. He contends that dirty hands actions can be an important weapon of the weak in their fight against injustice and oppression.

We would like to thank the editors and support staff of The Journal of Ethics, in particular Wim Dubbink, and the reviewers for their time and effort in enabling this Special Issue to mark the 50th year of the publication of Michael Walzer’s seminal essay.